From J.J.Smith at gcu.ac.uk  Thu May  1 00:04:38 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 00:04:38 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <CAO2PinO6UbK_7t2UznUBQ9i_MWfa4C+eHwbz2Y1LANbxMofscA@mail.gmail.com>
References: <009701cf5325$4e6040e0$eb20c2a0$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CEBBA7B4B@EXCHANGE1.ad.nottingham.ac.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFADBDF5F@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<006b01cf645b$3b70e650$b252b2f0$@co.uk>
	<CAO2PinNOMKe0hBT4S8gNwCDcfoaLWKFu9NgoTEjKosqoa=_j2A@mail.gmail.com>
	<01ae01cf64a6$2aa18190$7fe484b0$@co.uk>
	<CAO2PinPKmj8T=-4p1WJ50mWz_hfg_4Curx8Qwc-==P8iHxdrYw@mail.gmail.com>
	<01e201cf64b1$12f26c20$38d74460$@co.uk>,
	<CAO2PinO6UbK_7t2UznUBQ9i_MWfa4C+eHwbz2Y1LANbxMofscA@mail.gmail.com>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101AF@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Anyone still awake? Anyone able to test a new export zip script?

Unzip and drop it in /modules/xerte/

Uses the unix command zip and pipes the stream back into PHP in 8192 chunks

Not sure it will work any better with large files but I tested from a Linux server with a 79 Mb zip without changing my settings (not sure if they are at 128Mb but wouldn't have thought they would be more than that...

Also, might not work on Windows...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Pat L (pgogy) [xerte at pgogywebstuff.com]
Sent: 30 April 2014 22:04
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

62MB is a big MP4 - especially given the limits on display
128 might work, but memory_limit and max_execution_time problems are dependent on uploads, so it is more likely to work, but not a guarantee - so "requirement" is perhaps not enough


On Wed, Apr 30, 2014 at 9:16 PM, Ron Mitchell <ronm at mitchellmedia.co.uk<mailto:ronm at mitchellmedia.co.uk>> wrote:
I don't think an alert is displayed to the user when it fails and the example_zipper files are left in the project folder and I think increase after each failed attempt.

I know on the one hand it's not best practice to upload mp4's direct and better to use a streaming server but there are obviously occasions where people do need to upload direct and 62MB as an example isn't exactly huge.

Sounds like non-default memory_limit or at least increase from 128MB has become a system requirement?

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of Pat L (pgogy)
Sent: 30 April 2014 21:03

To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

It's not the code, it's the file sizes.
MP4s will be bigger when zipped than they are as files
Example zipper is the temp file name the code uses, i assume when it runs out of memory it dumps the file.
There is a logic to doing a file size check before exporting and warning with large files


On Wed, Apr 30, 2014 at 7:58 PM, Ron Mitchell <ronm at mitchellmedia.co.uk<mailto:ronm at mitchellmedia.co.uk>> wrote:
Yep it's in 2.1 https://github.com/thexerteproject/xerteonlinetoolkits/blob/2.1/modules/xerte/export.php line 61<https://github.com/thexerteproject/xerteonlinetoolkits/blob/2.1/modules/xerte/export.php%20line%2061>

and in develop too:  https://github.com/thexerteproject/xerteonlinetoolkits/blob/develop/modules/xerte/export.php line 61<https://github.com/thexerteproject/xerteonlinetoolkits/blob/develop/modules/xerte/export.php%20line%2061>

I've been testing further on a server where I can change anything and it does seem like increasing memory_limit resolves the issue but I'm not sure what the equation would be for file size v required limit or if any of the other php settings have a part to play too. I can't be sure but this seems like a relatively new issue too.  My tests so far?

1. Small projects don't seem to be a problem - no example_zipper created and projects export ok.

2. I was alerted to the problem again by a user of the Techdis sandpit who reported not being able to export here project yet was able to previously. The particular project included a 62MB mp4 so that's what I've been testing with now but on a different server. On this server with the memory_limit set to 128MB example_zipper files were created and part of the exported zip as well as left on the server.

3. Increasing the memory_limit to 300MB enable the export to work and no example_zipper created.

4. I mentioned I had a similar issue with a Wordpress install recently where the server host had to increase the memory_limit and in that case to 256MB and that seems to be a regular problem especially for people on shared hosting where they can't change the memory limit themselves. 128MB seems to be the default but I guess it could vary quite a bit depending on hosts.

5. What I'm not sure about is if this is a new problem due to changes in Xerte code and whether it's been fixed previously as I think John suggested.?

6. There are so many variables here I can't test reliably without others testing too e.g. I don't know if the combined file size with lots of smaller files gives the same  result as this single 62MB file or if other PHP settings also have a part to play. My fear is that we don't know how big the problem is. Partly because export isn't as common as just creating and linking and that together with project size variation, settings variation etc and perhaps some not knowing that these files even exist means it could be quite common without many knowing.

Ron

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of Pat L (pgogy)
Sent: 30 April 2014 18:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Does exampler_zipper appear in the code, it sounds like something I left in.

I imagine it is a PHP timeout issue kicking in, or Memory being maxed out.

The zip script it supposed to do a clean up. Maybe do it before export now as well

On Wed, Apr 30, 2014 at 11:02 AM, Ron Mitchell <ronm at mitchellmedia.co.uk<mailto:ronm at mitchellmedia.co.uk>> wrote:
Hi all
I've updated some installations using the current 2.1 git download and had
hoped that the example_zipper issue I mentioned previously would be fixed
but it seems it hasn't? However it might only be showing up when large files
are included or with certain php or server configurations etc or perhaps
it's been fixed in develop but not applied to 2.1? This may also be an issue
going unoticed to some extent and leaving lots of redundant zip files in
project folders. So testing and questions...

What I'm finding with current 2.1 code on more than one server is that
exporting smaller projects works fine with no example_zipper files left on
the server or included in the zip download. However if a project includes
larger media files it either doesn't export or when it does it leaves
example_zipper files in the project folder as well as including them in the
zip download. I'm not sure where small - large overlaps but for instance
I've been testing export of a project containing a 60MB mp4.

Q1. Was this a known issue fixed in develop but not in current 2.1 branch?

Q2. Could this only be occurring with particular PHP/server settings and if
so what are the relevant settings?

Q3. Can others reproduce this using current 2.1 branch?

Cheers
Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of Smith, John
Sent: 08 April 2014 14:05
To: For Xerte technical developers
Subject: [Xerte-dev] Re: session issue + example_zipper issue

I've not really seen any session issues on a windows install and on xampp
mainly though. Have never seen the invalid templat id error either...

the example_zipper_new... file seems to be related to old export code
(https://code.google.com/p/xerteonlinetoolkits/source/browse/trunk/modules/x
erte/export.php?spec=svn696&r=696) with _xyz on the end nothing shows in
Google but remove _xyz and you get hits...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
Caledonian University Cowcaddens Road | Glasgow | G4 0BA
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
[xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of Julian Tenney
[Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>]
Sent: 08 April 2014 13:39
To: For Xerte technical developers
Subject: [Xerte-dev] Re: session issue + example_zipper issue

Never seen 'invalid template id'. But 'session ID not set' is doing my head
in.

I don't recognise 2.

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of Ron Mitchell
Sent: 08 April 2014 13:23
To: 'For Xerte technical developers'
Subject: [Xerte-dev] session issue + example_zipper issue

Hi all
I think an issue I mentioned in the thread entitled testing results got lost
in the discussion about re-running setup so I'm checking again along with a
separate issue I've noticed.

Issue 1
I downloaded xertetoolkits_2.1 Version:2.1 Last update  2014-03-14 from the
community site and from my testing this revision is not reliable e.g. one
minute I can create an LO and the next minute I get one of two errors when
trying to create a new lo:
Session ID not set
or
Invalid template_id (could not find in DB)

Q. Does this not happen for others?

Issue 2
In an older revision that I've been using which seems to be relatively
trouble free I've noticed that one or more files named something like
example_zipper_new_xyz. is getting created and included in any successful
export but also still residing in the project folder on the server.

Q. Does anyone recognise this? Was it fixed in a later revision that doesn't
have the session issue above?

Cheers
Ron




This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number
SC021474

Winner: Times Higher Education's Widening Participation Initiative of the
Year 2009 and Herald Society's Education Initiative of the Year 2009.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,6219,en
.html

Winner: Times Higher Education's Outstanding Support for Early Career
Researchers of the Year 2010, GCU as a lead with Universities Scotland
partners.
http://www.gcu.ac.uk/newsevents/news/bycategory/theuniversity/1/name,15691,e
n.html

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.






This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev



Glasgow Caledonian University is a registered Scottish charity, number SC021474
-------------- next part --------------
A non-text attachment was scrubbed...
Name: export.zip
Type: application/x-zip-compressed
Size: 2710 bytes
Desc: export.zip
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140501/280ca8e6/attachment.bin>

From david at palepurple.co.uk  Thu May  1 08:12:01 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Thu, 01 May 2014 08:12:01 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101AF@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <009701cf5325$4e6040e0$eb20c2a0$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CEBBA7B4B@EXCHANGE1.ad.nottingham.ac.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFADBDF5F@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<006b01cf645b$3b70e650$b252b2f0$@co.uk>
	<CAO2PinNOMKe0hBT4S8gNwCDcfoaLWKFu9NgoTEjKosqoa=_j2A@mail.gmail.com>
	<01ae01cf64a6$2aa18190$7fe484b0$@co.uk>
	<CAO2PinPKmj8T=-4p1WJ50mWz_hfg_4Curx8Qwc-==P8iHxdrYw@mail.gmail.com>
	<01e201cf64b1$12f26c20$38d74460$@co.uk>,
	<CAO2PinO6UbK_7t2UznUBQ9i_MWfa4C+eHwbz2Y1LANbxMofscA@mail.gmail.com>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101AF@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <5361F3C1.3070805@palepurple.co.uk>


<snip>
> Uses the unix command zip and pipes the stream back into PHP in 8192 
> chunks
>
<snip>
> Also, might not work on Windows...

So therefore it's not worth using?


Technically there is no benefit to doing the fread(....) over a 
fpassthru(...)
fpassthru does something like 8192 chunks internally.


David.

-- 
--
Pale Purple Ltd
http://www.palepurple.co.uk @PalePurpleLtd
0845 0046746 / 07792 380669
"Bespoke web and mobile application development"
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140501/1af11fbf/attachment.html>

From J.J.Smith at gcu.ac.uk  Thu May  1 08:37:41 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 08:37:41 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
Message-ID: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>

All i know Dave is that, with the commented code uncommented, my system couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no problem.

We may have to use what we have access to on different systems. There may be a similar way to pipe the result into windows or we fall back to the old code and crank the memory requirements up yo 300Mb which not everyone can most likely do anyway...

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII



David Goodwin <david at palepurple.co.uk> wrote:



<snip>
Uses the unix command zip and pipes the stream back into PHP in 8192 chunks

<snip>
Also, might not work on Windows...

So therefore it's not worth using?


Technically there is no benefit to doing the fread(....) over a fpassthru(...)
fpassthru does something like 8192 chunks internally.


David.

--
--
Pale Purple Ltd
http://www.palepurple.co.uk @PalePurpleLtd
0845 0046746 / 07792 380669
"Bespoke web and mobile application development"


This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number SC021474


From ronm at mitchellmedia.co.uk  Thu May  1 09:09:37 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Thu, 1 May 2014 09:09:37 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>
Message-ID: <002d01cf6514$b369cb40$1a3d61c0$@co.uk>

Hi John
I was just going to test your export code on a server where I don't have
access to change php settings so the memory limit is fixed at 128MB. I know
what David means that ideally we want cross platform code but as you say
there are inevitably different solutions/restrictions for some e.g. no
control over memory limit. In fact the upload limit on this particular
shared server is set at the default 2MB so I had to upload the test 62MB mp4
via ftp. However I'm now confused further because this is the same XOT code
as the other servers I was using for testing so I expected the export to
fail when testing before replacing the export code with yours and yet it
didn't fail and I was able to download the zip which didn't contain any
example_zipper files.

I'm glad you're able to reproduce the issue but not sure how a server with
128MB memory limit is able to export ok without a change of code or memory
limit :-(



-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 08:38
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: example_zipper issue revisited

All i know Dave is that, with the commented code uncommented, my system
couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no
problem.

We may have to use what we have access to on different systems. There may be
a similar way to pipe the result into windows or we fall back to the old
code and crank the memory requirements up yo 300Mb which not everyone can
most likely do anyway...

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII



David Goodwin <david at palepurple.co.uk> wrote:



<snip>
Uses the unix command zip and pipes the stream back into PHP in 8192 chunks

<snip>
Also, might not work on Windows...

So therefore it's not worth using?


Technically there is no benefit to doing the fread(....) over a
fpassthru(...) fpassthru does something like 8192 chunks internally.


David.

--
--
Pale Purple Ltd
http://www.palepurple.co.uk @PalePurpleLtd
0845 0046746 / 07792 380669
"Bespoke web and mobile application development"


This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.






From J.J.Smith at gcu.ac.uk  Thu May  1 09:28:50 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 09:28:50 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
Message-ID: <ibmy0eg959kqth82hsw9inef.1398932587378@email.android.com>

Hi Ron

I get what Dave is saying completely and agree that we need cross platform support. It must be the php zip library that is crashing while building the zip.

I think it will be interesting to get my new code tested on a server which was already failing to see if we can establish an upper limit. Because the 128Mb is php memory and its taking in 8k chunks and dumping straight to user then the memory might be more efficiently used or might even be able to use more than 128Mb, i have no idea.

Why yours doesn't fail though at 128 and others do though is a mystery but could be related to what else is on the box - php can ask for 128Mb but if that isn't available then it won't get it all.

You can watch the files via ftp. Watch a user folder then start a zip download. While download is in progress refresh the ftp and you should see all the files requiring zipping. They are delete by the code at very end so that isn't being reached on the problem servers.

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII



Ron Mitchell <ronm at mitchellmedia.co.uk> wrote:


Hi John
I was just going to test your export code on a server where I don't have
access to change php settings so the memory limit is fixed at 128MB. I know
what David means that ideally we want cross platform code but as you say
there are inevitably different solutions/restrictions for some e.g. no
control over memory limit. In fact the upload limit on this particular
shared server is set at the default 2MB so I had to upload the test 62MB mp4
via ftp. However I'm now confused further because this is the same XOT code
as the other servers I was using for testing so I expected the export to
fail when testing before replacing the export code with yours and yet it
didn't fail and I was able to download the zip which didn't contain any
example_zipper files.

I'm glad you're able to reproduce the issue but not sure how a server with
128MB memory limit is able to export ok without a change of code or memory
limit :-(



-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 08:38
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: example_zipper issue revisited

All i know Dave is that, with the commented code uncommented, my system
couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no
problem.

We may have to use what we have access to on different systems. There may be
a similar way to pipe the result into windows or we fall back to the old
code and crank the memory requirements up yo 300Mb which not everyone can
most likely do anyway...

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII



David Goodwin <david at palepurple.co.uk> wrote:



<snip>
Uses the unix command zip and pipes the stream back into PHP in 8192 chunks

<snip>
Also, might not work on Windows...

So therefore it's not worth using?


Technically there is no benefit to doing the fread(....) over a
fpassthru(...) fpassthru does something like 8192 chunks internally.


David.

--
--
Pale Purple Ltd
http://www.palepurple.co.uk @PalePurpleLtd
0845 0046746 / 07792 380669
"Bespoke web and mobile application development"


This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From xerte at pgogywebstuff.com  Thu May  1 09:44:35 2014
From: xerte at pgogywebstuff.com (Pat L (pgogy))
Date: Thu, 1 May 2014 09:44:35 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>
Message-ID: <CAO2PinMGJr8fKpXhK1=DQT-5KvVPxC5uUpXi20=bPPpLHp+3vQ@mail.gmail.com>

Do we still use the old zip library, or the native php stuff.
Native php might be better? Or another class?
On 1 May 2014 08:37, "Smith, John" <J.J.Smith at gcu.ac.uk> wrote:

> All i know Dave is that, with the commented code uncommented, my system
> couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no
> problem.
>
> We may have to use what we have access to on different systems. There may
> be a similar way to pipe the result into windows or we fall back to the old
> code and crank the memory requirements up yo 300Mb which not everyone can
> most likely do anyway...
>
> Regards
>
> John Smith
> Learning Technologist
> School of Health and Life Sciences
>
> Sent from Samsung Galaxy SII
>
>
>
> David Goodwin <david at palepurple.co.uk> wrote:
>
>
>
> <snip>
> Uses the unix command zip and pipes the stream back into PHP in 8192 chunks
>
> <snip>
> Also, might not work on Windows...
>
> So therefore it's not worth using?
>
>
> Technically there is no benefit to doing the fread(....) over a
> fpassthru(...)
> fpassthru does something like 8192 chunks internally.
>
>
> David.
>
> --
> --
> Pale Purple Ltd
> http://www.palepurple.co.uk @PalePurpleLtd
> 0845 0046746 / 07792 380669
> "Bespoke web and mobile application development"
>
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer
> system, you are advised to perform your own checks. Email communications
> with the University of Nottingham may be monitored as permitted by UK
> legislation.
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140501/dc1ead5d/attachment.html>

From david at palepurple.co.uk  Thu May  1 09:56:30 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Thu, 01 May 2014 09:56:30 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <002d01cf6514$b369cb40$1a3d61c0$@co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
Message-ID: <53620C3E.2060704@palepurple.co.uk>



On 01/05/14 09:09, Ron Mitchell wrote:
> Hi John
> I was just going to test your export code on a server where I don't have
> access to change php settings so the memory limit is fixed at 128MB. I know
> what David means that ideally we want cross platform code but as you say
> there are inevitably different solutions/restrictions for some e.g. no
> control over memory limit. In fact the upload limit on this particular
> shared server is set at the default 2MB so I had to upload the test 62MB mp4
> via ftp. However I'm now confused further because this is the same XOT code
> as the other servers I was using for testing so I expected the export to
> fail when testing before replacing the export code with yours and yet it
> didn't fail and I was able to download the zip which didn't contain any
> example_zipper files.
>
> I'm glad you're able to reproduce the issue but not sure how a server with
> 128MB memory limit is able to export ok without a change of code or memory
> limit :-(
>
>

John's code for export.php avoids using php/scorm/archive.php - or at 
least doesn't use it fully.

My understanding is that :

a. PHP's default memory limit has been 128mb for a few years now (since 
around version 5.2 I think).... and I'd have thought it to be reasonable 
to expect any new installs to have a memory limit of around this value.


b. PHP has a 'zip' extension - which is probably installed by default 
(perhaps?). If so, this ought to be better to use than the 
php/scorm/archive.php variant.


c. I don't think anyone's actually done any profiling of the code yet to 
get an idea of what is gobbling up memory.


David.


-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From J.J.Smith at gcu.ac.uk  Thu May  1 09:57:25 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 09:57:25 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <002d01cf6514$b369cb40$1a3d61c0$@co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Hi Ron,

Have been thinking about your issue and just wondering whether you might have inadvertently chosen the worst file size possible for the video with 128Mb memory available to PHP. Let me explain (and this is entirely hypothetical)...

The zipper creates a blank object (<0.1Mb memory used)
The library loads all the other files individually and zips them into the object (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or under)
The library then loads the mp4 (now it has the object = 2Mb and the video = 62Mb in memory)
It adds it to the zip object but it's also still in memory (mp4 doesn't compress at all so we have 2 + 62 + 62 =126Mb which is less than 128Mb)
It then closes the connection to the mp4 and destroys the object (memory now = 64Mb for the zip)

I'm wondering whether a 64 Mb or slightly bigger will cause you to be able to crash that server too. Is that something you could try?

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell [ronm at mitchellmedia.co.uk]
Sent: 01 May 2014 09:09
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi John
I was just going to test your export code on a server where I don't have
access to change php settings so the memory limit is fixed at 128MB. I know
what David means that ideally we want cross platform code but as you say
there are inevitably different solutions/restrictions for some e.g. no
control over memory limit. In fact the upload limit on this particular
shared server is set at the default 2MB so I had to upload the test 62MB mp4
via ftp. However I'm now confused further because this is the same XOT code
as the other servers I was using for testing so I expected the export to
fail when testing before replacing the export code with yours and yet it
didn't fail and I was able to download the zip which didn't contain any
example_zipper files.

I'm glad you're able to reproduce the issue but not sure how a server with
128MB memory limit is able to export ok without a change of code or memory
limit :-(



-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 08:38
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: example_zipper issue revisited

All i know Dave is that, with the commented code uncommented, my system
couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no
problem.

We may have to use what we have access to on different systems. There may be
a similar way to pipe the result into windows or we fall back to the old
code and crank the memory requirements up yo 300Mb which not everyone can
most likely do anyway...

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII



David Goodwin <david at palepurple.co.uk> wrote:



<snip>
Uses the unix command zip and pipes the stream back into PHP in 8192 chunks

<snip>
Also, might not work on Windows...

So therefore it's not worth using?


Technically there is no benefit to doing the fread(....) over a
fpassthru(...) fpassthru does something like 8192 chunks internally.


David.

--
--
Pale Purple Ltd
http://www.palepurple.co.uk @PalePurpleLtd
0845 0046746 / 07792 380669
"Bespoke web and mobile application development"


This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From J.J.Smith at gcu.ac.uk  Thu May  1 10:13:54 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 10:13:54 +0100
Subject: [Xerte-dev] Re: Fancy it
In-Reply-To: <CAO2PinPefgFjO7SkJSUYeNhTHLxj51QTRZ=s6GCz_PCqPnHrEQ@mail.gmail.com>
References: <CAO2PinPefgFjO7SkJSUYeNhTHLxj51QTRZ=s6GCz_PCqPnHrEQ@mail.gmail.com>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B9@ITSEMBXCLUS.enterprise.gcal.ac.uk>

I think for some areas it would be great, mainly the modules folder which contains a lot of client side stuff like xenith and scorms stuff. Problems with them often sit dormant until a particular page model or browser is used...

I think Ron has made a good point aginst this in the past for a blanket auto update though in that people have tweaked some of the php files to do speciic things and we could break some peoples installs. I think that was my idea around having Wordpress style hooks in xerte. It means people can keep their code seperate from the core and then, if properly managed, their code should work on newer versions.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Pat L (pgogy) [xerte at pgogywebstuff.com]
Sent: 30 April 2014 21:25
To: For Xerte technical developers
Subject: [Xerte-dev] Fancy it

http://www.phpclasses.org/blog/post/233-Automatic-Update-of-Packages-from-Version-Control-Repositories.html

auto update from git?

Glasgow Caledonian University is a registered Scottish charity, number SC021474


From david at palepurple.co.uk  Thu May  1 10:23:56 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Thu, 01 May 2014 10:23:56 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <536212AC.1070202@palepurple.co.uk>

Hi,

> The zipper creates a blank object (<0.1Mb memory used)
> The library loads all the other files individually and zips them into the object (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or under)
> The library then loads the mp4 (now it has the object = 2Mb and the video = 62Mb in memory)

Except it's not in memory.

The zip_file class has an option set (inmemory = 0) which should make it 
write to a temporary file (and not store the zip itself in memory).


David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From J.J.Smith at gcu.ac.uk  Thu May  1 10:31:58 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 10:31:58 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <536212AC.1070202@palepurple.co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<536212AC.1070202@palepurple.co.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101BA@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Fair enough, I don't know enough about how it works to know exactly what is going on under the hood and it was hypothetical...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin [david at palepurple.co.uk]
Sent: 01 May 2014 10:23
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi,

> The zipper creates a blank object (<0.1Mb memory used)
> The library loads all the other files individually and zips them into the object (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or under)
> The library then loads the mp4 (now it has the object = 2Mb and the video = 62Mb in memory)

Except it's not in memory.

The zip_file class has an option set (inmemory = 0) which should make it
write to a temporary file (and not store the zip itself in memory).


David.

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From david at palepurple.co.uk  Thu May  1 10:39:17 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Thu, 01 May 2014 10:39:17 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101BA@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<536212AC.1070202@palepurple.co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101BA@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <53621645.4000403@palepurple.co.uk>



On 01/05/14 10:31, Smith, John wrote:
> Fair enough, I don't know enough about how it works to know exactly what is going on under the hood and it was hypothetical...


Well - something is misbehaving there. And it's probably something to do 
with opening/writing/creating the zip.... and evidence points to it 
being related to adding large files - so I suspect you're on the right 
track .... it's just not obvious where the issue is.

It's not like the code is doing :

foreach($files as $file) {
      $zip .= file_get_contents($file);
}

Which would at least be easy to  spot - and fix!

I'll see if we can replicate it internally ...


David.


-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From ronm at mitchellmedia.co.uk  Thu May  1 10:58:43 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Thu, 1 May 2014 10:58:43 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <007801cf6523$f7747920$e65d6b60$@co.uk>

Hi John
the problem is when I originally raised this issue on 8th April it coincided
with discovering loads of these un-removed example_zipper files on the
sandpit install. At that time I also mentioned the separate session issue as
a reason I hadn't updated to a newer revision and I think you mentioned that
you thought Tom had fixed this zipper issue in a later revision. So once
we'd identified the cause of the session issue (or at least Julian's attempt
at fixing his FF issue) I updated the sandpit code as well as other servers.
Now I've been using the 2.1 branch for this rather than develop so I'm not
sure if Tom's fix is in develop but not 2.1 but I think the code is the same
at the moment.

In short although I've been using a 62MB file from a project a user was
unable to export, previously there were loads of these zipper files in loads
of different user/project folders and there are probably quite a few again
now prior to my changing the memory limit.

I could try resetting the memory_limit back on my own server and test again
but my previous tests with the 62MB file certainly weren't crashing the
server. Let me know if you think its worth testing with larger files and the
default 128MB limit or if your and David's later responses have ruled that
out as being worthwhile? 

I know Tom said last night that he was going to revisit his fix for this so
would be good to hear back from Tom once he's done that.

Cheers
Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 09:57
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi Ron,

Have been thinking about your issue and just wondering whether you might
have inadvertently chosen the worst file size possible for the video with
128Mb memory available to PHP. Let me explain (and this is entirely
hypothetical)...

The zipper creates a blank object (<0.1Mb memory used) The library loads all
the other files individually and zips them into the object (0Mb ->
compressed size of all js/html is perhaps around 1-2Mb or under) The library
then loads the mp4 (now it has the object = 2Mb and the video = 62Mb in
memory) It adds it to the zip object but it's also still in memory (mp4
doesn't compress at all so we have 2 + 62 + 62 =126Mb which is less than
128Mb) It then closes the connection to the mp4 and destroys the object
(memory now = 64Mb for the zip)

I'm wondering whether a 64 Mb or slightly bigger will cause you to be able
to crash that server too. Is that something you could try?

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk
[xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
[ronm at mitchellmedia.co.uk]
Sent: 01 May 2014 09:09
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi John
I was just going to test your export code on a server where I don't have
access to change php settings so the memory limit is fixed at 128MB. I know
what David means that ideally we want cross platform code but as you say
there are inevitably different solutions/restrictions for some e.g. no
control over memory limit. In fact the upload limit on this particular
shared server is set at the default 2MB so I had to upload the test 62MB mp4
via ftp. However I'm now confused further because this is the same XOT code
as the other servers I was using for testing so I expected the export to
fail when testing before replacing the export code with yours and yet it
didn't fail and I was able to download the zip which didn't contain any
example_zipper files.

I'm glad you're able to reproduce the issue but not sure how a server with
128MB memory limit is able to export ok without a change of code or memory
limit :-(



-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 08:38
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: example_zipper issue revisited

All i know Dave is that, with the commented code uncommented, my system
couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no
problem.

We may have to use what we have access to on different systems. There may be
a similar way to pipe the result into windows or we fall back to the old
code and crank the memory requirements up yo 300Mb which not everyone can
most likely do anyway...

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII



David Goodwin <david at palepurple.co.uk> wrote:



<snip>
Uses the unix command zip and pipes the stream back into PHP in 8192 chunks

<snip>
Also, might not work on Windows...

So therefore it's not worth using?


Technically there is no benefit to doing the fread(....) over a
fpassthru(...) fpassthru does something like 8192 chunks internally.


David.

--
--
Pale Purple Ltd
http://www.palepurple.co.uk @PalePurpleLtd
0845 0046746 / 07792 380669
"Bespoke web and mobile application development"


This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.






From J.J.Smith at gcu.ac.uk  Thu May  1 11:16:45 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 11:16:45 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <007801cf6523$f7747920$e65d6b60$@co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101BD@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Hi Ron,

I don't know if it's worth keeping testing different eventualities until we have a better idea of what is going on. The problem with these zipper files is that each time a new one is created and not deleted then it will be included in further zip attempts because it shared the folder with the copied files. I did read somewhere that the zip command that I am using ignores files/folders begin with a . (dot) and perhaps the php library can be made to do similar which would at least keep the temporary zipper folders out of the downloaded zip.

I'm a bit busy this morning but will in the afternoon try to test how the inclusion of different sized files correlates to memory limits and see if I can predict whether it will crash the script like the example that I based on your test case.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell [ronm at mitchellmedia.co.uk]
Sent: 01 May 2014 10:58
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi John
the problem is when I originally raised this issue on 8th April it coincided
with discovering loads of these un-removed example_zipper files on the
sandpit install. At that time I also mentioned the separate session issue as
a reason I hadn't updated to a newer revision and I think you mentioned that
you thought Tom had fixed this zipper issue in a later revision. So once
we'd identified the cause of the session issue (or at least Julian's attempt
at fixing his FF issue) I updated the sandpit code as well as other servers.
Now I've been using the 2.1 branch for this rather than develop so I'm not
sure if Tom's fix is in develop but not 2.1 but I think the code is the same
at the moment.

In short although I've been using a 62MB file from a project a user was
unable to export, previously there were loads of these zipper files in loads
of different user/project folders and there are probably quite a few again
now prior to my changing the memory limit.

I could try resetting the memory_limit back on my own server and test again
but my previous tests with the 62MB file certainly weren't crashing the
server. Let me know if you think its worth testing with larger files and the
default 128MB limit or if your and David's later responses have ruled that
out as being worthwhile?

I know Tom said last night that he was going to revisit his fix for this so
would be good to hear back from Tom once he's done that.

Cheers
Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 09:57
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi Ron,

Have been thinking about your issue and just wondering whether you might
have inadvertently chosen the worst file size possible for the video with
128Mb memory available to PHP. Let me explain (and this is entirely
hypothetical)...

The zipper creates a blank object (<0.1Mb memory used) The library loads all
the other files individually and zips them into the object (0Mb ->
compressed size of all js/html is perhaps around 1-2Mb or under) The library
then loads the mp4 (now it has the object = 2Mb and the video = 62Mb in
memory) It adds it to the zip object but it's also still in memory (mp4
doesn't compress at all so we have 2 + 62 + 62 =126Mb which is less than
128Mb) It then closes the connection to the mp4 and destroys the object
(memory now = 64Mb for the zip)

I'm wondering whether a 64 Mb or slightly bigger will cause you to be able
to crash that server too. Is that something you could try?

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk
[xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
[ronm at mitchellmedia.co.uk]
Sent: 01 May 2014 09:09
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi John
I was just going to test your export code on a server where I don't have
access to change php settings so the memory limit is fixed at 128MB. I know
what David means that ideally we want cross platform code but as you say
there are inevitably different solutions/restrictions for some e.g. no
control over memory limit. In fact the upload limit on this particular
shared server is set at the default 2MB so I had to upload the test 62MB mp4
via ftp. However I'm now confused further because this is the same XOT code
as the other servers I was using for testing so I expected the export to
fail when testing before replacing the export code with yours and yet it
didn't fail and I was able to download the zip which didn't contain any
example_zipper files.

I'm glad you're able to reproduce the issue but not sure how a server with
128MB memory limit is able to export ok without a change of code or memory
limit :-(



-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 08:38
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: example_zipper issue revisited

All i know Dave is that, with the commented code uncommented, my system
couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no
problem.

We may have to use what we have access to on different systems. There may be
a similar way to pipe the result into windows or we fall back to the old
code and crank the memory requirements up yo 300Mb which not everyone can
most likely do anyway...

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII



David Goodwin <david at palepurple.co.uk> wrote:



<snip>
Uses the unix command zip and pipes the stream back into PHP in 8192 chunks

<snip>
Also, might not work on Windows...

So therefore it's not worth using?


Technically there is no benefit to doing the fread(....) over a
fpassthru(...) fpassthru does something like 8192 chunks internally.


David.

--
--
Pale Purple Ltd
http://www.palepurple.co.uk @PalePurpleLtd
0845 0046746 / 07792 380669
"Bespoke web and mobile application development"


This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From xerte at pgogywebstuff.com  Thu May  1 12:10:56 2014
From: xerte at pgogywebstuff.com (Pat L (pgogy))
Date: Thu, 1 May 2014 12:10:56 +0100
Subject: [Xerte-dev] Re: Fancy it
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B9@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <CAO2PinPefgFjO7SkJSUYeNhTHLxj51QTRZ=s6GCz_PCqPnHrEQ@mail.gmail.com>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B9@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <CAO2PinM=uUriyV04NaiBCaHrg8bRN98W99Hy_y_nOXDygt0PyA@mail.gmail.com>

Question is trade off of modifications vs "You need to upgrade" problems
Logically plugins solves this - you shouldn't really play with core


On Thu, May 1, 2014 at 10:13 AM, Smith, John <J.J.Smith at gcu.ac.uk> wrote:

> I think for some areas it would be great, mainly the modules folder which
> contains a lot of client side stuff like xenith and scorms stuff. Problems
> with them often sit dormant until a particular page model or browser is
> used...
>
> I think Ron has made a good point aginst this in the past for a blanket
> auto update though in that people have tweaked some of the php files to do
> speciic things and we could break some peoples installs. I think that was
> my idea around having Wordpress style hooks in xerte. It means people can
> keep their code seperate from the core and then, if properly managed, their
> code should work on newer versions.
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
> Glasgow Caledonian University
> Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
> School Learning Technologists will pick up the job. This will ensure that
> all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk [
> xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Pat L (pgogy) [
> xerte at pgogywebstuff.com]
> Sent: 30 April 2014 21:25
> To: For Xerte technical developers
> Subject: [Xerte-dev] Fancy it
>
>
> http://www.phpclasses.org/blog/post/233-Automatic-Update-of-Packages-from-Version-Control-Repositories.html
>
> auto update from git?
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140501/56aae730/attachment.html>

From J.J.Smith at gcu.ac.uk  Thu May  1 12:32:33 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 12:32:33 +0100
Subject: [Xerte-dev] Re: Fancy it
In-Reply-To: <CAO2PinM=uUriyV04NaiBCaHrg8bRN98W99Hy_y_nOXDygt0PyA@mail.gmail.com>
References: <CAO2PinPefgFjO7SkJSUYeNhTHLxj51QTRZ=s6GCz_PCqPnHrEQ@mail.gmail.com>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B9@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<CAO2PinM=uUriyV04NaiBCaHrg8bRN98W99Hy_y_nOXDygt0PyA@mail.gmail.com>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101BE@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Well we all know that but guys like RonM2 who need specific functionality have no other way to do it.. then ended up locked at v1.7 for a while... i'm in the process of rewriting Xenith to use a pub/sub approach and at some point I will probably revisit the plugins idea as there are many other areas that i'd like to provide hook points...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Pat L (pgogy) [xerte at pgogywebstuff.com]
Sent: 01 May 2014 12:10
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Fancy it

Question is trade off of modifications vs "You need to upgrade" problems
Logically plugins solves this - you shouldn't really play with core


On Thu, May 1, 2014 at 10:13 AM, Smith, John <J.J.Smith at gcu.ac.uk<mailto:J.J.Smith at gcu.ac.uk>> wrote:
I think for some areas it would be great, mainly the modules folder which contains a lot of client side stuff like xenith and scorms stuff. Problems with them often sit dormant until a particular page model or browser is used...

I think Ron has made a good point aginst this in the past for a blanket auto update though in that people have tweaked some of the php files to do speciic things and we could break some peoples installs. I think that was my idea around having Wordpress style hooks in xerte. It means people can keep their code seperate from the core and then, if properly managed, their code should work on newer versions.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk<mailto:hlsblt at gcu.ac.uk> where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of Pat L (pgogy) [xerte at pgogywebstuff.com<mailto:xerte at pgogywebstuff.com>]
Sent: 30 April 2014 21:25
To: For Xerte technical developers
Subject: [Xerte-dev] Fancy it

http://www.phpclasses.org/blog/post/233-Automatic-Update-of-Packages-from-Version-Control-Repositories.html

auto update from git?

Glasgow Caledonian University is a registered Scottish charity, number SC021474


Glasgow Caledonian University is a registered Scottish charity, number SC021474


From ronm at mitchellmedia.co.uk  Thu May  1 13:55:26 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Thu, 1 May 2014 13:55:26 +0100
Subject: [Xerte-dev] Re: Fancy it
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101BE@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <CAO2PinPefgFjO7SkJSUYeNhTHLxj51QTRZ=s6GCz_PCqPnHrEQ@mail.gmail.com>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B9@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<CAO2PinM=uUriyV04NaiBCaHrg8bRN98W99Hy_y_nOXDygt0PyA@mail.gmail.com>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101BE@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <00d801cf653c$a6cdebf0$f469c3d0$@co.uk>

There are all sorts of reason to avoid auto-updating exactly because there
isn't really a clear difference between core and some of the things people
want/need to customise. The most common but not only customisation often
required is themes and although there are ways to achieve that without
changing any core files it's a bit more complex than that especially as
we've had very different versions over time too. 

I agree this could be desirable and worth planning for at some future point
but there are other customisations that I'm aware of which at the moment can
only be achieved by customisation of core files and not really a set list
that we could say exclude from auto updates as a pre-defined group. A simple
example is static.php and I know that could easily be part of a pre-defined
exclusion list but that's just a simple example.

So to achieve this I think there would need to be some fundamental changes
and/or the ability for an admin to easily control exclusions.

Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 12:33
To: xerte at pgogywebstuff.com; For Xerte technical developers
Subject: [Xerte-dev] Re: Fancy it

Well we all know that but guys like RonM2 who need specific functionality
have no other way to do it.. then ended up locked at v1.7 for a while... i'm
in the process of rewriting Xenith to use a pub/sub approach and at some
point I will probably revisit the plugins idea as there are many other areas
that i'd like to provide hook points...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk
[xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Pat L (pgogy)
[xerte at pgogywebstuff.com]
Sent: 01 May 2014 12:10
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Fancy it

Question is trade off of modifications vs "You need to upgrade" problems
Logically plugins solves this - you shouldn't really play with core


On Thu, May 1, 2014 at 10:13 AM, Smith, John
<J.J.Smith at gcu.ac.uk<mailto:J.J.Smith at gcu.ac.uk>> wrote:
I think for some areas it would be great, mainly the modules folder which
contains a lot of client side stuff like xenith and scorms stuff. Problems
with them often sit dormant until a particular page model or browser is
used...

I think Ron has made a good point aginst this in the past for a blanket auto
update though in that people have tweaked some of the php files to do
speciic things and we could break some peoples installs. I think that was my
idea around having Wordpress style hooks in xerte. It means people can keep
their code seperate from the core and then, if properly managed, their code
should work on newer versions.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to
hlsblt at gcu.ac.uk<mailto:hlsblt at gcu.ac.uk> where one of the School Learning
Technologists will pick up the job. This will ensure that all jobs are
completed as promptly as possible.
________________________________________
From:
xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nott
ingham.ac.uk>
[xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.not
tingham.ac.uk>] On Behalf Of Pat L (pgogy)
[xerte at pgogywebstuff.com<mailto:xerte at pgogywebstuff.com>]
Sent: 30 April 2014 21:25
To: For Xerte technical developers
Subject: [Xerte-dev] Fancy it

http://www.phpclasses.org/blog/post/233-Automatic-Update-of-Packages-from-Ve
rsion-Control-Repositories.html

auto update from git?

Glasgow Caledonian University is a registered Scottish charity, number
SC021474


Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.






From J.J.Smith at gcu.ac.uk  Thu May  1 14:38:11 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 14:38:11 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <007801cf6523$f7747920$e65d6b60$@co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Well. Preliminary testing looks good with my new code but i haven't tested on Windows yet - would be interested in getting Tom's take on all of this...

I just added loads of videos via ftp (upload limit set to 7M on this server) and checked the memory limit and it is

memory_limit    90M     90M

and with each video being used on a page then i've just successfully downloaded a 316Mb archive zip without any problems (it took several minutes too so the script didn't timeout) and was able to play the videos contained within... and this is just a cheap Dreamhost shared server, nothing fancy...

On to Windoze...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell [ronm at mitchellmedia.co.uk]
Sent: 01 May 2014 10:58
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi John
the problem is when I originally raised this issue on 8th April it coincided
with discovering loads of these un-removed example_zipper files on the
sandpit install. At that time I also mentioned the separate session issue as
a reason I hadn't updated to a newer revision and I think you mentioned that
you thought Tom had fixed this zipper issue in a later revision. So once
we'd identified the cause of the session issue (or at least Julian's attempt
at fixing his FF issue) I updated the sandpit code as well as other servers.
Now I've been using the 2.1 branch for this rather than develop so I'm not
sure if Tom's fix is in develop but not 2.1 but I think the code is the same
at the moment.

In short although I've been using a 62MB file from a project a user was
unable to export, previously there were loads of these zipper files in loads
of different user/project folders and there are probably quite a few again
now prior to my changing the memory limit.

I could try resetting the memory_limit back on my own server and test again
but my previous tests with the 62MB file certainly weren't crashing the
server. Let me know if you think its worth testing with larger files and the
default 128MB limit or if your and David's later responses have ruled that
out as being worthwhile?

I know Tom said last night that he was going to revisit his fix for this so
would be good to hear back from Tom once he's done that.

Cheers
Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 09:57
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi Ron,

Have been thinking about your issue and just wondering whether you might
have inadvertently chosen the worst file size possible for the video with
128Mb memory available to PHP. Let me explain (and this is entirely
hypothetical)...

The zipper creates a blank object (<0.1Mb memory used) The library loads all
the other files individually and zips them into the object (0Mb ->
compressed size of all js/html is perhaps around 1-2Mb or under) The library
then loads the mp4 (now it has the object = 2Mb and the video = 62Mb in
memory) It adds it to the zip object but it's also still in memory (mp4
doesn't compress at all so we have 2 + 62 + 62 =126Mb which is less than
128Mb) It then closes the connection to the mp4 and destroys the object
(memory now = 64Mb for the zip)

I'm wondering whether a 64 Mb or slightly bigger will cause you to be able
to crash that server too. Is that something you could try?

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk
[xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
[ronm at mitchellmedia.co.uk]
Sent: 01 May 2014 09:09
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi John
I was just going to test your export code on a server where I don't have
access to change php settings so the memory limit is fixed at 128MB. I know
what David means that ideally we want cross platform code but as you say
there are inevitably different solutions/restrictions for some e.g. no
control over memory limit. In fact the upload limit on this particular
shared server is set at the default 2MB so I had to upload the test 62MB mp4
via ftp. However I'm now confused further because this is the same XOT code
as the other servers I was using for testing so I expected the export to
fail when testing before replacing the export code with yours and yet it
didn't fail and I was able to download the zip which didn't contain any
example_zipper files.

I'm glad you're able to reproduce the issue but not sure how a server with
128MB memory limit is able to export ok without a change of code or memory
limit :-(



-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 08:38
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: example_zipper issue revisited

All i know Dave is that, with the commented code uncommented, my system
couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no
problem.

We may have to use what we have access to on different systems. There may be
a similar way to pipe the result into windows or we fall back to the old
code and crank the memory requirements up yo 300Mb which not everyone can
most likely do anyway...

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII



David Goodwin <david at palepurple.co.uk> wrote:



<snip>
Uses the unix command zip and pipes the stream back into PHP in 8192 chunks

<snip>
Also, might not work on Windows...

So therefore it's not worth using?


Technically there is no benefit to doing the fread(....) over a
fpassthru(...) fpassthru does something like 8192 chunks internally.


David.

--
--
Pale Purple Ltd
http://www.palepurple.co.uk @PalePurpleLtd
0845 0046746 / 07792 380669
"Bespoke web and mobile application development"


This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474


From reijnders at tor.nl  Thu May  1 14:46:15 2014
From: reijnders at tor.nl (Tom Reijnders)
Date: Thu, 01 May 2014 15:46:15 +0200
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <53625027.2030801@tor.nl>

I will look into this, but I don't have much time at the moment. Perhaps 
topnight. Otherwise, it'll have to be next week.

Tom

Smith, John schreef op 1-5-2014 15:38:
> Well. Preliminary testing looks good with my new code but i haven't tested on Windows yet - would be interested in getting Tom's take on all of this...
>
> I just added loads of videos via ftp (upload limit set to 7M on this server) and checked the memory limit and it is
>
> memory_limit    90M     90M
>
> and with each video being used on a page then i've just successfully downloaded a 316Mb archive zip without any problems (it took several minutes too so the script didn't timeout) and was able to play the videos contained within... and this is just a cheap Dreamhost shared server, nothing fancy...
>
> On to Windoze...
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
> Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 10:58
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> the problem is when I originally raised this issue on 8th April it coincided
> with discovering loads of these un-removed example_zipper files on the
> sandpit install. At that time I also mentioned the separate session issue as
> a reason I hadn't updated to a newer revision and I think you mentioned that
> you thought Tom had fixed this zipper issue in a later revision. So once
> we'd identified the cause of the session issue (or at least Julian's attempt
> at fixing his FF issue) I updated the sandpit code as well as other servers.
> Now I've been using the 2.1 branch for this rather than develop so I'm not
> sure if Tom's fix is in develop but not 2.1 but I think the code is the same
> at the moment.
>
> In short although I've been using a 62MB file from a project a user was
> unable to export, previously there were loads of these zipper files in loads
> of different user/project folders and there are probably quite a few again
> now prior to my changing the memory limit.
>
> I could try resetting the memory_limit back on my own server and test again
> but my previous tests with the 62MB file certainly weren't crashing the
> server. Let me know if you think its worth testing with larger files and the
> default 128MB limit or if your and David's later responses have ruled that
> out as being worthwhile?
>
> I know Tom said last night that he was going to revisit his fix for this so
> would be good to hear back from Tom once he's done that.
>
> Cheers
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
> Sent: 01 May 2014 09:57
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi Ron,
>
> Have been thinking about your issue and just wondering whether you might
> have inadvertently chosen the worst file size possible for the video with
> 128Mb memory available to PHP. Let me explain (and this is entirely
> hypothetical)...
>
> The zipper creates a blank object (<0.1Mb memory used) The library loads all
> the other files individually and zips them into the object (0Mb ->
> compressed size of all js/html is perhaps around 1-2Mb or under) The library
> then loads the mp4 (now it has the object = 2Mb and the video = 62Mb in
> memory) It adds it to the zip object but it's also still in memory (mp4
> doesn't compress at all so we have 2 + 62 + 62 =126Mb which is less than
> 128Mb) It then closes the connection to the mp4 and destroys the object
> (memory now = 64Mb for the zip)
>
> I'm wondering whether a 64 Mb or slightly bigger will cause you to be able
> to crash that server too. Is that something you could try?
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
> Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
> School Learning Technologists will pick up the job. This will ensure that
> all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 09:09
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> I was just going to test your export code on a server where I don't have
> access to change php settings so the memory limit is fixed at 128MB. I know
> what David means that ideally we want cross platform code but as you say
> there are inevitably different solutions/restrictions for some e.g. no
> control over memory limit. In fact the upload limit on this particular
> shared server is set at the default 2MB so I had to upload the test 62MB mp4
> via ftp. However I'm now confused further because this is the same XOT code
> as the other servers I was using for testing so I expected the export to
> fail when testing before replacing the export code with yours and yet it
> didn't fail and I was able to download the zip which didn't contain any
> example_zipper files.
>
> I'm glad you're able to reproduce the issue but not sure how a server with
> 128MB memory limit is able to export ok without a change of code or memory
> limit :-(
>
>
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
> Sent: 01 May 2014 08:38
> To: xerte-dev at lists.nottingham.ac.uk
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> All i know Dave is that, with the commented code uncommented, my system
> couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no
> problem.
>
> We may have to use what we have access to on different systems. There may be
> a similar way to pipe the result into windows or we fall back to the old
> code and crank the memory requirements up yo 300Mb which not everyone can
> most likely do anyway...
>
> Regards
>
> John Smith
> Learning Technologist
> School of Health and Life Sciences
>
> Sent from Samsung Galaxy SII
>
>
>
> David Goodwin <david at palepurple.co.uk> wrote:
>
>
>
> <snip>
> Uses the unix command zip and pipes the stream back into PHP in 8192 chunks
>
> <snip>
> Also, might not work on Windows...
>
> So therefore it's not worth using?
>
>
> Technically there is no benefit to doing the fread(....) over a
> fpassthru(...) fpassthru does something like 8192 chunks internally.
>
>
> David.
>
> --
> --
> Pale Purple Ltd
> http://www.palepurple.co.uk @PalePurpleLtd
> 0845 0046746 / 07792 380669
> "Bespoke web and mobile application development"
>
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>

-- 
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196



From J.J.Smith at gcu.ac.uk  Thu May  1 14:51:10 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 1 May 2014 14:51:10 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <53625027.2030801@tor.nl>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<53625027.2030801@tor.nl>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Hi Tom,

Yes, definitely, no worries. I'm about to investigate a windows install with both the original code and this new code to see whether I can get it to run. Will update you.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders [reijnders at tor.nl]
Sent: 01 May 2014 14:46
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

I will look into this, but I don't have much time at the moment. Perhaps
topnight. Otherwise, it'll have to be next week.

Tom

Smith, John schreef op 1-5-2014 15:38:
> Well. Preliminary testing looks good with my new code but i haven't tested on Windows yet - would be interested in getting Tom's take on all of this...
>
> I just added loads of videos via ftp (upload limit set to 7M on this server) and checked the memory limit and it is
>
> memory_limit    90M     90M
>
> and with each video being used on a page then i've just successfully downloaded a 316Mb archive zip without any problems (it took several minutes too so the script didn't timeout) and was able to play the videos contained within... and this is just a cheap Dreamhost shared server, nothing fancy...
>
> On to Windoze...
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
> Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 10:58
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> the problem is when I originally raised this issue on 8th April it coincided
> with discovering loads of these un-removed example_zipper files on the
> sandpit install. At that time I also mentioned the separate session issue as
> a reason I hadn't updated to a newer revision and I think you mentioned that
> you thought Tom had fixed this zipper issue in a later revision. So once
> we'd identified the cause of the session issue (or at least Julian's attempt
> at fixing his FF issue) I updated the sandpit code as well as other servers.
> Now I've been using the 2.1 branch for this rather than develop so I'm not
> sure if Tom's fix is in develop but not 2.1 but I think the code is the same
> at the moment.
>
> In short although I've been using a 62MB file from a project a user was
> unable to export, previously there were loads of these zipper files in loads
> of different user/project folders and there are probably quite a few again
> now prior to my changing the memory limit.
>
> I could try resetting the memory_limit back on my own server and test again
> but my previous tests with the 62MB file certainly weren't crashing the
> server. Let me know if you think its worth testing with larger files and the
> default 128MB limit or if your and David's later responses have ruled that
> out as being worthwhile?
>
> I know Tom said last night that he was going to revisit his fix for this so
> would be good to hear back from Tom once he's done that.
>
> Cheers
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
> Sent: 01 May 2014 09:57
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi Ron,
>
> Have been thinking about your issue and just wondering whether you might
> have inadvertently chosen the worst file size possible for the video with
> 128Mb memory available to PHP. Let me explain (and this is entirely
> hypothetical)...
>
> The zipper creates a blank object (<0.1Mb memory used) The library loads all
> the other files individually and zips them into the object (0Mb ->
> compressed size of all js/html is perhaps around 1-2Mb or under) The library
> then loads the mp4 (now it has the object = 2Mb and the video = 62Mb in
> memory) It adds it to the zip object but it's also still in memory (mp4
> doesn't compress at all so we have 2 + 62 + 62 =126Mb which is less than
> 128Mb) It then closes the connection to the mp4 and destroys the object
> (memory now = 64Mb for the zip)
>
> I'm wondering whether a 64 Mb or slightly bigger will cause you to be able
> to crash that server too. Is that something you could try?
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
> Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
> School Learning Technologists will pick up the job. This will ensure that
> all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 09:09
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> I was just going to test your export code on a server where I don't have
> access to change php settings so the memory limit is fixed at 128MB. I know
> what David means that ideally we want cross platform code but as you say
> there are inevitably different solutions/restrictions for some e.g. no
> control over memory limit. In fact the upload limit on this particular
> shared server is set at the default 2MB so I had to upload the test 62MB mp4
> via ftp. However I'm now confused further because this is the same XOT code
> as the other servers I was using for testing so I expected the export to
> fail when testing before replacing the export code with yours and yet it
> didn't fail and I was able to download the zip which didn't contain any
> example_zipper files.
>
> I'm glad you're able to reproduce the issue but not sure how a server with
> 128MB memory limit is able to export ok without a change of code or memory
> limit :-(
>
>
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
> Sent: 01 May 2014 08:38
> To: xerte-dev at lists.nottingham.ac.uk
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> All i know Dave is that, with the commented code uncommented, my system
> couldn't download a 56Mb an 79Mb zip but with this code 79Mb came down no
> problem.
>
> We may have to use what we have access to on different systems. There may be
> a similar way to pipe the result into windows or we fall back to the old
> code and crank the memory requirements up yo 300Mb which not everyone can
> most likely do anyway...
>
> Regards
>
> John Smith
> Learning Technologist
> School of Health and Life Sciences
>
> Sent from Samsung Galaxy SII
>
>
>
> David Goodwin <david at palepurple.co.uk> wrote:
>
>
>
> <snip>
> Uses the unix command zip and pipes the stream back into PHP in 8192 chunks
>
> <snip>
> Also, might not work on Windows...
>
> So therefore it's not worth using?
>
>
> Technically there is no benefit to doing the fread(....) over a
> fpassthru(...) fpassthru does something like 8192 chunks internally.
>
>
> David.
>
> --
> --
> Pale Purple Ltd
> http://www.palepurple.co.uk @PalePurpleLtd
> 0845 0046746 / 07792 380669
> "Bespoke web and mobile application development"
>
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>

--
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196


_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From ronm at mitchellmedia.co.uk  Thu May  1 15:35:14 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Thu, 1 May 2014 15:35:14 +0100
Subject: [Xerte-dev] bootstrap logo code
Message-ID: <00f501cf654a$9cff3530$d6fd9f90$@co.uk>

Hi all 

I added this as an issue/bug/feature on github but might be quicker to
discuss here.

 

In the bootstrap play and preview files in current 2.1 and current develop
we have code to load a logo but no logo in the location it looks so shows a
broken image holder and arguably wasted vertical space. This applies to
existing LO's as well as new LO's.

 

I think ideally this ought to be an optional property together with options
for logo upload and placement. Until we have time to add that should we
remove these divs from \modules\site\play.php and \modules\site\preview.php?

 

Cheers

Ron

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140501/179b7eca/attachment.html>

From ronm at mitchellmedia.co.uk  Fri May  2 11:59:14 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Fri, 2 May 2014 11:59:14 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<53625027.2030801@tor.nl>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <008d01cf65f5$91db1240$b59136c0$@co.uk>

Hi all
My inclinations is to scream aaaarrrrrgh! Having had what seems like
coincidental support requests but is clearly all part of the same issue so
I've lost track of how long this issue has existed and how widespread it
might be. To explain...

Late yesterday I received a request from one of the HEA projects regarding
two of their students being unable to export their projects. After
investigation it turns out both students have been uploading quite a few
videos as part of their LO's in this case .flv's and so sure enough the
projects weren't exporting and creatining more and more example_zipper files
after each attempt.

One of these particular projects showed 158MB of media via media & quota and
the other 242MB
One had 47 zipper files the other 21 and across all the user folders 200+
zipper files!

I'm a bit concerned how widespread this could be with people unaware of the
issue. I know it's not best practice to upload loads of video direct but
lets be realistic users are going to do this and sometimes with good reason.

John - This server was also set to 128MB memory limit by defauls so I tried
your export.php but that was generating an empty zip download.

I've increased the memory limit to 300MB and with the current 2.1 code and
original export.php the above two projects export ok without leaving
example_zipper files behind. 

However I also noticed looking at the folder via ftp as you suggested John,
that if the user cancels the export, even with the memory limit increased,
the example zipper doesn't then get removed. I don't know if this would be
different with develop or if your export.php would only work with develop
John? 

But I am wondering if this is a bit of a ticking timebomb for lots of
installations! If anyone has SSH access to their installation(s) try find .
-type f -name 'example_zipper*'

Perhaps the temp file name should have been example_zippo! ;-)

We obviously need to crack this one asap...

Ron 

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 14:51
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi Tom,

Yes, definitely, no worries. I'm about to investigate a windows install with
both the original code and this new code to see whether I can get it to run.
Will update you.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk
[xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders
[reijnders at tor.nl]
Sent: 01 May 2014 14:46
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

I will look into this, but I don't have much time at the moment. Perhaps
topnight. Otherwise, it'll have to be next week.

Tom

Smith, John schreef op 1-5-2014 15:38:
> Well. Preliminary testing looks good with my new code but i haven't tested
on Windows yet - would be interested in getting Tom's take on all of this...
>
> I just added loads of videos via ftp (upload limit set to 7M on this 
> server) and checked the memory limit and it is
>
> memory_limit    90M     90M
>
> and with each video being used on a page then i've just successfully
downloaded a 316Mb archive zip without any problems (it took several minutes
too so the script didn't timeout) and was able to play the videos contained
within... and this is just a cheap Dreamhost shared server, nothing fancy...
>
> On to Windoze...
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | 
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk 
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell 
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 10:58
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> the problem is when I originally raised this issue on 8th April it 
> coincided with discovering loads of these un-removed example_zipper 
> files on the sandpit install. At that time I also mentioned the 
> separate session issue as a reason I hadn't updated to a newer 
> revision and I think you mentioned that you thought Tom had fixed this 
> zipper issue in a later revision. So once we'd identified the cause of 
> the session issue (or at least Julian's attempt at fixing his FF issue) I
updated the sandpit code as well as other servers.
> Now I've been using the 2.1 branch for this rather than develop so I'm 
> not sure if Tom's fix is in develop but not 2.1 but I think the code 
> is the same at the moment.
>
> In short although I've been using a 62MB file from a project a user 
> was unable to export, previously there were loads of these zipper 
> files in loads of different user/project folders and there are 
> probably quite a few again now prior to my changing the memory limit.
>
> I could try resetting the memory_limit back on my own server and test 
> again but my previous tests with the 62MB file certainly weren't 
> crashing the server. Let me know if you think its worth testing with 
> larger files and the default 128MB limit or if your and David's later 
> responses have ruled that out as being worthwhile?
>
> I know Tom said last night that he was going to revisit his fix for 
> this so would be good to hear back from Tom once he's done that.
>
> Cheers
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, 
> John
> Sent: 01 May 2014 09:57
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi Ron,
>
> Have been thinking about your issue and just wondering whether you 
> might have inadvertently chosen the worst file size possible for the 
> video with 128Mb memory available to PHP. Let me explain (and this is 
> entirely hypothetical)...
>
> The zipper creates a blank object (<0.1Mb memory used) The library 
> loads all the other files individually and zips them into the object 
> (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or 
> under) The library then loads the mp4 (now it has the object = 2Mb and 
> the video = 62Mb in
> memory) It adds it to the zip object but it's also still in memory 
> (mp4 doesn't compress at all so we have 2 + 62 + 62 =126Mb which is 
> less than
> 128Mb) It then closes the connection to the mp4 and destroys the 
> object (memory now = 64Mb for the zip)
>
> I'm wondering whether a 64 Mb or slightly bigger will cause you to be 
> able to crash that server too. Is that something you could try?
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | 
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of 
> the School Learning Technologists will pick up the job. This will 
> ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell 
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 09:09
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> I was just going to test your export code on a server where I don't 
> have access to change php settings so the memory limit is fixed at 
> 128MB. I know what David means that ideally we want cross platform 
> code but as you say there are inevitably different 
> solutions/restrictions for some e.g. no control over memory limit. In 
> fact the upload limit on this particular shared server is set at the 
> default 2MB so I had to upload the test 62MB mp4 via ftp. However I'm 
> now confused further because this is the same XOT code as the other 
> servers I was using for testing so I expected the export to fail when 
> testing before replacing the export code with yours and yet it didn't 
> fail and I was able to download the zip which didn't contain any
example_zipper files.
>
> I'm glad you're able to reproduce the issue but not sure how a server 
> with 128MB memory limit is able to export ok without a change of code 
> or memory limit :-(
>
>
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, 
> John
> Sent: 01 May 2014 08:38
> To: xerte-dev at lists.nottingham.ac.uk
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> All i know Dave is that, with the commented code uncommented, my 
> system couldn't download a 56Mb an 79Mb zip but with this code 79Mb 
> came down no problem.
>
> We may have to use what we have access to on different systems. There 
> may be a similar way to pipe the result into windows or we fall back 
> to the old code and crank the memory requirements up yo 300Mb which 
> not everyone can most likely do anyway...
>
> Regards
>
> John Smith
> Learning Technologist
> School of Health and Life Sciences
>
> Sent from Samsung Galaxy SII
>
>
>
> David Goodwin <david at palepurple.co.uk> wrote:
>
>
>
> <snip>
> Uses the unix command zip and pipes the stream back into PHP in 8192 
> chunks
>
> <snip>
> Also, might not work on Windows...
>
> So therefore it's not worth using?
>
>
> Technically there is no benefit to doing the fread(....) over a
> fpassthru(...) fpassthru does something like 8192 chunks internally.
>
>
> David.
>
> --
> --
> Pale Purple Ltd
> http://www.palepurple.co.uk @PalePurpleLtd
> 0845 0046746 / 07792 380669
> "Bespoke web and mobile application development"
>
>
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in 
> any attachment.  Any views or opinions expressed by the author of this 
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in 
> any attachment.  Any views or opinions expressed by the author of this 
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in 
> any attachment.  Any views or opinions expressed by the author of this 
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in 
> any attachment.  Any views or opinions expressed by the author of this 
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email
communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number 
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email
communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>

--
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196


_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.






From david at palepurple.co.uk  Fri May  2 12:01:34 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Fri, 02 May 2014 12:01:34 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <008d01cf65f5$91db1240$b59136c0$@co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<53625027.2030801@tor.nl>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<008d01cf65f5$91db1240$b59136c0$@co.uk>
Message-ID: <53637B0E.5090409@palepurple.co.uk>


>
> But I am wondering if this is a bit of a ticking timebomb for lots of
> installations! If anyone has SSH access to their installation(s) try find .
> -type f -name 'example_zipper*'
>


I appear to have a few lying around on one customer's site ..... [so 
it's not just you]

David.


-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From J.J.Smith at gcu.ac.uk  Fri May  2 12:47:51 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Fri, 2 May 2014 12:47:51 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <008d01cf65f5$91db1240$b59136c0$@co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<53625027.2030801@tor.nl>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<008d01cf65f5$91db1240$b59136c0$@co.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101CE@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Hi Ron,

Yes I've got a few lying around also so it is a bit of an issue...

I've tried that code on 4 servers and 2 worked and two didn't. I haven't checked yet but it's probably either safe_mode being on or zip not being available (both of which we could check for). I've also see a few code samples using 7z, tar etc so it is possible that we could check for these eventualities and use the most appropriate technique...

That being said, i've done some testing and can't seem to get executable detection to work consistently on Windows servers which is where command line zip is most likely to not be available...

Also, i've not tested it but if the user cancelling the zip stops the rest of the php code running then that could be very problematic. We would have to delete files at the start of the code although we may have to change the format of the temporary file. The number at the end is obviously to allow two or more different people to export at the same time but these then get orphaned if the process quits. What we could do is have two numbers with the first calculable, zipper_xx_yyy and then we can delete all the zipper_xx_* except for the xx in use today... or we delete all zipper folders that were created >x hours ago.

I think we first need to get some export routines that work on different systems though as we can't expect everyone can crank their memory limit up to 300Mb...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell [ronm at mitchellmedia.co.uk]
Sent: 02 May 2014 11:59
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi all
My inclinations is to scream aaaarrrrrgh! Having had what seems like
coincidental support requests but is clearly all part of the same issue so
I've lost track of how long this issue has existed and how widespread it
might be. To explain...

Late yesterday I received a request from one of the HEA projects regarding
two of their students being unable to export their projects. After
investigation it turns out both students have been uploading quite a few
videos as part of their LO's in this case .flv's and so sure enough the
projects weren't exporting and creatining more and more example_zipper files
after each attempt.

One of these particular projects showed 158MB of media via media & quota and
the other 242MB
One had 47 zipper files the other 21 and across all the user folders 200+
zipper files!

I'm a bit concerned how widespread this could be with people unaware of the
issue. I know it's not best practice to upload loads of video direct but
lets be realistic users are going to do this and sometimes with good reason.

John - This server was also set to 128MB memory limit by defauls so I tried
your export.php but that was generating an empty zip download.

I've increased the memory limit to 300MB and with the current 2.1 code and
original export.php the above two projects export ok without leaving
example_zipper files behind.

However I also noticed looking at the folder via ftp as you suggested John,
that if the user cancels the export, even with the memory limit increased,
the example zipper doesn't then get removed. I don't know if this would be
different with develop or if your export.php would only work with develop
John?

But I am wondering if this is a bit of a ticking timebomb for lots of
installations! If anyone has SSH access to their installation(s) try find .
-type f -name 'example_zipper*'

Perhaps the temp file name should have been example_zippo! ;-)

We obviously need to crack this one asap...

Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 14:51
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi Tom,

Yes, definitely, no worries. I'm about to investigate a windows install with
both the original code and this new code to see whether I can get it to run.
Will update you.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk
[xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders
[reijnders at tor.nl]
Sent: 01 May 2014 14:46
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

I will look into this, but I don't have much time at the moment. Perhaps
topnight. Otherwise, it'll have to be next week.

Tom

Smith, John schreef op 1-5-2014 15:38:
> Well. Preliminary testing looks good with my new code but i haven't tested
on Windows yet - would be interested in getting Tom's take on all of this...
>
> I just added loads of videos via ftp (upload limit set to 7M on this
> server) and checked the memory limit and it is
>
> memory_limit    90M     90M
>
> and with each video being used on a page then i've just successfully
downloaded a 316Mb archive zip without any problems (it took several minutes
too so the script didn't timeout) and was able to play the videos contained
within... and this is just a cheap Dreamhost shared server, nothing fancy...
>
> On to Windoze...
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 10:58
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> the problem is when I originally raised this issue on 8th April it
> coincided with discovering loads of these un-removed example_zipper
> files on the sandpit install. At that time I also mentioned the
> separate session issue as a reason I hadn't updated to a newer
> revision and I think you mentioned that you thought Tom had fixed this
> zipper issue in a later revision. So once we'd identified the cause of
> the session issue (or at least Julian's attempt at fixing his FF issue) I
updated the sandpit code as well as other servers.
> Now I've been using the 2.1 branch for this rather than develop so I'm
> not sure if Tom's fix is in develop but not 2.1 but I think the code
> is the same at the moment.
>
> In short although I've been using a 62MB file from a project a user
> was unable to export, previously there were loads of these zipper
> files in loads of different user/project folders and there are
> probably quite a few again now prior to my changing the memory limit.
>
> I could try resetting the memory_limit back on my own server and test
> again but my previous tests with the 62MB file certainly weren't
> crashing the server. Let me know if you think its worth testing with
> larger files and the default 128MB limit or if your and David's later
> responses have ruled that out as being worthwhile?
>
> I know Tom said last night that he was going to revisit his fix for
> this so would be good to hear back from Tom once he's done that.
>
> Cheers
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
> John
> Sent: 01 May 2014 09:57
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi Ron,
>
> Have been thinking about your issue and just wondering whether you
> might have inadvertently chosen the worst file size possible for the
> video with 128Mb memory available to PHP. Let me explain (and this is
> entirely hypothetical)...
>
> The zipper creates a blank object (<0.1Mb memory used) The library
> loads all the other files individually and zips them into the object
> (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or
> under) The library then loads the mp4 (now it has the object = 2Mb and
> the video = 62Mb in
> memory) It adds it to the zip object but it's also still in memory
> (mp4 doesn't compress at all so we have 2 + 62 + 62 =126Mb which is
> less than
> 128Mb) It then closes the connection to the mp4 and destroys the
> object (memory now = 64Mb for the zip)
>
> I'm wondering whether a 64 Mb or slightly bigger will cause you to be
> able to crash that server too. Is that something you could try?
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of
> the School Learning Technologists will pick up the job. This will
> ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 09:09
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> I was just going to test your export code on a server where I don't
> have access to change php settings so the memory limit is fixed at
> 128MB. I know what David means that ideally we want cross platform
> code but as you say there are inevitably different
> solutions/restrictions for some e.g. no control over memory limit. In
> fact the upload limit on this particular shared server is set at the
> default 2MB so I had to upload the test 62MB mp4 via ftp. However I'm
> now confused further because this is the same XOT code as the other
> servers I was using for testing so I expected the export to fail when
> testing before replacing the export code with yours and yet it didn't
> fail and I was able to download the zip which didn't contain any
example_zipper files.
>
> I'm glad you're able to reproduce the issue but not sure how a server
> with 128MB memory limit is able to export ok without a change of code
> or memory limit :-(
>
>
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
> John
> Sent: 01 May 2014 08:38
> To: xerte-dev at lists.nottingham.ac.uk
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> All i know Dave is that, with the commented code uncommented, my
> system couldn't download a 56Mb an 79Mb zip but with this code 79Mb
> came down no problem.
>
> We may have to use what we have access to on different systems. There
> may be a similar way to pipe the result into windows or we fall back
> to the old code and crank the memory requirements up yo 300Mb which
> not everyone can most likely do anyway...
>
> Regards
>
> John Smith
> Learning Technologist
> School of Health and Life Sciences
>
> Sent from Samsung Galaxy SII
>
>
>
> David Goodwin <david at palepurple.co.uk> wrote:
>
>
>
> <snip>
> Uses the unix command zip and pipes the stream back into PHP in 8192
> chunks
>
> <snip>
> Also, might not work on Windows...
>
> So therefore it's not worth using?
>
>
> Technically there is no benefit to doing the fread(....) over a
> fpassthru(...) fpassthru does something like 8192 chunks internally.
>
>
> David.
>
> --
> --
> Pale Purple Ltd
> http://www.palepurple.co.uk @PalePurpleLtd
> 0845 0046746 / 07792 380669
> "Bespoke web and mobile application development"
>
>
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email
communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email
communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>

--
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196


_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474


From david at palepurple.co.uk  Fri May  2 12:58:03 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Fri, 02 May 2014 12:58:03 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <008d01cf65f5$91db1240$b59136c0$@co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<53625027.2030801@tor.nl>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<008d01cf65f5$91db1240$b59136c0$@co.uk>
Message-ID: <5363884B.20903@palepurple.co.uk>

I'm working on a fix for the example_zipper thing - the code can use 
php's tempnam() function instead - which should result in the file being 
cleaned up by PHP afterwards.... or at least not existing in the current 
working directory (i.e. putting it in /tmp or equivalent instead).

David.


On 02/05/14 11:59, Ron Mitchell wrote:
> Hi all
> My inclinations is to scream aaaarrrrrgh! Having had what seems like
> coincidental support requests but is clearly all part of the same issue so
> I've lost track of how long this issue has existed and how widespread it
> might be. To explain...
>
> Late yesterday I received a request from one of the HEA projects regarding
> two of their students being unable to export their projects. After
> investigation it turns out both students have been uploading quite a few
> videos as part of their LO's in this case .flv's and so sure enough the
> projects weren't exporting and creatining more and more example_zipper files
> after each attempt.
>
> One of these particular projects showed 158MB of media via media & quota and
> the other 242MB
> One had 47 zipper files the other 21 and across all the user folders 200+
> zipper files!
>
> I'm a bit concerned how widespread this could be with people unaware of the
> issue. I know it's not best practice to upload loads of video direct but
> lets be realistic users are going to do this and sometimes with good reason.
>
> John - This server was also set to 128MB memory limit by defauls so I tried
> your export.php but that was generating an empty zip download.
>
> I've increased the memory limit to 300MB and with the current 2.1 code and
> original export.php the above two projects export ok without leaving
> example_zipper files behind.
>
> However I also noticed looking at the folder via ftp as you suggested John,
> that if the user cancels the export, even with the memory limit increased,
> the example zipper doesn't then get removed. I don't know if this would be
> different with develop or if your export.php would only work with develop
> John?
>
> But I am wondering if this is a bit of a ticking timebomb for lots of
> installations! If anyone has SSH access to their installation(s) try find .
> -type f -name 'example_zipper*'
>
> Perhaps the temp file name should have been example_zippo! ;-)
>
> We obviously need to crack this one asap...
>
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
> Sent: 01 May 2014 14:51
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi Tom,
>
> Yes, definitely, no worries. I'm about to investigate a windows install with
> both the original code and this new code to see whether I can get it to run.
> Will update you.
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
> Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
> School Learning Technologists will pick up the job. This will ensure that
> all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders
> [reijnders at tor.nl]
> Sent: 01 May 2014 14:46
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> I will look into this, but I don't have much time at the moment. Perhaps
> topnight. Otherwise, it'll have to be next week.
>
> Tom
>
> Smith, John schreef op 1-5-2014 15:38:
>> Well. Preliminary testing looks good with my new code but i haven't tested
> on Windows yet - would be interested in getting Tom's take on all of this...
>>
>> I just added loads of videos via ftp (upload limit set to 7M on this
>> server) and checked the memory limit and it is
>>
>> memory_limit    90M     90M
>>
>> and with each video being used on a page then i've just successfully
> downloaded a 316Mb archive zip without any problems (it took several minutes
> too so the script didn't timeout) and was able to play the videos contained
> within... and this is just a cheap Dreamhost shared server, nothing fancy...
>>
>> On to Windoze...
>>
>> Regards,
>>
>> John Smith | Learning Technologist
>> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
>> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>>
>> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
> School Learning Technologists will pick up the job. This will ensure that
> all jobs are completed as promptly as possible.
>> ________________________________________
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
>> [ronm at mitchellmedia.co.uk]
>> Sent: 01 May 2014 10:58
>> To: 'For Xerte technical developers'
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> Hi John
>> the problem is when I originally raised this issue on 8th April it
>> coincided with discovering loads of these un-removed example_zipper
>> files on the sandpit install. At that time I also mentioned the
>> separate session issue as a reason I hadn't updated to a newer
>> revision and I think you mentioned that you thought Tom had fixed this
>> zipper issue in a later revision. So once we'd identified the cause of
>> the session issue (or at least Julian's attempt at fixing his FF issue) I
> updated the sandpit code as well as other servers.
>> Now I've been using the 2.1 branch for this rather than develop so I'm
>> not sure if Tom's fix is in develop but not 2.1 but I think the code
>> is the same at the moment.
>>
>> In short although I've been using a 62MB file from a project a user
>> was unable to export, previously there were loads of these zipper
>> files in loads of different user/project folders and there are
>> probably quite a few again now prior to my changing the memory limit.
>>
>> I could try resetting the memory_limit back on my own server and test
>> again but my previous tests with the 62MB file certainly weren't
>> crashing the server. Let me know if you think its worth testing with
>> larger files and the default 128MB limit or if your and David's later
>> responses have ruled that out as being worthwhile?
>>
>> I know Tom said last night that he was going to revisit his fix for
>> this so would be good to hear back from Tom once he's done that.
>>
>> Cheers
>> Ron
>>
>> -----Original Message-----
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
>> John
>> Sent: 01 May 2014 09:57
>> To: For Xerte technical developers
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> Hi Ron,
>>
>> Have been thinking about your issue and just wondering whether you
>> might have inadvertently chosen the worst file size possible for the
>> video with 128Mb memory available to PHP. Let me explain (and this is
>> entirely hypothetical)...
>>
>> The zipper creates a blank object (<0.1Mb memory used) The library
>> loads all the other files individually and zips them into the object
>> (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or
>> under) The library then loads the mp4 (now it has the object = 2Mb and
>> the video = 62Mb in
>> memory) It adds it to the zip object but it's also still in memory
>> (mp4 doesn't compress at all so we have 2 + 62 + 62 =126Mb which is
>> less than
>> 128Mb) It then closes the connection to the mp4 and destroys the
>> object (memory now = 64Mb for the zip)
>>
>> I'm wondering whether a 64 Mb or slightly bigger will cause you to be
>> able to crash that server too. Is that something you could try?
>>
>> Regards,
>>
>> John Smith | Learning Technologist
>> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
>> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>>
>> Please address ALL support requests to hlsblt at gcu.ac.uk where one of
>> the School Learning Technologists will pick up the job. This will
>> ensure that all jobs are completed as promptly as possible.
>> ________________________________________
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
>> [ronm at mitchellmedia.co.uk]
>> Sent: 01 May 2014 09:09
>> To: 'For Xerte technical developers'
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> Hi John
>> I was just going to test your export code on a server where I don't
>> have access to change php settings so the memory limit is fixed at
>> 128MB. I know what David means that ideally we want cross platform
>> code but as you say there are inevitably different
>> solutions/restrictions for some e.g. no control over memory limit. In
>> fact the upload limit on this particular shared server is set at the
>> default 2MB so I had to upload the test 62MB mp4 via ftp. However I'm
>> now confused further because this is the same XOT code as the other
>> servers I was using for testing so I expected the export to fail when
>> testing before replacing the export code with yours and yet it didn't
>> fail and I was able to download the zip which didn't contain any
> example_zipper files.
>>
>> I'm glad you're able to reproduce the issue but not sure how a server
>> with 128MB memory limit is able to export ok without a change of code
>> or memory limit :-(
>>
>>
>>
>> -----Original Message-----
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
>> John
>> Sent: 01 May 2014 08:38
>> To: xerte-dev at lists.nottingham.ac.uk
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> All i know Dave is that, with the commented code uncommented, my
>> system couldn't download a 56Mb an 79Mb zip but with this code 79Mb
>> came down no problem.
>>
>> We may have to use what we have access to on different systems. There
>> may be a similar way to pipe the result into windows or we fall back
>> to the old code and crank the memory requirements up yo 300Mb which
>> not everyone can most likely do anyway...
>>
>> Regards
>>
>> John Smith
>> Learning Technologist
>> School of Health and Life Sciences
>>
>> Sent from Samsung Galaxy SII
>>
>>
>>
>> David Goodwin <david at palepurple.co.uk> wrote:
>>
>>
>>
>> <snip>
>> Uses the unix command zip and pipes the stream back into PHP in 8192
>> chunks
>>
>> <snip>
>> Also, might not work on Windows...
>>
>> So therefore it's not worth using?
>>
>>
>> Technically there is no benefit to doing the fread(....) over a
>> fpassthru(...) fpassthru does something like 8192 chunks internally.
>>
>>
>> David.
>>
>> --
>> --
>> Pale Purple Ltd
>> http://www.palepurple.co.uk @PalePurpleLtd
>> 0845 0046746 / 07792 380669
>> "Bespoke web and mobile application development"
>>
>>
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
> message in
>> error, please send it back to me, and immediately delete it.   Please do
> not
>> use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of
> Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>> Glasgow Caledonian University is a registered Scottish charity, number
>> SC021474
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
> message in
>> error, please send it back to me, and immediately delete it.   Please do
> not
>> use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of
> Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>>
>>
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
> message in
>> error, please send it back to me, and immediately delete it.   Please do
> not
>> use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of
> Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>>
>>
>>
>> Glasgow Caledonian University is a registered Scottish charity, number
>> SC021474
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
> message in
>> error, please send it back to me, and immediately delete it.   Please do
> not
>> use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of
> Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>>
>>
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>>
>>
>> Glasgow Caledonian University is a registered Scottish charity, number
>> SC021474
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>>
>>
>
> --
> --
>
> Tom Reijnders
> TOR Informatica
> Chopinlaan 27
> 5242HM Rosmalen
> Tel: 073 5226191
> Fax: 073 5226196
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any
> attachment.  Any views or opinions expressed by the author of this email do
> not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system,
> you are advised to perform your own checks. Email communications with the
> University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From Julian.Tenney at nottingham.ac.uk  Fri May  2 14:08:12 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Fri, 2 May 2014 14:08:12 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <008d01cf65f5$91db1240$b59136c0$@co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<53625027.2030801@tor.nl>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<008d01cf65f5$91db1240$b59136c0$@co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0CECE82BC0@EXCHANGE1.ad.nottingham.ac.uk>

I haven't followed all the details here, but can you summarise in words I'd understand?

thanks

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 02 May 2014 11:59
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi all
My inclinations is to scream aaaarrrrrgh! Having had what seems like coincidental support requests but is clearly all part of the same issue so I've lost track of how long this issue has existed and how widespread it might be. To explain...

Late yesterday I received a request from one of the HEA projects regarding two of their students being unable to export their projects. After investigation it turns out both students have been uploading quite a few videos as part of their LO's in this case .flv's and so sure enough the projects weren't exporting and creatining more and more example_zipper files after each attempt.

One of these particular projects showed 158MB of media via media & quota and the other 242MB One had 47 zipper files the other 21 and across all the user folders 200+ zipper files!

I'm a bit concerned how widespread this could be with people unaware of the issue. I know it's not best practice to upload loads of video direct but lets be realistic users are going to do this and sometimes with good reason.

John - This server was also set to 128MB memory limit by defauls so I tried your export.php but that was generating an empty zip download.

I've increased the memory limit to 300MB and with the current 2.1 code and original export.php the above two projects export ok without leaving example_zipper files behind.

However I also noticed looking at the folder via ftp as you suggested John, that if the user cancels the export, even with the memory limit increased, the example zipper doesn't then get removed. I don't know if this would be different with develop or if your export.php would only work with develop John?

But I am wondering if this is a bit of a ticking timebomb for lots of installations! If anyone has SSH access to their installation(s) try find .
-type f -name 'example_zipper*'

Perhaps the temp file name should have been example_zippo! ;-)

We obviously need to crack this one asap...

Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 14:51
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi Tom,

Yes, definitely, no worries. I'm about to investigate a windows install with both the original code and this new code to see whether I can get it to run.
Will update you.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk
[xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders [reijnders at tor.nl]
Sent: 01 May 2014 14:46
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

I will look into this, but I don't have much time at the moment. Perhaps topnight. Otherwise, it'll have to be next week.

Tom

Smith, John schreef op 1-5-2014 15:38:
> Well. Preliminary testing looks good with my new code but i haven't
> tested
on Windows yet - would be interested in getting Tom's take on all of this...
>
> I just added loads of videos via ftp (upload limit set to 7M on this
> server) and checked the memory limit and it is
>
> memory_limit    90M     90M
>
> and with each video being used on a page then i've just successfully
downloaded a 316Mb archive zip without any problems (it took several minutes too so the script didn't timeout) and was able to play the videos contained within... and this is just a cheap Dreamhost shared server, nothing fancy...
>
> On to Windoze...
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of
> the
School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 10:58
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> the problem is when I originally raised this issue on 8th April it
> coincided with discovering loads of these un-removed example_zipper
> files on the sandpit install. At that time I also mentioned the
> separate session issue as a reason I hadn't updated to a newer
> revision and I think you mentioned that you thought Tom had fixed this
> zipper issue in a later revision. So once we'd identified the cause of
> the session issue (or at least Julian's attempt at fixing his FF
> issue) I
updated the sandpit code as well as other servers.
> Now I've been using the 2.1 branch for this rather than develop so I'm
> not sure if Tom's fix is in develop but not 2.1 but I think the code
> is the same at the moment.
>
> In short although I've been using a 62MB file from a project a user
> was unable to export, previously there were loads of these zipper
> files in loads of different user/project folders and there are
> probably quite a few again now prior to my changing the memory limit.
>
> I could try resetting the memory_limit back on my own server and test
> again but my previous tests with the 62MB file certainly weren't
> crashing the server. Let me know if you think its worth testing with
> larger files and the default 128MB limit or if your and David's later
> responses have ruled that out as being worthwhile?
>
> I know Tom said last night that he was going to revisit his fix for
> this so would be good to hear back from Tom once he's done that.
>
> Cheers
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
> John
> Sent: 01 May 2014 09:57
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi Ron,
>
> Have been thinking about your issue and just wondering whether you
> might have inadvertently chosen the worst file size possible for the
> video with 128Mb memory available to PHP. Let me explain (and this is
> entirely hypothetical)...
>
> The zipper creates a blank object (<0.1Mb memory used) The library
> loads all the other files individually and zips them into the object
> (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or
> under) The library then loads the mp4 (now it has the object = 2Mb and
> the video = 62Mb in
> memory) It adds it to the zip object but it's also still in memory
> (mp4 doesn't compress at all so we have 2 + 62 + 62 =126Mb which is
> less than
> 128Mb) It then closes the connection to the mp4 and destroys the
> object (memory now = 64Mb for the zip)
>
> I'm wondering whether a 64 Mb or slightly bigger will cause you to be
> able to crash that server too. Is that something you could try?
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of
> the School Learning Technologists will pick up the job. This will
> ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 09:09
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> I was just going to test your export code on a server where I don't
> have access to change php settings so the memory limit is fixed at
> 128MB. I know what David means that ideally we want cross platform
> code but as you say there are inevitably different
> solutions/restrictions for some e.g. no control over memory limit. In
> fact the upload limit on this particular shared server is set at the
> default 2MB so I had to upload the test 62MB mp4 via ftp. However I'm
> now confused further because this is the same XOT code as the other
> servers I was using for testing so I expected the export to fail when
> testing before replacing the export code with yours and yet it didn't
> fail and I was able to download the zip which didn't contain any
example_zipper files.
>
> I'm glad you're able to reproduce the issue but not sure how a server
> with 128MB memory limit is able to export ok without a change of code
> or memory limit :-(
>
>
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
> John
> Sent: 01 May 2014 08:38
> To: xerte-dev at lists.nottingham.ac.uk
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> All i know Dave is that, with the commented code uncommented, my
> system couldn't download a 56Mb an 79Mb zip but with this code 79Mb
> came down no problem.
>
> We may have to use what we have access to on different systems. There
> may be a similar way to pipe the result into windows or we fall back
> to the old code and crank the memory requirements up yo 300Mb which
> not everyone can most likely do anyway...
>
> Regards
>
> John Smith
> Learning Technologist
> School of Health and Life Sciences
>
> Sent from Samsung Galaxy SII
>
>
>
> David Goodwin <david at palepurple.co.uk> wrote:
>
>
>
> <snip>
> Uses the unix command zip and pipes the stream back into PHP in 8192
> chunks
>
> <snip>
> Also, might not work on Windows...
>
> So therefore it's not worth using?
>
>
> Technically there is no benefit to doing the fread(....) over a
> fpassthru(...) fpassthru does something like 8192 chunks internally.
>
>
> David.
>
> --
> --
> Pale Purple Ltd
> http://www.palepurple.co.uk @PalePurpleLtd
> 0845 0046746 / 07792 380669
> "Bespoke web and mobile application development"
>
>
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>

--
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196


_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From ronm at mitchellmedia.co.uk  Fri May  2 14:55:07 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Fri, 2 May 2014 14:55:07 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0CECE82BC0@EXCHANGE1.ad.nottingham.ac.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<53625027.2030801@tor.nl>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>	<008d01cf65f5$91db1240$b59136c0$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CECE82BC0@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <00cb01cf660e$2369df30$6a3d9d90$@co.uk>

For projects containing lots or large files when the user tries to export it
fails and leaves behind files named example_zipper_ and doesn't clear them
up so the situation gets worse each time. It only occurs if the size of the
project and temporary zip exceeds the memory_limit set on the server. Look
in USER-FILES for example_zipper files?

HTH
Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 02 May 2014 14:08
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

I haven't followed all the details here, but can you summarise in words I'd
understand?

thanks

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 02 May 2014 11:59
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi all
My inclinations is to scream aaaarrrrrgh! Having had what seems like
coincidental support requests but is clearly all part of the same issue so
I've lost track of how long this issue has existed and how widespread it
might be. To explain...

Late yesterday I received a request from one of the HEA projects regarding
two of their students being unable to export their projects. After
investigation it turns out both students have been uploading quite a few
videos as part of their LO's in this case .flv's and so sure enough the
projects weren't exporting and creatining more and more example_zipper files
after each attempt.

One of these particular projects showed 158MB of media via media & quota and
the other 242MB One had 47 zipper files the other 21 and across all the user
folders 200+ zipper files!

I'm a bit concerned how widespread this could be with people unaware of the
issue. I know it's not best practice to upload loads of video direct but
lets be realistic users are going to do this and sometimes with good reason.

John - This server was also set to 128MB memory limit by defauls so I tried
your export.php but that was generating an empty zip download.

I've increased the memory limit to 300MB and with the current 2.1 code and
original export.php the above two projects export ok without leaving
example_zipper files behind.

However I also noticed looking at the folder via ftp as you suggested John,
that if the user cancels the export, even with the memory limit increased,
the example zipper doesn't then get removed. I don't know if this would be
different with develop or if your export.php would only work with develop
John?

But I am wondering if this is a bit of a ticking timebomb for lots of
installations! If anyone has SSH access to their installation(s) try find .
-type f -name 'example_zipper*'

Perhaps the temp file name should have been example_zippo! ;-)

We obviously need to crack this one asap...

Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 14:51
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi Tom,

Yes, definitely, no worries. I'm about to investigate a windows install with
both the original code and this new code to see whether I can get it to run.
Will update you.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow
Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk
[xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders
[reijnders at tor.nl]
Sent: 01 May 2014 14:46
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

I will look into this, but I don't have much time at the moment. Perhaps
topnight. Otherwise, it'll have to be next week.

Tom

Smith, John schreef op 1-5-2014 15:38:
> Well. Preliminary testing looks good with my new code but i haven't 
> tested
on Windows yet - would be interested in getting Tom's take on all of this...
>
> I just added loads of videos via ftp (upload limit set to 7M on this
> server) and checked the memory limit and it is
>
> memory_limit    90M     90M
>
> and with each video being used on a page then i've just successfully
downloaded a 316Mb archive zip without any problems (it took several minutes
too so the script didn't timeout) and was able to play the videos contained
within... and this is just a cheap Dreamhost shared server, nothing fancy...
>
> On to Windoze...
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | 
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of 
> the
School Learning Technologists will pick up the job. This will ensure that
all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell 
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 10:58
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> the problem is when I originally raised this issue on 8th April it 
> coincided with discovering loads of these un-removed example_zipper 
> files on the sandpit install. At that time I also mentioned the 
> separate session issue as a reason I hadn't updated to a newer 
> revision and I think you mentioned that you thought Tom had fixed this 
> zipper issue in a later revision. So once we'd identified the cause of 
> the session issue (or at least Julian's attempt at fixing his FF
> issue) I
updated the sandpit code as well as other servers.
> Now I've been using the 2.1 branch for this rather than develop so I'm 
> not sure if Tom's fix is in develop but not 2.1 but I think the code 
> is the same at the moment.
>
> In short although I've been using a 62MB file from a project a user 
> was unable to export, previously there were loads of these zipper 
> files in loads of different user/project folders and there are 
> probably quite a few again now prior to my changing the memory limit.
>
> I could try resetting the memory_limit back on my own server and test 
> again but my previous tests with the 62MB file certainly weren't 
> crashing the server. Let me know if you think its worth testing with 
> larger files and the default 128MB limit or if your and David's later 
> responses have ruled that out as being worthwhile?
>
> I know Tom said last night that he was going to revisit his fix for 
> this so would be good to hear back from Tom once he's done that.
>
> Cheers
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, 
> John
> Sent: 01 May 2014 09:57
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi Ron,
>
> Have been thinking about your issue and just wondering whether you 
> might have inadvertently chosen the worst file size possible for the 
> video with 128Mb memory available to PHP. Let me explain (and this is 
> entirely hypothetical)...
>
> The zipper creates a blank object (<0.1Mb memory used) The library 
> loads all the other files individually and zips them into the object 
> (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or
> under) The library then loads the mp4 (now it has the object = 2Mb and 
> the video = 62Mb in
> memory) It adds it to the zip object but it's also still in memory
> (mp4 doesn't compress at all so we have 2 + 62 + 62 =126Mb which is 
> less than
> 128Mb) It then closes the connection to the mp4 and destroys the 
> object (memory now = 64Mb for the zip)
>
> I'm wondering whether a 64 Mb or slightly bigger will cause you to be 
> able to crash that server too. Is that something you could try?
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | 
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of 
> the School Learning Technologists will pick up the job. This will 
> ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell 
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 09:09
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> I was just going to test your export code on a server where I don't 
> have access to change php settings so the memory limit is fixed at 
> 128MB. I know what David means that ideally we want cross platform 
> code but as you say there are inevitably different 
> solutions/restrictions for some e.g. no control over memory limit. In 
> fact the upload limit on this particular shared server is set at the 
> default 2MB so I had to upload the test 62MB mp4 via ftp. However I'm 
> now confused further because this is the same XOT code as the other 
> servers I was using for testing so I expected the export to fail when 
> testing before replacing the export code with yours and yet it didn't 
> fail and I was able to download the zip which didn't contain any
example_zipper files.
>
> I'm glad you're able to reproduce the issue but not sure how a server 
> with 128MB memory limit is able to export ok without a change of code 
> or memory limit :-(
>
>
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, 
> John
> Sent: 01 May 2014 08:38
> To: xerte-dev at lists.nottingham.ac.uk
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> All i know Dave is that, with the commented code uncommented, my 
> system couldn't download a 56Mb an 79Mb zip but with this code 79Mb 
> came down no problem.
>
> We may have to use what we have access to on different systems. There 
> may be a similar way to pipe the result into windows or we fall back 
> to the old code and crank the memory requirements up yo 300Mb which 
> not everyone can most likely do anyway...
>
> Regards
>
> John Smith
> Learning Technologist
> School of Health and Life Sciences
>
> Sent from Samsung Galaxy SII
>
>
>
> David Goodwin <david at palepurple.co.uk> wrote:
>
>
>
> <snip>
> Uses the unix command zip and pipes the stream back into PHP in 8192 
> chunks
>
> <snip>
> Also, might not work on Windows...
>
> So therefore it's not worth using?
>
>
> Technically there is no benefit to doing the fread(....) over a
> fpassthru(...) fpassthru does something like 8192 chunks internally.
>
>
> David.
>
> --
> --
> Pale Purple Ltd
> http://www.palepurple.co.uk @PalePurpleLtd
> 0845 0046746 / 07792 380669
> "Bespoke web and mobile application development"
>
>
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in 
> any attachment.  Any views or opinions expressed by the author of this 
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in 
> any attachment.  Any views or opinions expressed by the author of this 
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in 
> any attachment.  Any views or opinions expressed by the author of this 
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in 
> any attachment.  Any views or opinions expressed by the author of this 
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee 
> and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email
communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee 
> and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email
communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>

--
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196


_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.






From Julian.Tenney at nottingham.ac.uk  Fri May  2 14:52:54 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Fri, 2 May 2014 14:52:54 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited
In-Reply-To: <00cb01cf660e$2369df30$6a3d9d90$@co.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<53625027.2030801@tor.nl>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<008d01cf65f5$91db1240$b59136c0$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CECE82BC0@EXCHANGE1.ad.nottingham.ac.uk>
	<00cb01cf660e$2369df30$6a3d9d90$@co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0CECE82C11@EXCHANGE1.ad.nottingham.ac.uk>

Got it. OK.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 02 May 2014 14:55
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

For projects containing lots or large files when the user tries to export it fails and leaves behind files named example_zipper_ and doesn't clear them up so the situation gets worse each time. It only occurs if the size of the project and temporary zip exceeds the memory_limit set on the server. Look in USER-FILES for example_zipper files?

HTH
Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 02 May 2014 14:08
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

I haven't followed all the details here, but can you summarise in words I'd understand?

thanks

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 02 May 2014 11:59
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi all
My inclinations is to scream aaaarrrrrgh! Having had what seems like coincidental support requests but is clearly all part of the same issue so I've lost track of how long this issue has existed and how widespread it might be. To explain...

Late yesterday I received a request from one of the HEA projects regarding two of their students being unable to export their projects. After investigation it turns out both students have been uploading quite a few videos as part of their LO's in this case .flv's and so sure enough the projects weren't exporting and creatining more and more example_zipper files after each attempt.

One of these particular projects showed 158MB of media via media & quota and the other 242MB One had 47 zipper files the other 21 and across all the user folders 200+ zipper files!

I'm a bit concerned how widespread this could be with people unaware of the issue. I know it's not best practice to upload loads of video direct but lets be realistic users are going to do this and sometimes with good reason.

John - This server was also set to 128MB memory limit by defauls so I tried your export.php but that was generating an empty zip download.

I've increased the memory limit to 300MB and with the current 2.1 code and original export.php the above two projects export ok without leaving example_zipper files behind.

However I also noticed looking at the folder via ftp as you suggested John, that if the user cancels the export, even with the memory limit increased, the example zipper doesn't then get removed. I don't know if this would be different with develop or if your export.php would only work with develop John?

But I am wondering if this is a bit of a ticking timebomb for lots of installations! If anyone has SSH access to their installation(s) try find .
-type f -name 'example_zipper*'

Perhaps the temp file name should have been example_zippo! ;-)

We obviously need to crack this one asap...

Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 01 May 2014 14:51
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

Hi Tom,

Yes, definitely, no worries. I'm about to investigate a windows install with both the original code and this new code to see whether I can get it to run.
Will update you.

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk
[xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders [reijnders at tor.nl]
Sent: 01 May 2014 14:46
To: For Xerte technical developers
Subject: [Xerte-dev] Re: example_zipper issue revisited

I will look into this, but I don't have much time at the moment. Perhaps topnight. Otherwise, it'll have to be next week.

Tom

Smith, John schreef op 1-5-2014 15:38:
> Well. Preliminary testing looks good with my new code but i haven't
> tested
on Windows yet - would be interested in getting Tom's take on all of this...
>
> I just added loads of videos via ftp (upload limit set to 7M on this
> server) and checked the memory limit and it is
>
> memory_limit    90M     90M
>
> and with each video being used on a page then i've just successfully
downloaded a 316Mb archive zip without any problems (it took several minutes too so the script didn't timeout) and was able to play the videos contained within... and this is just a cheap Dreamhost shared server, nothing fancy...
>
> On to Windoze...
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of
> the
School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 10:58
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> the problem is when I originally raised this issue on 8th April it
> coincided with discovering loads of these un-removed example_zipper
> files on the sandpit install. At that time I also mentioned the
> separate session issue as a reason I hadn't updated to a newer
> revision and I think you mentioned that you thought Tom had fixed this
> zipper issue in a later revision. So once we'd identified the cause of
> the session issue (or at least Julian's attempt at fixing his FF
> issue) I
updated the sandpit code as well as other servers.
> Now I've been using the 2.1 branch for this rather than develop so I'm
> not sure if Tom's fix is in develop but not 2.1 but I think the code
> is the same at the moment.
>
> In short although I've been using a 62MB file from a project a user
> was unable to export, previously there were loads of these zipper
> files in loads of different user/project folders and there are
> probably quite a few again now prior to my changing the memory limit.
>
> I could try resetting the memory_limit back on my own server and test
> again but my previous tests with the 62MB file certainly weren't
> crashing the server. Let me know if you think its worth testing with
> larger files and the default 128MB limit or if your and David's later
> responses have ruled that out as being worthwhile?
>
> I know Tom said last night that he was going to revisit his fix for
> this so would be good to hear back from Tom once he's done that.
>
> Cheers
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
> John
> Sent: 01 May 2014 09:57
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi Ron,
>
> Have been thinking about your issue and just wondering whether you
> might have inadvertently chosen the worst file size possible for the
> video with 128Mb memory available to PHP. Let me explain (and this is
> entirely hypothetical)...
>
> The zipper creates a blank object (<0.1Mb memory used) The library
> loads all the other files individually and zips them into the object
> (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or
> under) The library then loads the mp4 (now it has the object = 2Mb and
> the video = 62Mb in
> memory) It adds it to the zip object but it's also still in memory
> (mp4 doesn't compress at all so we have 2 + 62 + 62 =126Mb which is
> less than
> 128Mb) It then closes the connection to the mp4 and destroys the
> object (memory now = 64Mb for the zip)
>
> I'm wondering whether a 64 Mb or slightly bigger will cause you to be
> able to crash that server too. Is that something you could try?
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of
> the School Learning Technologists will pick up the job. This will
> ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> [ronm at mitchellmedia.co.uk]
> Sent: 01 May 2014 09:09
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi John
> I was just going to test your export code on a server where I don't
> have access to change php settings so the memory limit is fixed at
> 128MB. I know what David means that ideally we want cross platform
> code but as you say there are inevitably different
> solutions/restrictions for some e.g. no control over memory limit. In
> fact the upload limit on this particular shared server is set at the
> default 2MB so I had to upload the test 62MB mp4 via ftp. However I'm
> now confused further because this is the same XOT code as the other
> servers I was using for testing so I expected the export to fail when
> testing before replacing the export code with yours and yet it didn't
> fail and I was able to download the zip which didn't contain any
example_zipper files.
>
> I'm glad you're able to reproduce the issue but not sure how a server
> with 128MB memory limit is able to export ok without a change of code
> or memory limit :-(
>
>
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
> John
> Sent: 01 May 2014 08:38
> To: xerte-dev at lists.nottingham.ac.uk
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> All i know Dave is that, with the commented code uncommented, my
> system couldn't download a 56Mb an 79Mb zip but with this code 79Mb
> came down no problem.
>
> We may have to use what we have access to on different systems. There
> may be a similar way to pipe the result into windows or we fall back
> to the old code and crank the memory requirements up yo 300Mb which
> not everyone can most likely do anyway...
>
> Regards
>
> John Smith
> Learning Technologist
> School of Health and Life Sciences
>
> Sent from Samsung Galaxy SII
>
>
>
> David Goodwin <david at palepurple.co.uk> wrote:
>
>
>
> <snip>
> Uses the unix command zip and pipes the stream back into PHP in 8192
> chunks
>
> <snip>
> Also, might not work on Windows...
>
> So therefore it's not worth using?
>
>
> Technically there is no benefit to doing the fread(....) over a
> fpassthru(...) fpassthru does something like 8192 chunks internally.
>
>
> David.
>
> --
> --
> Pale Purple Ltd
> http://www.palepurple.co.uk @PalePurpleLtd
> 0845 0046746 / 07792 380669
> "Bespoke web and mobile application development"
>
>
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
message in
> error, please send it back to me, and immediately delete it.   Please do
not
> use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of
Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee
> and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>

--
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196


_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number
SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From david at palepurple.co.uk  Fri May  2 15:28:04 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Fri, 02 May 2014 15:28:04 +0100
Subject: [Xerte-dev] Re: example_zipper issue revisited [pull request]
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0CECE82C11@EXCHANGE1.ad.nottingham.ac.uk>
References: <jls9e2ctv83cofmw668se8cb.1398929858592@email.android.com>,
	<002d01cf6514$b369cb40$1a3d61c0$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101B7@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<007801cf6523$f7747920$e65d6b60$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C5@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<53625027.2030801@tor.nl>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101C6@ITSEMBXCLUS.enterprise.gcal.ac.uk>	<008d01cf65f5$91db1240$b59136c0$@co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0CECE82BC0@EXCHANGE1.ad.nottingham.ac.uk>	<00cb01cf660e$2369df30$6a3d9d90$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CECE82C11@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <5363AB74.1010804@palepurple.co.uk>

Shorter summary - get someone to merge :

https://github.com/thexerteproject/xerteonlinetoolkits/pull/57


It includes a fix for the Zip file memory issue. If PHP has the 'zip' 
extension loaded, then it uses it to create the zip file. If it doesn't, 
then we fall back to using the existing variant.

It also moves the zip file creation out of USER-FILES into wherever the 
system stores temporary files ... and won't clobber others / isn't 
predictable etc.

I've tested it where with LOs of ~300mb, with a 64mb PHP memory limit. 
It works.


David.


On 02/05/14 14:52, Julian Tenney wrote:
> Got it. OK.
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> Sent: 02 May 2014 14:55
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> For projects containing lots or large files when the user tries to export it fails and leaves behind files named example_zipper_ and doesn't clear them up so the situation gets worse each time. It only occurs if the size of the project and temporary zip exceeds the memory_limit set on the server. Look in USER-FILES for example_zipper files?
>
> HTH
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
> Sent: 02 May 2014 14:08
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> I haven't followed all the details here, but can you summarise in words I'd understand?
>
> thanks
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
> Sent: 02 May 2014 11:59
> To: 'For Xerte technical developers'
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi all
> My inclinations is to scream aaaarrrrrgh! Having had what seems like coincidental support requests but is clearly all part of the same issue so I've lost track of how long this issue has existed and how widespread it might be. To explain...
>
> Late yesterday I received a request from one of the HEA projects regarding two of their students being unable to export their projects. After investigation it turns out both students have been uploading quite a few videos as part of their LO's in this case .flv's and so sure enough the projects weren't exporting and creatining more and more example_zipper files after each attempt.
>
> One of these particular projects showed 158MB of media via media & quota and the other 242MB One had 47 zipper files the other 21 and across all the user folders 200+ zipper files!
>
> I'm a bit concerned how widespread this could be with people unaware of the issue. I know it's not best practice to upload loads of video direct but lets be realistic users are going to do this and sometimes with good reason.
>
> John - This server was also set to 128MB memory limit by defauls so I tried your export.php but that was generating an empty zip download.
>
> I've increased the memory limit to 300MB and with the current 2.1 code and original export.php the above two projects export ok without leaving example_zipper files behind.
>
> However I also noticed looking at the folder via ftp as you suggested John, that if the user cancels the export, even with the memory limit increased, the example zipper doesn't then get removed. I don't know if this would be different with develop or if your export.php would only work with develop John?
>
> But I am wondering if this is a bit of a ticking timebomb for lots of installations! If anyone has SSH access to their installation(s) try find .
> -type f -name 'example_zipper*'
>
> Perhaps the temp file name should have been example_zippo! ;-)
>
> We obviously need to crack this one asap...
>
> Ron
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
> Sent: 01 May 2014 14:51
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> Hi Tom,
>
> Yes, definitely, no worries. I'm about to investigate a windows install with both the original code and this new code to see whether I can get it to run.
> Will update you.
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders [reijnders at tor.nl]
> Sent: 01 May 2014 14:46
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: example_zipper issue revisited
>
> I will look into this, but I don't have much time at the moment. Perhaps topnight. Otherwise, it'll have to be next week.
>
> Tom
>
> Smith, John schreef op 1-5-2014 15:38:
>> Well. Preliminary testing looks good with my new code but i haven't
>> tested
> on Windows yet - would be interested in getting Tom's take on all of this...
>>
>> I just added loads of videos via ftp (upload limit set to 7M on this
>> server) and checked the memory limit and it is
>>
>> memory_limit    90M     90M
>>
>> and with each video being used on a page then i've just successfully
> downloaded a 316Mb archive zip without any problems (it took several minutes too so the script didn't timeout) and was able to play the videos contained within... and this is just a cheap Dreamhost shared server, nothing fancy...
>>
>> On to Windoze...
>>
>> Regards,
>>
>> John Smith | Learning Technologist
>> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
>> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>>
>> Please address ALL support requests to hlsblt at gcu.ac.uk where one of
>> the
> School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
>> ________________________________________
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
>> [ronm at mitchellmedia.co.uk]
>> Sent: 01 May 2014 10:58
>> To: 'For Xerte technical developers'
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> Hi John
>> the problem is when I originally raised this issue on 8th April it
>> coincided with discovering loads of these un-removed example_zipper
>> files on the sandpit install. At that time I also mentioned the
>> separate session issue as a reason I hadn't updated to a newer
>> revision and I think you mentioned that you thought Tom had fixed this
>> zipper issue in a later revision. So once we'd identified the cause of
>> the session issue (or at least Julian's attempt at fixing his FF
>> issue) I
> updated the sandpit code as well as other servers.
>> Now I've been using the 2.1 branch for this rather than develop so I'm
>> not sure if Tom's fix is in develop but not 2.1 but I think the code
>> is the same at the moment.
>>
>> In short although I've been using a 62MB file from a project a user
>> was unable to export, previously there were loads of these zipper
>> files in loads of different user/project folders and there are
>> probably quite a few again now prior to my changing the memory limit.
>>
>> I could try resetting the memory_limit back on my own server and test
>> again but my previous tests with the 62MB file certainly weren't
>> crashing the server. Let me know if you think its worth testing with
>> larger files and the default 128MB limit or if your and David's later
>> responses have ruled that out as being worthwhile?
>>
>> I know Tom said last night that he was going to revisit his fix for
>> this so would be good to hear back from Tom once he's done that.
>>
>> Cheers
>> Ron
>>
>> -----Original Message-----
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
>> John
>> Sent: 01 May 2014 09:57
>> To: For Xerte technical developers
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> Hi Ron,
>>
>> Have been thinking about your issue and just wondering whether you
>> might have inadvertently chosen the worst file size possible for the
>> video with 128Mb memory available to PHP. Let me explain (and this is
>> entirely hypothetical)...
>>
>> The zipper creates a blank object (<0.1Mb memory used) The library
>> loads all the other files individually and zips them into the object
>> (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or
>> under) The library then loads the mp4 (now it has the object = 2Mb and
>> the video = 62Mb in
>> memory) It adds it to the zip object but it's also still in memory
>> (mp4 doesn't compress at all so we have 2 + 62 + 62 =126Mb which is
>> less than
>> 128Mb) It then closes the connection to the mp4 and destroys the
>> object (memory now = 64Mb for the zip)
>>
>> I'm wondering whether a 64 Mb or slightly bigger will cause you to be
>> able to crash that server too. Is that something you could try?
>>
>> Regards,
>>
>> John Smith | Learning Technologist
>> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
>> Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>>
>> Please address ALL support requests to hlsblt at gcu.ac.uk where one of
>> the School Learning Technologists will pick up the job. This will
>> ensure that all jobs are completed as promptly as possible.
>> ________________________________________
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
>> [ronm at mitchellmedia.co.uk]
>> Sent: 01 May 2014 09:09
>> To: 'For Xerte technical developers'
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> Hi John
>> I was just going to test your export code on a server where I don't
>> have access to change php settings so the memory limit is fixed at
>> 128MB. I know what David means that ideally we want cross platform
>> code but as you say there are inevitably different
>> solutions/restrictions for some e.g. no control over memory limit. In
>> fact the upload limit on this particular shared server is set at the
>> default 2MB so I had to upload the test 62MB mp4 via ftp. However I'm
>> now confused further because this is the same XOT code as the other
>> servers I was using for testing so I expected the export to fail when
>> testing before replacing the export code with yours and yet it didn't
>> fail and I was able to download the zip which didn't contain any
> example_zipper files.
>>
>> I'm glad you're able to reproduce the issue but not sure how a server
>> with 128MB memory limit is able to export ok without a change of code
>> or memory limit :-(
>>
>>
>>
>> -----Original Message-----
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
>> John
>> Sent: 01 May 2014 08:38
>> To: xerte-dev at lists.nottingham.ac.uk
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> All i know Dave is that, with the commented code uncommented, my
>> system couldn't download a 56Mb an 79Mb zip but with this code 79Mb
>> came down no problem.
>>
>> We may have to use what we have access to on different systems. There
>> may be a similar way to pipe the result into windows or we fall back
>> to the old code and crank the memory requirements up yo 300Mb which
>> not everyone can most likely do anyway...
>>
>> Regards
>>
>> John Smith
>> Learning Technologist
>> School of Health and Life Sciences
>>
>> Sent from Samsung Galaxy SII
>>
>>
>>
>> David Goodwin <david at palepurple.co.uk> wrote:
>>
>>
>>
>> <snip>
>> Uses the unix command zip and pipes the stream back into PHP in 8192
>> chunks
>>
>> <snip>
>> Also, might not work on Windows...
>>
>> So therefore it's not worth using?
>>
>>
>> Technically there is no benefit to doing the fread(....) over a
>> fpassthru(...) fpassthru does something like 8192 chunks internally.
>>
>>
>> David.
>>
>> --
>> --
>> Pale Purple Ltd
>> http://www.palepurple.co.uk @PalePurpleLtd
>> 0845 0046746 / 07792 380669
>> "Bespoke web and mobile application development"
>>
>>
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
> message in
>> error, please send it back to me, and immediately delete it.   Please do
> not
>> use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of
> Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>> Glasgow Caledonian University is a registered Scottish charity, number
>> SC021474
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
> message in
>> error, please send it back to me, and immediately delete it.   Please do
> not
>> use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of
> Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>>
>>
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
> message in
>> error, please send it back to me, and immediately delete it.   Please do
> not
>> use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of
> Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>>
>>
>>
>> Glasgow Caledonian University is a registered Scottish charity, number
>> SC021474
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee
>> and may contain confidential information. If you have received this
> message in
>> error, please send it back to me, and immediately delete it.   Please do
> not
>> use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of
> Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>>
>>
>>
>>
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee
>> and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>>
>>
>>
>>
>> Glasgow Caledonian University is a registered Scottish charity, number
>> SC021474
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee
>> and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>>
>>
>>
>>
>
> --
> --
>
> Tom Reijnders
> TOR Informatica
> Chopinlaan 27
> 5242HM Rosmalen
> Tel: 073 5226191
> Fax: 073 5226196
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do not
> use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
>
>
>
>

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From xerte at pgogywebstuff.com  Fri May  2 18:37:39 2014
From: xerte at pgogywebstuff.com (Pat L (pgogy))
Date: Fri, 2 May 2014 18:37:39 +0100
Subject: [Xerte-dev] Re: example_zipper issue
Message-ID: <CAO2PinNODhMr9vmvfJeGj65AddAvU6hfgBP0xHee-FZ+KHn_fQ@mail.gmail.com>

This has been there from day one. We had Nottingham users before it was
open source who had this problem.

David has fixed size, but timeout might also happen as well.

Why are people exporting? Is it to edit elsewhere or to use? If it is edit
then maybe they don't need all of it? Leave media relative export? If it is
to use then maybe making media into links makes sense. Smaller zip?
>
> On 2 May 2014 11:59, "Ron Mitchell" <ronm at mitchellmedia.co.uk> wrote:
>>
>> Hi all
>> My inclinations is to scream aaaarrrrrgh! Having had what seems like
>> coincidental support requests but is clearly all part of the same issue
so
>> I've lost track of how long this issue has existed and how widespread it
>> might be. To explain...
>>
>> Late yesterday I received a request from one of the HEA projects
regarding
>> two of their students being unable to export their projects. After
>> investigation it turns out both students have been uploading quite a few
>> videos as part of their LO's in this case .flv's and so sure enough the
>> projects weren't exporting and creatining more and more example_zipper
files
>> after each attempt.
>>
>> One of these particular projects showed 158MB of media via media & quota
and
>> the other 242MB
>> One had 47 zipper files the other 21 and across all the user folders 200+
>> zipper files!
>>
>> I'm a bit concerned how widespread this could be with people unaware of
the
>> issue. I know it's not best practice to upload loads of video direct but
>> lets be realistic users are going to do this and sometimes with good
reason.
>>
>> John - This server was also set to 128MB memory limit by defauls so I
tried
>> your export.php but that was generating an empty zip download.
>>
>> I've increased the memory limit to 300MB and with the current 2.1 code
and
>> original export.php the above two projects export ok without leaving
>> example_zipper files behind.
>>
>> However I also noticed looking at the folder via ftp as you suggested
John,
>> that if the user cancels the export, even with the memory limit
increased,
>> the example zipper doesn't then get removed. I don't know if this would
be
>> different with develop or if your export.php would only work with develop
>> John?
>>
>> But I am wondering if this is a bit of a ticking timebomb for lots of
>> installations! If anyone has SSH access to their installation(s) try
find .
>> -type f -name 'example_zipper*'
>>
>> Perhaps the temp file name should have been example_zippo! ;-)
>>
>> We obviously need to crack this one asap...
>>
>> Ron
>>
>> -----Original Message-----
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
John
>> Sent: 01 May 2014 14:51
>> To: For Xerte technical developers
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> Hi Tom,
>>
>> Yes, definitely, no worries. I'm about to investigate a windows install
with
>> both the original code and this new code to see whether I can get it to
run.
>> Will update you.
>>
>> Regards,
>>
>> John Smith | Learning Technologist
>> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
Glasgow
>> Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>>
>> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
>> School Learning Technologists will pick up the job. This will ensure that
>> all jobs are completed as promptly as possible.
>> ________________________________________
>> From: xerte-dev-bounces at lists.nottingham.ac.uk
>> [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders
>> [reijnders at tor.nl]
>> Sent: 01 May 2014 14:46
>> To: For Xerte technical developers
>> Subject: [Xerte-dev] Re: example_zipper issue revisited
>>
>> I will look into this, but I don't have much time at the moment. Perhaps
>> topnight. Otherwise, it'll have to be next week.
>>
>> Tom
>>
>> Smith, John schreef op 1-5-2014 15:38:
>> > Well. Preliminary testing looks good with my new code but i haven't
tested
>> on Windows yet - would be interested in getting Tom's take on all of
this...
>> >
>> > I just added loads of videos via ftp (upload limit set to 7M on this
>> > server) and checked the memory limit and it is
>> >
>> > memory_limit    90M     90M
>> >
>> > and with each video being used on a page then i've just successfully
>> downloaded a 316Mb archive zip without any problems (it took several
minutes
>> too so the script didn't timeout) and was able to play the videos
contained
>> within... and this is just a cheap Dreamhost shared server, nothing
fancy...
>> >
>> > On to Windoze...
>> >
>> > Regards,
>> >
>> > John Smith | Learning Technologist
>> > Room A251, Govan Mbeki Building | School of Health & Life Sciences |
>> > Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>> >
>> > Please address ALL support requests to hlsblt at gcu.ac.uk where one of
the
>> School Learning Technologists will pick up the job. This will ensure that
>> all jobs are completed as promptly as possible.
>> > ________________________________________
>> > From: xerte-dev-bounces at lists.nottingham.ac.uk
>> > [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
>> > [ronm at mitchellmedia.co.uk]
>> > Sent: 01 May 2014 10:58
>> > To: 'For Xerte technical developers'
>> > Subject: [Xerte-dev] Re: example_zipper issue revisited
>> >
>> > Hi John
>> > the problem is when I originally raised this issue on 8th April it
>> > coincided with discovering loads of these un-removed example_zipper
>> > files on the sandpit install. At that time I also mentioned the
>> > separate session issue as a reason I hadn't updated to a newer
>> > revision and I think you mentioned that you thought Tom had fixed this
>> > zipper issue in a later revision. So once we'd identified the cause of
>> > the session issue (or at least Julian's attempt at fixing his FF
issue) I
>> updated the sandpit code as well as other servers.
>> > Now I've been using the 2.1 branch for this rather than develop so I'm
>> > not sure if Tom's fix is in develop but not 2.1 but I think the code
>> > is the same at the moment.
>> >
>> > In short although I've been using a 62MB file from a project a user
>> > was unable to export, previously there were loads of these zipper
>> > files in loads of different user/project folders and there are
>> > probably quite a few again now prior to my changing the memory limit.
>> >
>> > I could try resetting the memory_limit back on my own server and test
>> > again but my previous tests with the 62MB file certainly weren't
>> > crashing the server. Let me know if you think its worth testing with
>> > larger files and the default 128MB limit or if your and David's later
>> > responses have ruled that out as being worthwhile?
>> >
>> > I know Tom said last night that he was going to revisit his fix for
>> > this so would be good to hear back from Tom once he's done that.
>> >
>> > Cheers
>> > Ron
>> >
>> > -----Original Message-----
>> > From: xerte-dev-bounces at lists.nottingham.ac.uk
>> > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
>> > John
>> > Sent: 01 May 2014 09:57
>> > To: For Xerte technical developers
>> > Subject: [Xerte-dev] Re: example_zipper issue revisited
>> >
>> > Hi Ron,
>> >
>> > Have been thinking about your issue and just wondering whether you
>> > might have inadvertently chosen the worst file size possible for the
>> > video with 128Mb memory available to PHP. Let me explain (and this is
>> > entirely hypothetical)...
>> >
>> > The zipper creates a blank object (<0.1Mb memory used) The library
>> > loads all the other files individually and zips them into the object
>> > (0Mb -> compressed size of all js/html is perhaps around 1-2Mb or
>> > under) The library then loads the mp4 (now it has the object = 2Mb and
>> > the video = 62Mb in
>> > memory) It adds it to the zip object but it's also still in memory
>> > (mp4 doesn't compress at all so we have 2 + 62 + 62 =126Mb which is
>> > less than
>> > 128Mb) It then closes the connection to the mp4 and destroys the
>> > object (memory now = 64Mb for the zip)
>> >
>> > I'm wondering whether a 64 Mb or slightly bigger will cause you to be
>> > able to crash that server too. Is that something you could try?
>> >
>> > Regards,
>> >
>> > John Smith | Learning Technologist
>> > Room A251, Govan Mbeki Building | School of Health & Life Sciences |
>> > Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA
>> >
>> > Please address ALL support requests to hlsblt at gcu.ac.uk where one of
>> > the School Learning Technologists will pick up the job. This will
>> > ensure that all jobs are completed as promptly as possible.
>> > ________________________________________
>> > From: xerte-dev-bounces at lists.nottingham.ac.uk
>> > [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
>> > [ronm at mitchellmedia.co.uk]
>> > Sent: 01 May 2014 09:09
>> > To: 'For Xerte technical developers'
>> > Subject: [Xerte-dev] Re: example_zipper issue revisited
>> >
>> > Hi John
>> > I was just going to test your export code on a server where I don't
>> > have access to change php settings so the memory limit is fixed at
>> > 128MB. I know what David means that ideally we want cross platform
>> > code but as you say there are inevitably different
>> > solutions/restrictions for some e.g. no control over memory limit. In
>> > fact the upload limit on this particular shared server is set at the
>> > default 2MB so I had to upload the test 62MB mp4 via ftp. However I'm
>> > now confused further because this is the same XOT code as the other
>> > servers I was using for testing so I expected the export to fail when
>> > testing before replacing the export code with yours and yet it didn't
>> > fail and I was able to download the zip which didn't contain any
>> example_zipper files.
>> >
>> > I'm glad you're able to reproduce the issue but not sure how a server
>> > with 128MB memory limit is able to export ok without a change of code
>> > or memory limit :-(
>> >
>> >
>> >
>> > -----Original Message-----
>> > From: xerte-dev-bounces at lists.nottingham.ac.uk
>> > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith,
>> > John
>> > Sent: 01 May 2014 08:38
>> > To: xerte-dev at lists.nottingham.ac.uk
>> > Subject: [Xerte-dev] Re: example_zipper issue revisited
>> >
>> > All i know Dave is that, with the commented code uncommented, my
>> > system couldn't download a 56Mb an 79Mb zip but with this code 79Mb
>> > came down no problem.
>> >
>> > We may have to use what we have access to on different systems. There
>> > may be a similar way to pipe the result into windows or we fall back
>> > to the old code and crank the memory requirements up yo 300Mb which
>> > not everyone can most likely do anyway...
>> >
>> > Regards
>> >
>> > John Smith
>> > Learning Technologist
>> > School of Health and Life Sciences
>> >
>> > Sent from Samsung Galaxy SII
>> >
>> >
>> >
>> > David Goodwin <david at palepurple.co.uk> wrote:
>> >
>> >
>> >
>> > <snip>
>> > Uses the unix command zip and pipes the stream back into PHP in 8192
>> > chunks
>> >
>> > <snip>
>> > Also, might not work on Windows...
>> >
>> > So therefore it's not worth using?
>> >
>> >
>> > Technically there is no benefit to doing the fread(....) over a
>> > fpassthru(...) fpassthru does something like 8192 chunks internally.
>> >
>> >
>> > David.
>> >
>> > --
>> > --
>> > Pale Purple Ltd
>> > http://www.palepurple.co.uk @PalePurpleLtd
>> > 0845 0046746 / 07792 380669
>> > "Bespoke web and mobile application development"
>> >
>> >
>> > This message and any attachment are intended solely for the addressee
>> > and may contain confidential information. If you have received this
>> message in
>> > error, please send it back to me, and immediately delete it.   Please
do
>> not
>> > use, copy or disclose the information contained in this message or in
>> > any attachment.  Any views or opinions expressed by the author of this
>> > email do not necessarily reflect the views of the University of
>> Nottingham.
>> >
>> > This message has been checked for viruses but the contents of an
>> > attachment may still contain software viruses which could damage your
>> > computer system, you are advised to perform your own checks. Email
>> > communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>> >
>> >
>> > Glasgow Caledonian University is a registered Scottish charity, number
>> > SC021474
>> >
>> > _______________________________________________
>> > Xerte-dev mailing list
>> > Xerte-dev at lists.nottingham.ac.uk
>> > http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> > This message and any attachment are intended solely for the addressee
>> > and may contain confidential information. If you have received this
>> message in
>> > error, please send it back to me, and immediately delete it.   Please
do
>> not
>> > use, copy or disclose the information contained in this message or in
>> > any attachment.  Any views or opinions expressed by the author of this
>> > email do not necessarily reflect the views of the University of
>> Nottingham.
>> >
>> > This message has been checked for viruses but the contents of an
>> > attachment may still contain software viruses which could damage your
>> > computer system, you are advised to perform your own checks. Email
>> > communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>> >
>> >
>> >
>> >
>> >
>> > _______________________________________________
>> > Xerte-dev mailing list
>> > Xerte-dev at lists.nottingham.ac.uk
>> > http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> > This message and any attachment are intended solely for the addressee
>> > and may contain confidential information. If you have received this
>> message in
>> > error, please send it back to me, and immediately delete it.   Please
do
>> not
>> > use, copy or disclose the information contained in this message or in
>> > any attachment.  Any views or opinions expressed by the author of this
>> > email do not necessarily reflect the views of the University of
>> Nottingham.
>> >
>> > This message has been checked for viruses but the contents of an
>> > attachment may still contain software viruses which could damage your
>> > computer system, you are advised to perform your own checks. Email
>> > communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>> >
>> >
>> >
>> >
>> >
>> > Glasgow Caledonian University is a registered Scottish charity, number
>> > SC021474
>> >
>> > _______________________________________________
>> > Xerte-dev mailing list
>> > Xerte-dev at lists.nottingham.ac.uk
>> > http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> > This message and any attachment are intended solely for the addressee
>> > and may contain confidential information. If you have received this
>> message in
>> > error, please send it back to me, and immediately delete it.   Please
do
>> not
>> > use, copy or disclose the information contained in this message or in
>> > any attachment.  Any views or opinions expressed by the author of this
>> > email do not necessarily reflect the views of the University of
>> Nottingham.
>> >
>> > This message has been checked for viruses but the contents of an
>> > attachment may still contain software viruses which could damage your
>> > computer system, you are advised to perform your own checks. Email
>> > communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>> >
>> >
>> >
>> >
>> >
>> > _______________________________________________
>> > Xerte-dev mailing list
>> > Xerte-dev at lists.nottingham.ac.uk
>> > http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> > This message and any attachment are intended solely for the addressee
and
>> may contain confidential information. If you have received this message
in
>> error, please send it back to me, and immediately delete it.   Please do
not
>> use, copy or disclose the information contained in this message or in any
>> attachment.  Any views or opinions expressed by the author of this email
do
>> not necessarily reflect the views of the University of Nottingham.
>> >
>> > This message has been checked for viruses but the contents of an
>> > attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>> >
>> >
>> >
>> >
>> > Glasgow Caledonian University is a registered Scottish charity, number
>> > SC021474
>> >
>> > _______________________________________________
>> > Xerte-dev mailing list
>> > Xerte-dev at lists.nottingham.ac.uk
>> > http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> > This message and any attachment are intended solely for the addressee
and
>> may contain confidential information. If you have received this message
in
>> error, please send it back to me, and immediately delete it.   Please do
not
>> use, copy or disclose the information contained in this message or in any
>> attachment.  Any views or opinions expressed by the author of this email
do
>> not necessarily reflect the views of the University of Nottingham.
>> >
>> > This message has been checked for viruses but the contents of an
>> > attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>> >
>> >
>> >
>> >
>>
>> --
>> --
>>
>> Tom Reijnders
>> TOR Informatica
>> Chopinlaan 27
>> 5242HM Rosmalen
>> Tel: 073 5226191
>> Fax: 073 5226196
>>
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee and
>> may contain confidential information. If you have received this message
in
>> error, please send it back to me, and immediately delete it.   Please do
not
>> use, copy or disclose the information contained in this message or in any
>> attachment.  Any views or opinions expressed by the author of this email
do
>> not necessarily reflect the views of the University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
attachment
>> may still contain software viruses which could damage your computer
system,
>> you are advised to perform your own checks. Email communications with the
>> University of Nottingham may be monitored as permitted by UK legislation.
>>
>>
>>
>>
>>
>> Glasgow Caledonian University is a registered Scottish charity, number
>> SC021474
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee and
>> may contain confidential information. If you have received this message
in
>> error, please send it back to me, and immediately delete it.   Please do
not
>> use, copy or disclose the information contained in this message or in any
>> attachment.  Any views or opinions expressed by the author of this email
do
>> not necessarily reflect the views of the University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
attachment
>> may still contain software viruses which could damage your computer
system,
>> you are advised to perform your own checks. Email communications with the
>> University of Nottingham may be monitored as permitted by UK legislation.
>>
>>
>>
>>
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>> This message and any attachment are intended solely for the addressee
and may contain confidential information. If you have received this message
in error, please send it back to me, and immediately delete it.   Please do
not use, copy or disclose the information contained in this message or in
any attachment.  Any views or opinions expressed by the author of this
email do not necessarily reflect the views of the University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
attachment
>> may still contain software viruses which could damage your computer
system, you are advised to perform your own checks. Email communications
with the University of Nottingham may be monitored as permitted by UK
legislation.
>>
>>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140502/1c2b24f7/attachment-0001.html>

From Julian.Tenney at nottingham.ac.uk  Thu May  8 08:45:13 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Thu, 8 May 2014 08:45:13 +0100
Subject: [Xerte-dev] FW: Flickr API Going SSL-Only on June 27th, 2014
In-Reply-To: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>
References: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>

This is going to have an impact I think. I?ll investigate.

From: Flickr Mail [mailto:no-reply-flickrmail at flickr.com]
Sent: 07 May 2014 18:19
To: julian.tenney at nottingham.ac.uk
Subject: Flickr API Going SSL-Only on June 27th, 2014

We see that you own an active API key for the Flickr API, and wanted make sure you know about some upcoming changes to the API related to SSL.

As you may have read on the Flickr code blog<http://code.flickr.net/2014/04/30/flickr-api-going-ssl-only-on-june-27th-2014/>, we've updated all of the API endpoints to fully support SSL. You no longer need to use a special subdomain to access the API securely, and we want to make sure that all API communication adopts SSL.

Many of you have moved over to using SSL already, which is great! If you haven't, it shouldn't be much more work than updating your apps to call the API at: https://api.flickr.com/

We'll be updating the blog with information as we get close to the switchover date, but here's the most important info:

New API keys will be issued for HTTPS access only: as of 6 May 2014
First blackout test window: 3 June 2014, 10:00-12:00 Pacific Daylight Time (PDT)
Second blackout test window: 17 June 2014, 18:00-20:00 (PDT)
Non-SSL API deprecated: 27 June 2014, 10:00 (PDT)

During the blackout tests and after the 27th of June, calls to the Flickr API that are not over SSL will fail with an HTTP error code 403. We will also return an error message with (Flickr) code 95 and msg "SSL is required".

Here are example error responses:

XML (REST):

<?xml version="1.0" encoding="utf-8" ?> <rsp stat="fail"> <err code="95" msg="SSL is required" /> </rsp>

JSONP:

jsonFlickrApi({ "stat": "fail", "code": "95", "message": "SSL is required" })

JSON:

{ "stat": "fail", "code": "95", "message": "SSL is required" }

We realize that this change might be more difficult for some. We will follow the Developer Support Group<https://groups.yahoo.com/neo/groups/yws-flickr/info> closely, so please let us hear your questions.

If you have a question about the API transition to SSL that you feel is unique to your situation or is more appropriate to handle via conversation directly with Flickr, send an email to flickr-api-help at yahoo-inc.com<mailto:flickr-api-help at yahoo-inc.com>. We will monitor that mailbox through July 2014.

Thank you for being a member of our API community,
The Flickreenos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140508/f998f7c4/attachment.html>

From Julian.Tenney at nottingham.ac.uk  Thu May  8 08:59:07 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Thu, 8 May 2014 08:59:07 +0100
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>
References: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0CED09699D@EXCHANGE1.ad.nottingham.ac.uk>

It appears as simple as changing the url to https from http. I?ve committed those changes, but not tested,

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 08 May 2014 08:45
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Flickr API Going SSL-Only on June 27th, 2014

This is going to have an impact I think. I?ll investigate.

From: Flickr Mail [mailto:no-reply-flickrmail at flickr.com]
Sent: 07 May 2014 18:19
To: julian.tenney at nottingham.ac.uk<mailto:julian.tenney at nottingham.ac.uk>
Subject: Flickr API Going SSL-Only on June 27th, 2014

We see that you own an active API key for the Flickr API, and wanted make sure you know about some upcoming changes to the API related to SSL.

As you may have read on the Flickr code blog<http://code.flickr.net/2014/04/30/flickr-api-going-ssl-only-on-june-27th-2014/>, we've updated all of the API endpoints to fully support SSL. You no longer need to use a special subdomain to access the API securely, and we want to make sure that all API communication adopts SSL.

Many of you have moved over to using SSL already, which is great! If you haven't, it shouldn't be much more work than updating your apps to call the API at: https://api.flickr.com/

We'll be updating the blog with information as we get close to the switchover date, but here's the most important info:

New API keys will be issued for HTTPS access only: as of 6 May 2014
First blackout test window: 3 June 2014, 10:00-12:00 Pacific Daylight Time (PDT)
Second blackout test window: 17 June 2014, 18:00-20:00 (PDT)
Non-SSL API deprecated: 27 June 2014, 10:00 (PDT)

During the blackout tests and after the 27th of June, calls to the Flickr API that are not over SSL will fail with an HTTP error code 403. We will also return an error message with (Flickr) code 95 and msg "SSL is required".

Here are example error responses:

XML (REST):

<?xml version="1.0" encoding="utf-8" ?> <rsp stat="fail"> <err code="95" msg="SSL is required" /> </rsp>

JSONP:

jsonFlickrApi({ "stat": "fail", "code": "95", "message": "SSL is required" })

JSON:

{ "stat": "fail", "code": "95", "message": "SSL is required" }

We realize that this change might be more difficult for some. We will follow the Developer Support Group<https://groups.yahoo.com/neo/groups/yws-flickr/info> closely, so please let us hear your questions.

If you have a question about the API transition to SSL that you feel is unique to your situation or is more appropriate to handle via conversation directly with Flickr, send an email to flickr-api-help at yahoo-inc.com<mailto:flickr-api-help at yahoo-inc.com>. We will monitor that mailbox through July 2014.

Thank you for being a member of our API community,
The Flickreenos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140508/2c43ab2a/attachment.html>

From Julian.Tenney at nottingham.ac.uk  Thu May  8 09:26:01 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Thu, 8 May 2014 09:26:01 +0100
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
References: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0CED0969CF@EXCHANGE1.ad.nottingham.ac.uk>

Yep, seems to work on our server,

J

From: Tenney Julian
Sent: 08 May 2014 08:59
To: 'For Xerte technical developers'
Subject: RE: Flickr API Going SSL-Only on June 27th, 2014

It appears as simple as changing the url to https from http. I?ve committed those changes, but not tested,

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 08 May 2014 08:45
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Flickr API Going SSL-Only on June 27th, 2014

This is going to have an impact I think. I?ll investigate.

From: Flickr Mail [mailto:no-reply-flickrmail at flickr.com]
Sent: 07 May 2014 18:19
To: julian.tenney at nottingham.ac.uk<mailto:julian.tenney at nottingham.ac.uk>
Subject: Flickr API Going SSL-Only on June 27th, 2014

We see that you own an active API key for the Flickr API, and wanted make sure you know about some upcoming changes to the API related to SSL.

As you may have read on the Flickr code blog<http://code.flickr.net/2014/04/30/flickr-api-going-ssl-only-on-june-27th-2014/>, we've updated all of the API endpoints to fully support SSL. You no longer need to use a special subdomain to access the API securely, and we want to make sure that all API communication adopts SSL.

Many of you have moved over to using SSL already, which is great! If you haven't, it shouldn't be much more work than updating your apps to call the API at: https://api.flickr.com/

We'll be updating the blog with information as we get close to the switchover date, but here's the most important info:

New API keys will be issued for HTTPS access only: as of 6 May 2014
First blackout test window: 3 June 2014, 10:00-12:00 Pacific Daylight Time (PDT)
Second blackout test window: 17 June 2014, 18:00-20:00 (PDT)
Non-SSL API deprecated: 27 June 2014, 10:00 (PDT)

During the blackout tests and after the 27th of June, calls to the Flickr API that are not over SSL will fail with an HTTP error code 403. We will also return an error message with (Flickr) code 95 and msg "SSL is required".

Here are example error responses:

XML (REST):

<?xml version="1.0" encoding="utf-8" ?> <rsp stat="fail"> <err code="95" msg="SSL is required" /> </rsp>

JSONP:

jsonFlickrApi({ "stat": "fail", "code": "95", "message": "SSL is required" })

JSON:

{ "stat": "fail", "code": "95", "message": "SSL is required" }

We realize that this change might be more difficult for some. We will follow the Developer Support Group<https://groups.yahoo.com/neo/groups/yws-flickr/info> closely, so please let us hear your questions.

If you have a question about the API transition to SSL that you feel is unique to your situation or is more appropriate to handle via conversation directly with Flickr, send an email to flickr-api-help at yahoo-inc.com<mailto:flickr-api-help at yahoo-inc.com>. We will monitor that mailbox through July 2014.

Thank you for being a member of our API community,
The Flickreenos
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140508/1f4169d8/attachment-0001.html>

From J.J.Smith at gcu.ac.uk  Thu May  8 09:34:13 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 8 May 2014 09:34:13 +0100
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0CED0969CF@EXCHANGE1.ad.nottingham.ac.uk>
References: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0CED0969CF@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101E6@ITSEMBXCLUS.enterprise.gcal.ac.uk>

It might still have security implications on latest browsers where XOT sits on http and access Flickr on https... similar to recent Youtube etc problems...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 08 May 2014 09:26
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014

Yep, seems to work on our server,

J

From: Tenney Julian
Sent: 08 May 2014 08:59
To: 'For Xerte technical developers'
Subject: RE: Flickr API Going SSL-Only on June 27th, 2014

It appears as simple as changing the url to https from http. I?ve committed those changes, but not tested,

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 08 May 2014 08:45
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Flickr API Going SSL-Only on June 27th, 2014

This is going to have an impact I think. I?ll investigate.

From: Flickr Mail [mailto:no-reply-flickrmail at flickr.com]
Sent: 07 May 2014 18:19
To: julian.tenney at nottingham.ac.uk<mailto:julian.tenney at nottingham.ac.uk>
Subject: Flickr API Going SSL-Only on June 27th, 2014

We see that you own an active API key for the Flickr API, and wanted make sure you know about some upcoming changes to the API related to SSL.

As you may have read on the Flickr code blog<http://code.flickr.net/2014/04/30/flickr-api-going-ssl-only-on-june-27th-2014/>, we've updated all of the API endpoints to fully support SSL. You no longer need to use a special subdomain to access the API securely, and we want to make sure that all API communication adopts SSL.

Many of you have moved over to using SSL already, which is great! If you haven't, it shouldn't be much more work than updating your apps to call the API at: https://api.flickr.com/

We'll be updating the blog with information as we get close to the switchover date, but here's the most important info:

New API keys will be issued for HTTPS access only: as of 6 May 2014
First blackout test window: 3 June 2014, 10:00-12:00 Pacific Daylight Time (PDT)
Second blackout test window: 17 June 2014, 18:00-20:00 (PDT)
Non-SSL API deprecated: 27 June 2014, 10:00 (PDT)

During the blackout tests and after the 27th of June, calls to the Flickr API that are not over SSL will fail with an HTTP error code 403. We will also return an error message with (Flickr) code 95 and msg "SSL is required".

Here are example error responses:

XML (REST):

<?xml version="1.0" encoding="utf-8" ?> <rsp stat="fail"> <err code="95" msg="SSL is required" /> </rsp>

JSONP:

jsonFlickrApi({ "stat": "fail", "code": "95", "message": "SSL is required" })

JSON:

{ "stat": "fail", "code": "95", "message": "SSL is required" }

We realize that this change might be more difficult for some. We will follow the Developer Support Group<https://groups.yahoo.com/neo/groups/yws-flickr/info> closely, so please let us hear your questions.

If you have a question about the API transition to SSL that you feel is unique to your situation or is more appropriate to handle via conversation directly with Flickr, send an email to flickr-api-help at yahoo-inc.com<mailto:flickr-api-help at yahoo-inc.com>. We will monitor that mailbox through July 2014.

Thank you for being a member of our API community,
The Flickreenos

Glasgow Caledonian University is a registered Scottish charity, number SC021474


From Julian.Tenney at nottingham.ac.uk  Thu May  8 09:43:07 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Thu, 8 May 2014 09:43:07 +0100
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101E6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0CED0969CF@EXCHANGE1.ad.nottingham.ac.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101E6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0CED0969EE@EXCHANGE1.ad.nottingham.ac.uk>

Maybe, but we don't have any choice but to use https to access the flickr api,

J

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 08 May 2014 09:34
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014

It might still have security implications on latest browsers where XOT sits on http and access Flickr on https... similar to recent Youtube etc problems...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 08 May 2014 09:26
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014

Yep, seems to work on our server,

J

From: Tenney Julian
Sent: 08 May 2014 08:59
To: 'For Xerte technical developers'
Subject: RE: Flickr API Going SSL-Only on June 27th, 2014

It appears as simple as changing the url to https from http. I've committed those changes, but not tested,

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 08 May 2014 08:45
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Flickr API Going SSL-Only on June 27th, 2014

This is going to have an impact I think. I'll investigate.

From: Flickr Mail [mailto:no-reply-flickrmail at flickr.com]
Sent: 07 May 2014 18:19
To: julian.tenney at nottingham.ac.uk<mailto:julian.tenney at nottingham.ac.uk>
Subject: Flickr API Going SSL-Only on June 27th, 2014

We see that you own an active API key for the Flickr API, and wanted make sure you know about some upcoming changes to the API related to SSL.

As you may have read on the Flickr code blog<http://code.flickr.net/2014/04/30/flickr-api-going-ssl-only-on-june-27th-2014/>, we've updated all of the API endpoints to fully support SSL. You no longer need to use a special subdomain to access the API securely, and we want to make sure that all API communication adopts SSL.

Many of you have moved over to using SSL already, which is great! If you haven't, it shouldn't be much more work than updating your apps to call the API at: https://api.flickr.com/

We'll be updating the blog with information as we get close to the switchover date, but here's the most important info:

New API keys will be issued for HTTPS access only: as of 6 May 2014 First blackout test window: 3 June 2014, 10:00-12:00 Pacific Daylight Time (PDT) Second blackout test window: 17 June 2014, 18:00-20:00 (PDT) Non-SSL API deprecated: 27 June 2014, 10:00 (PDT)

During the blackout tests and after the 27th of June, calls to the Flickr API that are not over SSL will fail with an HTTP error code 403. We will also return an error message with (Flickr) code 95 and msg "SSL is required".

Here are example error responses:

XML (REST):

<?xml version="1.0" encoding="utf-8" ?> <rsp stat="fail"> <err code="95" msg="SSL is required" /> </rsp>

JSONP:

jsonFlickrApi({ "stat": "fail", "code": "95", "message": "SSL is required" })

JSON:

{ "stat": "fail", "code": "95", "message": "SSL is required" }

We realize that this change might be more difficult for some. We will follow the Developer Support Group<https://groups.yahoo.com/neo/groups/yws-flickr/info> closely, so please let us hear your questions.

If you have a question about the API transition to SSL that you feel is unique to your situation or is more appropriate to handle via conversation directly with Flickr, send an email to flickr-api-help at yahoo-inc.com<mailto:flickr-api-help at yahoo-inc.com>. We will monitor that mailbox through July 2014.

Thank you for being a member of our API community, The Flickreenos

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From J.J.Smith at gcu.ac.uk  Thu May  8 09:46:31 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 8 May 2014 09:46:31 +0100
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0CED0969EE@EXCHANGE1.ad.nottingham.ac.uk>
References: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0CED0969CF@EXCHANGE1.ad.nottingham.ac.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101E6@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0CED0969EE@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101E9@ITSEMBXCLUS.enterprise.gcal.ac.uk>

True...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 08 May 2014 09:43
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014

Maybe, but we don't have any choice but to use https to access the flickr api,

J

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 08 May 2014 09:34
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014

It might still have security implications on latest browsers where XOT sits on http and access Flickr on https... similar to recent Youtube etc problems...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 08 May 2014 09:26
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014

Yep, seems to work on our server,

J

From: Tenney Julian
Sent: 08 May 2014 08:59
To: 'For Xerte technical developers'
Subject: RE: Flickr API Going SSL-Only on June 27th, 2014

It appears as simple as changing the url to https from http. I've committed those changes, but not tested,

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 08 May 2014 08:45
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Flickr API Going SSL-Only on June 27th, 2014

This is going to have an impact I think. I'll investigate.

From: Flickr Mail [mailto:no-reply-flickrmail at flickr.com]
Sent: 07 May 2014 18:19
To: julian.tenney at nottingham.ac.uk<mailto:julian.tenney at nottingham.ac.uk>
Subject: Flickr API Going SSL-Only on June 27th, 2014

We see that you own an active API key for the Flickr API, and wanted make sure you know about some upcoming changes to the API related to SSL.

As you may have read on the Flickr code blog<http://code.flickr.net/2014/04/30/flickr-api-going-ssl-only-on-june-27th-2014/>, we've updated all of the API endpoints to fully support SSL. You no longer need to use a special subdomain to access the API securely, and we want to make sure that all API communication adopts SSL.

Many of you have moved over to using SSL already, which is great! If you haven't, it shouldn't be much more work than updating your apps to call the API at: https://api.flickr.com/

We'll be updating the blog with information as we get close to the switchover date, but here's the most important info:

New API keys will be issued for HTTPS access only: as of 6 May 2014 First blackout test window: 3 June 2014, 10:00-12:00 Pacific Daylight Time (PDT) Second blackout test window: 17 June 2014, 18:00-20:00 (PDT) Non-SSL API deprecated: 27 June 2014, 10:00 (PDT)

During the blackout tests and after the 27th of June, calls to the Flickr API that are not over SSL will fail with an HTTP error code 403. We will also return an error message with (Flickr) code 95 and msg "SSL is required".

Here are example error responses:

XML (REST):

<?xml version="1.0" encoding="utf-8" ?> <rsp stat="fail"> <err code="95" msg="SSL is required" /> </rsp>

JSONP:

jsonFlickrApi({ "stat": "fail", "code": "95", "message": "SSL is required" })

JSON:

{ "stat": "fail", "code": "95", "message": "SSL is required" }

We realize that this change might be more difficult for some. We will follow the Developer Support Group<https://groups.yahoo.com/neo/groups/yws-flickr/info> closely, so please let us hear your questions.

If you have a question about the API transition to SSL that you feel is unique to your situation or is more appropriate to handle via conversation directly with Flickr, send an email to flickr-api-help at yahoo-inc.com<mailto:flickr-api-help at yahoo-inc.com>. We will monitor that mailbox through July 2014.

Thank you for being a member of our API community, The Flickreenos

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From xerte at pgogywebstuff.com  Thu May  8 09:52:53 2014
From: xerte at pgogywebstuff.com (Pat L (pgogy))
Date: Thu, 8 May 2014 09:52:53 +0100
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101E6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CED0969CF@EXCHANGE1.ad.nottingham.ac.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101E6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <CAO2PinMpMm_k-CMk0JA04oRRys8Vos-=z05C98XETjt5XHT1Vw@mail.gmail.com>

depends on where the script is loaded from? I have http js calling https
APIs without much trouble


On Thu, May 8, 2014 at 9:34 AM, Smith, John <J.J.Smith at gcu.ac.uk> wrote:

> It might still have security implications on latest browsers where XOT
> sits on http and access Flickr on https... similar to recent Youtube etc
> problems...
>
> Regards,
>
> John Smith | Learning Technologist
> Room A251, Govan Mbeki Building | School of Health & Life Sciences |
> Glasgow Caledonian University
> Cowcaddens Road | Glasgow | G4 0BA
>
> Please address ALL support requests to hlsblt at gcu.ac.uk where one of the
> School Learning Technologists will pick up the job. This will ensure that
> all jobs are completed as promptly as possible.
> ________________________________________
> From: xerte-dev-bounces at lists.nottingham.ac.uk [
> xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [
> Julian.Tenney at nottingham.ac.uk]
> Sent: 08 May 2014 09:26
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
>
> Yep, seems to work on our server,
>
> J
>
> From: Tenney Julian
> Sent: 08 May 2014 08:59
> To: 'For Xerte technical developers'
> Subject: RE: Flickr API Going SSL-Only on June 27th, 2014
>
> It appears as simple as changing the url to https from http. I?ve
> committed those changes, but not tested,
>
> From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:
> xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:
> xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
> Sent: 08 May 2014 08:45
> To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk
> <mailto:xerte-dev at lists.nottingham.ac.uk>)
> Subject: [Xerte-dev] FW: Flickr API Going SSL-Only on June 27th, 2014
>
> This is going to have an impact I think. I?ll investigate.
>
> From: Flickr Mail [mailto:no-reply-flickrmail at flickr.com]
> Sent: 07 May 2014 18:19
> To: julian.tenney at nottingham.ac.uk<mailto:julian.tenney at nottingham.ac.uk>
> Subject: Flickr API Going SSL-Only on June 27th, 2014
>
> We see that you own an active API key for the Flickr API, and wanted make
> sure you know about some upcoming changes to the API related to SSL.
>
> As you may have read on the Flickr code blog<
> http://code.flickr.net/2014/04/30/flickr-api-going-ssl-only-on-june-27th-2014/>,
> we've updated all of the API endpoints to fully support SSL. You no longer
> need to use a special subdomain to access the API securely, and we want to
> make sure that all API communication adopts SSL.
>
> Many of you have moved over to using SSL already, which is great! If you
> haven't, it shouldn't be much more work than updating your apps to call the
> API at: https://api.flickr.com/
>
> We'll be updating the blog with information as we get close to the
> switchover date, but here's the most important info:
>
> New API keys will be issued for HTTPS access only: as of 6 May 2014
> First blackout test window: 3 June 2014, 10:00-12:00 Pacific Daylight Time
> (PDT)
> Second blackout test window: 17 June 2014, 18:00-20:00 (PDT)
> Non-SSL API deprecated: 27 June 2014, 10:00 (PDT)
>
> During the blackout tests and after the 27th of June, calls to the Flickr
> API that are not over SSL will fail with an HTTP error code 403. We will
> also return an error message with (Flickr) code 95 and msg "SSL is
> required".
>
> Here are example error responses:
>
> XML (REST):
>
> <?xml version="1.0" encoding="utf-8" ?> <rsp stat="fail"> <err code="95"
> msg="SSL is required" /> </rsp>
>
> JSONP:
>
> jsonFlickrApi({ "stat": "fail", "code": "95", "message": "SSL is required"
> })
>
> JSON:
>
> { "stat": "fail", "code": "95", "message": "SSL is required" }
>
> We realize that this change might be more difficult for some. We will
> follow the Developer Support Group<
> https://groups.yahoo.com/neo/groups/yws-flickr/info> closely, so please
> let us hear your questions.
>
> If you have a question about the API transition to SSL that you feel is
> unique to your situation or is more appropriate to handle via conversation
> directly with Flickr, send an email to flickr-api-help at yahoo-inc.com
> <mailto:flickr-api-help at yahoo-inc.com>. We will monitor that mailbox
> through July 2014.
>
> Thank you for being a member of our API community,
> The Flickreenos
>
> Glasgow Caledonian University is a registered Scottish charity, number
> SC021474
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer
> system, you are advised to perform your own checks. Email communications
> with the University of Nottingham may be monitored as permitted by UK
> legislation.
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140508/0da6cff5/attachment-0001.html>

From reijnders at tor.nl  Thu May  8 11:26:42 2014
From: reijnders at tor.nl (Tom Reijnders)
Date: Thu, 08 May 2014 12:26:42 +0200
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
In-Reply-To: <CAO2PinMpMm_k-CMk0JA04oRRys8Vos-=z05C98XETjt5XHT1Vw@mail.gmail.com>
References: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>	<12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>	<12C67A1EEC419342AF5E59DA31562C3F0CED0969CF@EXCHANGE1.ad.nottingham.ac.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101E6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<CAO2PinMpMm_k-CMk0JA04oRRys8Vos-=z05C98XETjt5XHT1Vw@mail.gmail.com>
Message-ID: <536B5BE2.5010101@tor.nl>

As far as I know, only the other way around causes problems, i.e. 
calling http from https.

In that sense this would be a good change  :-)

Tom

Pat L (pgogy) schreef op 8-5-2014 10:52:
> depends on where the script is loaded from? I have http js calling 
> https APIs without much trouble
>
>
> On Thu, May 8, 2014 at 9:34 AM, Smith, John <J.J.Smith at gcu.ac.uk 
> <mailto:J.J.Smith at gcu.ac.uk>> wrote:
>
>     It might still have security implications on latest browsers where
>     XOT sits on http and access Flickr on https... similar to recent
>     Youtube etc problems...
>
>     Regards,
>
>     John Smith | Learning Technologist
>     Room A251, Govan Mbeki Building | School of Health & Life Sciences
>     | Glasgow Caledonian University
>     Cowcaddens Road | Glasgow | G4 0BA
>
>     Please address ALL support requests to hlsblt at gcu.ac.uk
>     <mailto:hlsblt at gcu.ac.uk> where one of the School Learning
>     Technologists will pick up the job. This will ensure that all jobs
>     are completed as promptly as possible.
>     ________________________________________
>     From: xerte-dev-bounces at lists.nottingham.ac.uk
>     <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
>     [xerte-dev-bounces at lists.nottingham.ac.uk
>     <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of
>     Julian Tenney [Julian.Tenney at nottingham.ac.uk
>     <mailto:Julian.Tenney at nottingham.ac.uk>]
>     Sent: 08 May 2014 09:26
>     To: For Xerte technical developers
>     Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
>
>     Yep, seems to work on our server,
>
>     J
>
>     From: Tenney Julian
>     Sent: 08 May 2014 08:59
>     To: 'For Xerte technical developers'
>     Subject: RE: Flickr API Going SSL-Only on June 27th, 2014
>
>     It appears as simple as changing the url to https from http. I've
>     committed those changes, but not tested,
>
>     From: xerte-dev-bounces at lists.nottingham.ac.uk
>     <mailto:xerte-dev-bounces at lists.nottingham.ac.uk><mailto:xerte-dev-bounces at lists.nottingham.ac.uk
>     <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>>
>     [mailto:xerte-dev-bounces at lists.nottingham.ac.uk
>     <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of
>     Julian Tenney
>     Sent: 08 May 2014 08:45
>     To: For Xerte technical developers
>     (xerte-dev at lists.nottingham.ac.uk
>     <mailto:xerte-dev at lists.nottingham.ac.uk><mailto:xerte-dev at lists.nottingham.ac.uk
>     <mailto:xerte-dev at lists.nottingham.ac.uk>>)
>     Subject: [Xerte-dev] FW: Flickr API Going SSL-Only on June 27th, 2014
>
>     This is going to have an impact I think. I'll investigate.
>
>     From: Flickr Mail [mailto:no-reply-flickrmail at flickr.com
>     <mailto:no-reply-flickrmail at flickr.com>]
>     Sent: 07 May 2014 18:19
>     To: julian.tenney at nottingham.ac.uk
>     <mailto:julian.tenney at nottingham.ac.uk><mailto:julian.tenney at nottingham.ac.uk
>     <mailto:julian.tenney at nottingham.ac.uk>>
>     Subject: Flickr API Going SSL-Only on June 27th, 2014
>
>     We see that you own an active API key for the Flickr API, and
>     wanted make sure you know about some upcoming changes to the API
>     related to SSL.
>
>     As you may have read on the Flickr code
>     blog<http://code.flickr.net/2014/04/30/flickr-api-going-ssl-only-on-june-27th-2014/>,
>     we've updated all of the API endpoints to fully support SSL. You
>     no longer need to use a special subdomain to access the API
>     securely, and we want to make sure that all API communication
>     adopts SSL.
>
>     Many of you have moved over to using SSL already, which is great!
>     If you haven't, it shouldn't be much more work than updating your
>     apps to call the API at: https://api.flickr.com/
>
>     We'll be updating the blog with information as we get close to the
>     switchover date, but here's the most important info:
>
>     New API keys will be issued for HTTPS access only: as of 6 May 2014
>     First blackout test window: 3 June 2014, 10:00-12:00 Pacific
>     Daylight Time (PDT)
>     Second blackout test window: 17 June 2014, 18:00-20:00 (PDT)
>     Non-SSL API deprecated: 27 June 2014, 10:00 (PDT)
>
>     During the blackout tests and after the 27th of June, calls to the
>     Flickr API that are not over SSL will fail with an HTTP error code
>     403. We will also return an error message with (Flickr) code 95
>     and msg "SSL is required".
>
>     Here are example error responses:
>
>     XML (REST):
>
>     <?xml version="1.0" encoding="utf-8" ?> <rsp stat="fail"> <err
>     code="95" msg="SSL is required" /> </rsp>
>
>     JSONP:
>
>     jsonFlickrApi({ "stat": "fail", "code": "95", "message": "SSL is
>     required" })
>
>     JSON:
>
>     { "stat": "fail", "code": "95", "message": "SSL is required" }
>
>     We realize that this change might be more difficult for some. We
>     will follow the Developer Support
>     Group<https://groups.yahoo.com/neo/groups/yws-flickr/info>
>     closely, so please let us hear your questions.
>
>     If you have a question about the API transition to SSL that you
>     feel is unique to your situation or is more appropriate to handle
>     via conversation directly with Flickr, send an email to
>     flickr-api-help at yahoo-inc.com
>     <mailto:flickr-api-help at yahoo-inc.com><mailto:flickr-api-help at yahoo-inc.com
>     <mailto:flickr-api-help at yahoo-inc.com>>. We will monitor that
>     mailbox through July 2014.
>
>     Thank you for being a member of our API community,
>     The Flickreenos
>
>     Glasgow Caledonian University is a registered Scottish charity,
>     number SC021474
>
>     _______________________________________________
>     Xerte-dev mailing list
>     Xerte-dev at lists.nottingham.ac.uk
>     <mailto:Xerte-dev at lists.nottingham.ac.uk>
>     http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>     This message and any attachment are intended solely for the
>     addressee and may contain confidential information. If you have
>     received this message in error, please send it back to me, and
>     immediately delete it.   Please do not use, copy or disclose the
>     information contained in this message or in any attachment.  Any
>     views or opinions expressed by the author of this email do not
>     necessarily reflect the views of the University of Nottingham.
>
>     This message has been checked for viruses but the contents of an
>     attachment
>     may still contain software viruses which could damage your
>     computer system, you are advised to perform your own checks. Email
>     communications with the University of Nottingham may be monitored
>     as permitted by UK legislation.
>
>
>
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

-- 
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140508/67196565/attachment.html>

From J.J.Smith at gcu.ac.uk  Thu May  8 11:37:49 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Thu, 8 May 2014 11:37:49 +0100
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014
In-Reply-To: <536B5BE2.5010101@tor.nl>
References: <9d6d4e22-8bc1-487e-b910-5cb20330dd04@UIWEXHUB02.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CED096986@EXCHANGE1.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0CED0969CF@EXCHANGE1.ad.nottingham.ac.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC101E6@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<CAO2PinMpMm_k-CMk0JA04oRRys8Vos-=z05C98XETjt5XHT1Vw@mail.gmail.com>,
	<536B5BE2.5010101@tor.nl>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC101ED@ITSEMBXCLUS.enterprise.gcal.ac.uk>

In some cases it can also depend on what comes down from the api but since it is just JSON data this time that is parsed then it should be fine...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Tom Reijnders [reijnders at tor.nl]
Sent: 08 May 2014 11:26
To: xerte at pgogywebstuff.com; For Xerte technical developers
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014

As far as I know, only the other way around causes problems, i.e. calling http from https.

In that sense this would be a good change  :-)

Tom

Pat L (pgogy) schreef op 8-5-2014 10:52:
depends on where the script is loaded from? I have http js calling https APIs without much trouble


On Thu, May 8, 2014 at 9:34 AM, Smith, John <J.J.Smith at gcu.ac.uk<mailto:J.J.Smith at gcu.ac.uk>> wrote:
It might still have security implications on latest browsers where XOT sits on http and access Flickr on https... similar to recent Youtube etc problems...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk<mailto:hlsblt at gcu.ac.uk> where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>]
Sent: 08 May 2014 09:26
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Flickr API Going SSL-Only on June 27th, 2014

Yep, seems to work on our server,

J

From: Tenney Julian
Sent: 08 May 2014 08:59
To: 'For Xerte technical developers'
Subject: RE: Flickr API Going SSL-Only on June 27th, 2014

It appears as simple as changing the url to https from http. I?ve committed those changes, but not tested,

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk><mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of Julian Tenney
Sent: 08 May 2014 08:45
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk><mailto:xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>>)
Subject: [Xerte-dev] FW: Flickr API Going SSL-Only on June 27th, 2014

This is going to have an impact I think. I?ll investigate.

From: Flickr Mail [mailto:no-reply-flickrmail at flickr.com<mailto:no-reply-flickrmail at flickr.com>]
Sent: 07 May 2014 18:19
To: julian.tenney at nottingham.ac.uk<mailto:julian.tenney at nottingham.ac.uk><mailto:julian.tenney at nottingham.ac.uk<mailto:julian.tenney at nottingham.ac.uk>>
Subject: Flickr API Going SSL-Only on June 27th, 2014

We see that you own an active API key for the Flickr API, and wanted make sure you know about some upcoming changes to the API related to SSL.

As you may have read on the Flickr code blog<http://code.flickr.net/2014/04/30/flickr-api-going-ssl-only-on-june-27th-2014/>, we've updated all of the API endpoints to fully support SSL. You no longer need to use a special subdomain to access the API securely, and we want to make sure that all API communication adopts SSL.

Many of you have moved over to using SSL already, which is great! If you haven't, it shouldn't be much more work than updating your apps to call the API at: https://api.flickr.com/

We'll be updating the blog with information as we get close to the switchover date, but here's the most important info:

New API keys will be issued for HTTPS access only: as of 6 May 2014
First blackout test window: 3 June 2014, 10:00-12:00 Pacific Daylight Time (PDT)
Second blackout test window: 17 June 2014, 18:00-20:00 (PDT)
Non-SSL API deprecated: 27 June 2014, 10:00 (PDT)

During the blackout tests and after the 27th of June, calls to the Flickr API that are not over SSL will fail with an HTTP error code 403. We will also return an error message with (Flickr) code 95 and msg "SSL is required".

Here are example error responses:

XML (REST):

<?xml version="1.0" encoding="utf-8" ?> <rsp stat="fail"> <err code="95" msg="SSL is required" /> </rsp>

JSONP:

jsonFlickrApi({ "stat": "fail", "code": "95", "message": "SSL is required" })

JSON:

{ "stat": "fail", "code": "95", "message": "SSL is required" }

We realize that this change might be more difficult for some. We will follow the Developer Support Group<https://groups.yahoo.com/neo/groups/yws-flickr/info> closely, so please let us hear your questions.

If you have a question about the API transition to SSL that you feel is unique to your situation or is more appropriate to handle via conversation directly with Flickr, send an email to flickr-api-help at yahoo-inc.com<mailto:flickr-api-help at yahoo-inc.com><mailto:flickr-api-help at yahoo-inc.com<mailto:flickr-api-help at yahoo-inc.com>>. We will monitor that mailbox through July 2014.

Thank you for being a member of our API community,
The Flickreenos

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.








_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev



--
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196




This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number SC021474


From Julian.Tenney at nottingham.ac.uk  Mon May 12 12:08:26 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 12:08:26 +0100
Subject: [Xerte-dev] FW: Further Xerte Data Loss
In-Reply-To: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>

I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters



<cMen



i.e. an incomplete tag, which I believe should be <cMenuConnector params="etc">



Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don't think the menu connectors are used much, which is why we haven't come across it before?



I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below...



Thanks a lot,



Julian





I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).



I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.



Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?



I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.



My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.



With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/17e80f26/attachment.html>

From d_b_burnett at hotmail.com  Mon May 12 12:14:43 2014
From: d_b_burnett at hotmail.com (Dave Burnett)
Date: Mon, 12 May 2014 07:14:43 -0400
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>


>after I included a further sub-menu connector
I take it the connectors have names?Wonder if apostrophes are escaped.

From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:08:26 +0100
Subject: [Xerte-dev] FW: Further Xerte Data Loss

I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters <cMen i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?> Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before? I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below? Thanks a lot, Julian   I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).  I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem. Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?  I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu. My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.  With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/edebb84c/attachment.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 12:20:39 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 12:20:39 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>

Apostrophes should be OK in either CDATA or in an xml attribute, yes?

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:15
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss


>after I included a further sub-menu connector

I take it the connectors have names?
Wonder if apostrophes are escaped.


________________________________
From: Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>
To: xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>
Date: Mon, 12 May 2014 12:08:26 +0100
Subject: [Xerte-dev] FW: Further Xerte Data Loss
I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters

<cMen

i.e. an incomplete tag, which I believe should be <cMenuConnector params="etc">

Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don't think the menu connectors are used much, which is why we haven't come across it before?

I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below...

Thanks a lot,

Julian


I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).

I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.

Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?

I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.

My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.

With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/4fbb1c57/attachment-0001.html>

From d_b_burnett at hotmail.com  Mon May 12 12:33:33 2014
From: d_b_burnett at hotmail.com (Dave Burnett)
Date: Mon, 12 May 2014 07:33:33 -0400
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <BLU177-W422D21842359208837B91DA7350@phx.gbl>

Not positive, but I thought attributes were picky.
From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:20:39 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Apostrophes should be OK in either CDATA or in an xml attribute, yes? From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:15
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss  >after I included a further sub-menu connector I take it the connectors have names?Wonder if apostrophes are escaped.  From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:08:26 +0100
Subject: [Xerte-dev] FW: Further Xerte Data LossI guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters <cMen i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?> Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before? I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below? Thanks a lot, Julian   I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).  I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem. Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?  I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu. My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.  With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/45b10ae8/attachment.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 12:35:05 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 12:35:05 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <BLU177-W422D21842359208837B91DA7350@phx.gbl>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>

They are about things like < > ", but I would have thought ' would be OK?

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:34
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Not positive, but I thought attributes were picky.

________________________________
From: Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>
To: xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>
Date: Mon, 12 May 2014 12:20:39 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
Apostrophes should be OK in either CDATA or in an xml attribute, yes?

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:15
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss


>after I included a further sub-menu connector

I take it the connectors have names?
Wonder if apostrophes are escaped.


________________________________
From: Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>
To: xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>
Date: Mon, 12 May 2014 12:08:26 +0100
Subject: [Xerte-dev] FW: Further Xerte Data Loss
I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters

<cMen

i.e. an incomplete tag, which I believe should be <cMenuConnector params="etc">

Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don't think the menu connectors are used much, which is why we haven't come across it before?

I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below...

Thanks a lot,

Julian


I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).

I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.

Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?

I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.

My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.

With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/b70afe04/attachment-0001.html>

From david at palepurple.co.uk  Mon May 12 12:36:57 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Mon, 12 May 2014 12:36:57 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <5370B259.3060606@palepurple.co.uk>

Is it possible that it's a unicode ` character?


I've seen a number of UTF-8 related issues with the data.xml etc files.

David.


On 12/05/14 12:35, Julian Tenney wrote:
> They are about things like < > ?, but I would have thought ? would be OK?
>
> *From:*xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] *On Behalf Of *Dave
> Burnett
> *Sent:* 12 May 2014 12:34
> *To:* For Xerte technical developers
> *Subject:* [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Not positive, but I thought attributes were picky.
>
> ------------------------------------------------------------------------
>
> From: Julian.Tenney at nottingham.ac.uk <mailto:Julian.Tenney at nottingham.ac.uk>
> To: xerte-dev at lists.nottingham.ac.uk
> <mailto:xerte-dev at lists.nottingham.ac.uk>
> Date: Mon, 12 May 2014 12:20:39 +0100
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Apostrophes should be OK in either CDATA or in an xml attribute, yes?
>
> *From:*xerte-dev-bounces at lists.nottingham.ac.uk
> <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] *On Behalf Of *Dave
> Burnett
> *Sent:* 12 May 2014 12:15
> *To:* For Xerte technical developers
> *Subject:* [Xerte-dev] Re: FW: Further Xerte Data Loss
>
>>after I included a further sub-menu connector
>
> I take it the connectors have names?
>
> Wonder if apostrophes are escaped.
>
> ------------------------------------------------------------------------
>
> From: Julian.Tenney at nottingham.ac.uk <mailto:Julian.Tenney at nottingham.ac.uk>
> To: xerte-dev at lists.nottingham.ac.uk
> <mailto:xerte-dev at lists.nottingham.ac.uk>
> Date: Mon, 12 May 2014 12:08:26 +0100
> Subject: [Xerte-dev] FW: Further Xerte Data Loss
>
> I guys, I need to get to the bottom of this issue: I have a fairly
> advanced user who has been periodically losing data: the email below
> explains the problem. A couple of weeks ago it happened and we lost 40
> slides, now it has happened again. In the file I looked at, the broken
> file ended with the characters
>
> <cMen
>
> i.e. an incomplete tag, which I believe should be <cMenuConnector
> params=?etc?>
>
> Does this ring any bells with anyone? Anyone experienced it before
> elsewhere? I don?t think the menu connectors are used much, which is why
> we haven?t come across it before?
>
> I have a feeling it is something to do with escaping characters as data
> is being saved, or as it is being read back to the editor. Anyone have
> any other bright ideas how we can troubleshoot it? See below?
>
> Thanks a lot,
>
> Julian
>
> I've experienced further data loss again this morning, this time on a
> completely different Xerte project
> (http://www.nottingham.ac.uk/toolkits/play_10189).
>
> I initially produced this resource using the Flash Xerte engine, but the
> academic I'm working with was keen for me to transfer the data over into
> HTML5. I made a direct duplicate copy of the Flash project, ticked the
> HTML5 engine and then set about reproducing the slides - no problem.
>
> Everything was absolutely fine (after working on the project for most of
> Friday) until around 10:30am this morning, when I lost around 10-15
> slides (including my review quiz with 20 MCQs). I'm now beginning to
> strongly suspect that the Xerte Menu Connector templates might be having
> something to do with this persistent problem?
>
> I have been using multiple connector menus in both my Grasslands and
> Meat Science packages. In both I have a dominant 'Main Menu' at the
> start, which then 'connects' via tabs to several sub-menus (of the same
> Connector template type) dotted throughout the resource. In each of the
> sub-menus I link back to the Main Menu.
>
> My last actions before the slides were lost today, happened after I
> included a further sub-menu connector (at around slide 70) and then
> linked this to this from the 'Main Menu' (slide 2) via the drop-down
> selector menu. I then published the changes.
>
> With that being said, I've attempted to recreate the problem using a
> dummy file, but I have so far failed to do it.
>
>
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
> message in error, please send it back to me, and immediately delete
> it.   Please do not use, copy or disclose the information contained in
> this message or in any attachment.  Any views or opinions expressed by
> the author of this email do not necessarily reflect the views of the
> University of Nottingham.
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
> _______________________________________________ Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> <mailto:Xerte-dev at lists.nottingham.ac.uk>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
> message in error, please send it back to me, and immediately delete
> it.   Please do not use, copy or disclose the information contained in
> this message or in any attachment.  Any views or opinions expressed by
> the author of this email do not necessarily reflect the views of the
> University of Nottingham.
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
> _______________________________________________ Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> <mailto:Xerte-dev at lists.nottingham.ac.uk>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
> message in error, please send it back to me, and immediately delete
> it.   Please do not use, copy or disclose the information contained in
> this message or in any attachment.  Any views or opinions expressed by
> the author of this email do not necessarily reflect the views of the
> University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From d_b_burnett at hotmail.com  Mon May 12 12:37:39 2014
From: d_b_burnett at hotmail.com (Dave Burnett)
Date: Mon, 12 May 2014 07:37:39 -0400
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	, , ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	, <BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>, ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <BLU177-W426EAB9CF8E100F651A7E9A7350@phx.gbl>


If the attribute is enclosed in double quotes.
From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:35:05 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

They are about things like < > ?, but I would have thought ? would be OK? From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:34
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss Not positive, but I thought attributes were picky. From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:20:39 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data LossApostrophes should be OK in either CDATA or in an xml attribute, yes? From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:15
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss  >after I included a further sub-menu connector I take it the connectors have names?Wonder if apostrophes are escaped.  From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:08:26 +0100
Subject: [Xerte-dev] FW: Further Xerte Data LossI guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters <cMen i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?> Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before? I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below? Thanks a lot, Julian   I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).  I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem. Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?  I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu. My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.  With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/455e8024/attachment-0001.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 12:38:16 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 12:38:16 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <BLU177-W426EAB9CF8E100F651A7E9A7350@phx.gbl>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	, , ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	, <BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>, ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W426EAB9CF8E100F651A7E9A7350@phx.gbl>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901ABB@EXCHANGE1.ad.nottingham.ac.uk>

It would be.

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:38
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss


If the attribute is enclosed in double quotes.

________________________________
From: Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>
To: xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>
Date: Mon, 12 May 2014 12:35:05 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
They are about things like < > ", but I would have thought ' would be OK?

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:34
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Not positive, but I thought attributes were picky.

________________________________
From: Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>
To: xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>
Date: Mon, 12 May 2014 12:20:39 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
Apostrophes should be OK in either CDATA or in an xml attribute, yes?

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:15
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss


>after I included a further sub-menu connector

I take it the connectors have names?
Wonder if apostrophes are escaped.


________________________________
From: Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>
To: xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>
Date: Mon, 12 May 2014 12:08:26 +0100
Subject: [Xerte-dev] FW: Further Xerte Data Loss
I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters

<cMen

i.e. an incomplete tag, which I believe should be <cMenuConnector params="etc">

Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don't think the menu connectors are used much, which is why we haven't come across it before?

I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below...

Thanks a lot,

Julian


I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).

I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.

Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?

I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.

My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.

With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/58570430/attachment-0001.html>

From d_b_burnett at hotmail.com  Mon May 12 12:39:01 2014
From: d_b_burnett at hotmail.com (Dave Burnett)
Date: Mon, 12 May 2014 07:39:01 -0400
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901ABB@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	, , , , ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	, , <BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>, , ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>,
	, <BLU177-W422D21842359208837B91DA7350@phx.gbl>, ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W426EAB9CF8E100F651A7E9A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901ABB@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <BLU177-W425AFCA74F1F72D95D92CFA7350@phx.gbl>

Ok.

From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:38:16 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

It would be. From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:38
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss  If the attribute is enclosed in double quotes. From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:35:05 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data LossThey are about things like < > ?, but I would have thought ? would be OK? From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:34
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss Not positive, but I thought attributes were picky. From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:20:39 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data LossApostrophes should be OK in either CDATA or in an xml attribute, yes? From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 12:15
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss  >after I included a further sub-menu connector I take it the connectors have names?Wonder if apostrophes are escaped.  From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 12:08:26 +0100
Subject: [Xerte-dev] FW: Further Xerte Data LossI guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters <cMen i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?> Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before? I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below? Thanks a lot, Julian   I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).  I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem. Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?  I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu. My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.  With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/f5b8b86b/attachment-0001.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 12:41:31 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 12:41:31 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <5370B259.3060606@palepurple.co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>

What beats me is why the file should break halfway through a tag name.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 12:37
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Is it possible that it's a unicode ` character?


I've seen a number of UTF-8 related issues with the data.xml etc files.

David.


On 12/05/14 12:35, Julian Tenney wrote:
> They are about things like < > ", but I would have thought ' would be OK?
>
> *From:*xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] *On Behalf Of *Dave 
> Burnett
> *Sent:* 12 May 2014 12:34
> *To:* For Xerte technical developers
> *Subject:* [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Not positive, but I thought attributes were picky.
>
> ----------------------------------------------------------------------
> --
>
> From: Julian.Tenney at nottingham.ac.uk 
> <mailto:Julian.Tenney at nottingham.ac.uk>
> To: xerte-dev at lists.nottingham.ac.uk
> <mailto:xerte-dev at lists.nottingham.ac.uk>
> Date: Mon, 12 May 2014 12:20:39 +0100
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Apostrophes should be OK in either CDATA or in an xml attribute, yes?
>
> *From:*xerte-dev-bounces at lists.nottingham.ac.uk
> <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] *On Behalf Of *Dave 
> Burnett
> *Sent:* 12 May 2014 12:15
> *To:* For Xerte technical developers
> *Subject:* [Xerte-dev] Re: FW: Further Xerte Data Loss
>
>>after I included a further sub-menu connector
>
> I take it the connectors have names?
>
> Wonder if apostrophes are escaped.
>
> ----------------------------------------------------------------------
> --
>
> From: Julian.Tenney at nottingham.ac.uk 
> <mailto:Julian.Tenney at nottingham.ac.uk>
> To: xerte-dev at lists.nottingham.ac.uk
> <mailto:xerte-dev at lists.nottingham.ac.uk>
> Date: Mon, 12 May 2014 12:08:26 +0100
> Subject: [Xerte-dev] FW: Further Xerte Data Loss
>
> I guys, I need to get to the bottom of this issue: I have a fairly 
> advanced user who has been periodically losing data: the email below 
> explains the problem. A couple of weeks ago it happened and we lost 40 
> slides, now it has happened again. In the file I looked at, the broken 
> file ended with the characters
>
> <cMen
>
> i.e. an incomplete tag, which I believe should be <cMenuConnector 
> params="etc">
>
> Does this ring any bells with anyone? Anyone experienced it before 
> elsewhere? I don't think the menu connectors are used much, which is 
> why we haven't come across it before?
>
> I have a feeling it is something to do with escaping characters as 
> data is being saved, or as it is being read back to the editor. Anyone 
> have any other bright ideas how we can troubleshoot it? See below...
>
> Thanks a lot,
>
> Julian
>
> I've experienced further data loss again this morning, this time on a 
> completely different Xerte project 
> (http://www.nottingham.ac.uk/toolkits/play_10189).
>
> I initially produced this resource using the Flash Xerte engine, but 
> the academic I'm working with was keen for me to transfer the data 
> over into HTML5. I made a direct duplicate copy of the Flash project, 
> ticked the
> HTML5 engine and then set about reproducing the slides - no problem.
>
> Everything was absolutely fine (after working on the project for most 
> of
> Friday) until around 10:30am this morning, when I lost around 10-15 
> slides (including my review quiz with 20 MCQs). I'm now beginning to 
> strongly suspect that the Xerte Menu Connector templates might be 
> having something to do with this persistent problem?
>
> I have been using multiple connector menus in both my Grasslands and 
> Meat Science packages. In both I have a dominant 'Main Menu' at the 
> start, which then 'connects' via tabs to several sub-menus (of the 
> same Connector template type) dotted throughout the resource. In each 
> of the sub-menus I link back to the Main Menu.
>
> My last actions before the slides were lost today, happened after I 
> included a further sub-menu connector (at around slide 70) and then 
> linked this to this from the 'Main Menu' (slide 2) via the drop-down 
> selector menu. I then published the changes.
>
> With that being said, I've attempted to recreate the problem using a 
> dummy file, but I have so far failed to do it.
>
>
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this 
> message in error, please send it back to me, and immediately delete
> it.   Please do not use, copy or disclose the information contained in
> this message or in any attachment.  Any views or opinions expressed by 
> the author of this email do not necessarily reflect the views of the 
> University of Nottingham.
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as 
> permitted by UK legislation.
>
>
> _______________________________________________ Xerte-dev mailing list 
> Xerte-dev at lists.nottingham.ac.uk 
> <mailto:Xerte-dev at lists.nottingham.ac.uk>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this 
> message in error, please send it back to me, and immediately delete
> it.   Please do not use, copy or disclose the information contained in
> this message or in any attachment.  Any views or opinions expressed by 
> the author of this email do not necessarily reflect the views of the 
> University of Nottingham.
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as 
> permitted by UK legislation.
>
>
> _______________________________________________ Xerte-dev mailing list 
> Xerte-dev at lists.nottingham.ac.uk 
> <mailto:Xerte-dev at lists.nottingham.ac.uk>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this 
> message in error, please send it back to me, and immediately delete
> it.   Please do not use, copy or disclose the information contained in
> this message or in any attachment.  Any views or opinions expressed by 
> the author of this email do not necessarily reflect the views of the 
> University of Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as 
> permitted by UK legislation.
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From J.J.Smith at gcu.ac.uk  Mon May 12 12:50:37 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Mon, 12 May 2014 12:50:37 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC10206@ITSEMBXCLUS.enterprise.gcal.ac.uk>

I'd investigate whether this line is actually doing what it is supposed to...

https://github.com/thexerteproject/xerteonlinetoolkits/blob/develop/modules/xerte/engine/save.php#L36

Though not sure whether a break in the upload would still allow save.php to complete...

We could try to rebuild the XML in save and validate that it is ok? and ends with </learningObject>

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 12 May 2014 12:41
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

What beats me is why the file should break halfway through a tag name.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 12:37
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Is it possible that it's a unicode ` character?


I've seen a number of UTF-8 related issues with the data.xml etc files.

David.


On 12/05/14 12:35, Julian Tenney wrote:
> They are about things like < > ", but I would have thought ' would be OK?
>
> *From:*xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] *On Behalf Of *Dave
> Burnett
> *Sent:* 12 May 2014 12:34
> *To:* For Xerte technical developers
> *Subject:* [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Not positive, but I thought attributes were picky.
>
> ----------------------------------------------------------------------
> --
>
> From: Julian.Tenney at nottingham.ac.uk
> <mailto:Julian.Tenney at nottingham.ac.uk>
> To: xerte-dev at lists.nottingham.ac.uk
> <mailto:xerte-dev at lists.nottingham.ac.uk>
> Date: Mon, 12 May 2014 12:20:39 +0100
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Apostrophes should be OK in either CDATA or in an xml attribute, yes?
>
> *From:*xerte-dev-bounces at lists.nottingham.ac.uk
> <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] *On Behalf Of *Dave
> Burnett
> *Sent:* 12 May 2014 12:15
> *To:* For Xerte technical developers
> *Subject:* [Xerte-dev] Re: FW: Further Xerte Data Loss
>
>>after I included a further sub-menu connector
>
> I take it the connectors have names?
>
> Wonder if apostrophes are escaped.
>
> ----------------------------------------------------------------------
> --
>
> From: Julian.Tenney at nottingham.ac.uk
> <mailto:Julian.Tenney at nottingham.ac.uk>
> To: xerte-dev at lists.nottingham.ac.uk
> <mailto:xerte-dev at lists.nottingham.ac.uk>
> Date: Mon, 12 May 2014 12:08:26 +0100
> Subject: [Xerte-dev] FW: Further Xerte Data Loss
>
> I guys, I need to get to the bottom of this issue: I have a fairly
> advanced user who has been periodically losing data: the email below
> explains the problem. A couple of weeks ago it happened and we lost 40
> slides, now it has happened again. In the file I looked at, the broken
> file ended with the characters
>
> <cMen
>
> i.e. an incomplete tag, which I believe should be <cMenuConnector
> params="etc">
>
> Does this ring any bells with anyone? Anyone experienced it before
> elsewhere? I don't think the menu connectors are used much, which is
> why we haven't come across it before?
>
> I have a feeling it is something to do with escaping characters as
> data is being saved, or as it is being read back to the editor. Anyone
> have any other bright ideas how we can troubleshoot it? See below...
>
> Thanks a lot,
>
> Julian
>
> I've experienced further data loss again this morning, this time on a
> completely different Xerte project
> (http://www.nottingham.ac.uk/toolkits/play_10189).
>
> I initially produced this resource using the Flash Xerte engine, but
> the academic I'm working with was keen for me to transfer the data
> over into HTML5. I made a direct duplicate copy of the Flash project,
> ticked the
> HTML5 engine and then set about reproducing the slides - no problem.
>
> Everything was absolutely fine (after working on the project for most
> of
> Friday) until around 10:30am this morning, when I lost around 10-15
> slides (including my review quiz with 20 MCQs). I'm now beginning to
> strongly suspect that the Xerte Menu Connector templates might be
> having something to do with this persistent problem?
>
> I have been using multiple connector menus in both my Grasslands and
> Meat Science packages. In both I have a dominant 'Main Menu' at the
> start, which then 'connects' via tabs to several sub-menus (of the
> same Connector template type) dotted throughout the resource. In each
> of the sub-menus I link back to the Main Menu.
>
> My last actions before the slides were lost today, happened after I
> included a further sub-menu connector (at around slide 70) and then
> linked this to this from the 'Main Menu' (slide 2) via the drop-down
> selector menu. I then published the changes.
>
> With that being said, I've attempted to recreate the problem using a
> dummy file, but I have so far failed to do it.
>
>
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
> message in error, please send it back to me, and immediately delete
> it.   Please do not use, copy or disclose the information contained in
> this message or in any attachment.  Any views or opinions expressed by
> the author of this email do not necessarily reflect the views of the
> University of Nottingham.
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
> _______________________________________________ Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> <mailto:Xerte-dev at lists.nottingham.ac.uk>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
> message in error, please send it back to me, and immediately delete
> it.   Please do not use, copy or disclose the information contained in
> this message or in any attachment.  Any views or opinions expressed by
> the author of this email do not necessarily reflect the views of the
> University of Nottingham.
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
> _______________________________________________ Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> <mailto:Xerte-dev at lists.nottingham.ac.uk>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
> This message and any attachment are intended solely for the addressee
> and may contain confidential information. If you have received this
> message in error, please send it back to me, and immediately delete
> it.   Please do not use, copy or disclose the information contained in
> this message or in any attachment.  Any views or opinions expressed by
> the author of this email do not necessarily reflect the views of the
> University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474


From david at palepurple.co.uk  Mon May 12 12:53:05 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Mon, 12 May 2014 12:53:05 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>	<BLU177-W422D21842359208837B91DA7350@phx.gbl>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <5370B621.4020601@palepurple.co.uk>



On 12/05/14 12:41, Julian Tenney wrote:
> What beats me is why the file should break halfway through a tag name.


What I've seen is that the PHP (or whatever) can't read the XML file as 
it has the wrong encoding specified, or it just somehow contains 
characters that are invalid.

see e.g.

http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-character-error

https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/


David.

>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
> Sent: 12 May 2014 12:37
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Is it possible that it's a unicode ` character?
>
>
> I've seen a number of UTF-8 related issues with the data.xml etc files.
>
> David.
>



-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From J.J.Smith at gcu.ac.uk  Mon May 12 13:00:16 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Mon, 12 May 2014 13:00:16 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <5370B621.4020601@palepurple.co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>,
	<5370B621.4020601@palepurple.co.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC10207@ITSEMBXCLUS.enterprise.gcal.ac.uk>

We don't actually try to build the XML which may be the issue. We just take the stream from Flash and write to the disk. Either Flash is sending truncated data (and the wrong size to match) or it is being truncated when received by save.php...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin [david at palepurple.co.uk]
Sent: 12 May 2014 12:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

On 12/05/14 12:41, Julian Tenney wrote:
> What beats me is why the file should break halfway through a tag name.


What I've seen is that the PHP (or whatever) can't read the XML file as
it has the wrong encoding specified, or it just somehow contains
characters that are invalid.

see e.g.

http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-character-error

https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/


David.

>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
> Sent: 12 May 2014 12:37
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Is it possible that it's a unicode ` character?
>
>
> I've seen a number of UTF-8 related issues with the data.xml etc files.
>
> David.
>



--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From Julian.Tenney at nottingham.ac.uk  Mon May 12 13:06:21 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 13:06:21 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC10207@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>,
	<5370B621.4020601@palepurple.co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC10207@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901B00@EXCHANGE1.ad.nottingham.ac.uk>

Implying invalid chars in the xml somehow?

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 12 May 2014 13:00
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

We don't actually try to build the XML which may be the issue. We just take the stream from Flash and write to the disk. Either Flash is sending truncated data (and the wrong size to match) or it is being truncated when received by save.php...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin [david at palepurple.co.uk]
Sent: 12 May 2014 12:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

On 12/05/14 12:41, Julian Tenney wrote:
> What beats me is why the file should break halfway through a tag name.


What I've seen is that the PHP (or whatever) can't read the XML file as it has the wrong encoding specified, or it just somehow contains characters that are invalid.

see e.g.

http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-character-error

https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/


David.

>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk 
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David 
> Goodwin
> Sent: 12 May 2014 12:37
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Is it possible that it's a unicode ` character?
>
>
> I've seen a number of UTF-8 related issues with the data.xml etc files.
>
> David.
>



--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From Julian.Tenney at nottingham.ac.uk  Mon May 12 13:16:45 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 13:16:45 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <5370B621.4020601@palepurple.co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B621.4020601@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901B13@EXCHANGE1.ad.nottingham.ac.uk>

On this occasion, the file has broken half way through a CDATA section, so everything ought to have escaped OK.

Just to be certain: data.xml is the published file, right, and preview.xml is the editing file?

In which case it is the published file that has broken, and the editing file is complete with a <learningObject> tag. I'll attach them, I'm just asking if he minds if I share the files.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 12:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss



On 12/05/14 12:41, Julian Tenney wrote:
> What beats me is why the file should break halfway through a tag name.


What I've seen is that the PHP (or whatever) can't read the XML file as it has the wrong encoding specified, or it just somehow contains characters that are invalid.

see e.g.

http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-character-error

https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/


David.

>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk 
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David 
> Goodwin
> Sent: 12 May 2014 12:37
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Is it possible that it's a unicode ` character?
>
>
> I've seen a number of UTF-8 related issues with the data.xml etc files.
>
> David.
>



--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From Julian.Tenney at nottingham.ac.uk  Mon May 12 13:23:40 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 13:23:40 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <5370B621.4020601@palepurple.co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B621.4020601@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>

This looks suspicious: that's not escaping, right? That looks to me like errant characters creeping in from somewhere? Paste from word, etc?



[cid:image001.png at 01CF6DE5.63F6A2C0]



-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 12:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss







On 12/05/14 12:41, Julian Tenney wrote:

> What beats me is why the file should break halfway through a tag name.





What I've seen is that the PHP (or whatever) can't read the XML file as it has the wrong encoding specified, or it just somehow contains characters that are invalid.



see e.g.



http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-character-error



https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/





David.



>

> -----Original Message-----

> From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>

> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David

> Goodwin

> Sent: 12 May 2014 12:37

> To: For Xerte technical developers

> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

>

> Is it possible that it's a unicode ` character?

>

>

> I've seen a number of UTF-8 related issues with the data.xml etc files.

>

> David.

>







--

Pale Purple Ltd



PHP Web application development and support



http://www.palepurple.co.uk

@PalePurpleLtd

07792 380669 / 0845 0046746



_______________________________________________

Xerte-dev mailing list

Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>

http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/4a6f497a/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 41424 bytes
Desc: image001.png
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/4a6f497a/attachment-0001.png>

From d_b_burnett at hotmail.com  Mon May 12 14:14:49 2014
From: d_b_burnett at hotmail.com (Dave Burnett)
Date: Mon, 12 May 2014 09:14:49 -0400
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	, , ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	, <BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>, ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>,
	<5370B259.3060606@palepurple.co.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>,
	<5370B621.4020601@palepurple.co.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <BLU177-W5046FE96F08B44A3F70E63A7350@phx.gbl>


It's a ? (RIGHT SINGLE QUOTATION MARK - U+2019) character which has been encoded as CP-1252 instead of UTF-8.
http://stackoverflow.com/questions/2477452/a-showing-on-page-instead-of
(and I assume the tilded version is the left single quote).
 

From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 13:23:40 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

This looks suspicious: that?s not escaping, right? That looks to me like errant characters creeping in from somewhere? Paste from word, etc?  -----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 12:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss   On 12/05/14 12:41, Julian Tenney wrote:> What beats me is why the file should break halfway through a tag name.  What I've seen is that the PHP (or whatever) can't read the XML file as it has the wrong encoding specified, or it just somehow contains characters that are invalid. see e.g. http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-character-error https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/  David. > > -----Original Message-----> From: xerte-dev-bounces at lists.nottingham.ac.uk > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David > Goodwin> Sent: 12 May 2014 12:37> To: For Xerte technical developers> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss> > Is it possible that it's a unicode ` character?> > > I've seen a number of UTF-8 related issues with the data.xml etc files.> > David.>    --Pale Purple Ltd PHP Web application development and support http://www.palepurple.co.uk at PalePurpleLtd07792 380669 / 0845 0046746 _______________________________________________Xerte-dev mailing listXerte-dev at lists.nottingham.ac.ukhttp://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/15693838/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 41424 bytes
Desc: not available
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/15693838/attachment-0001.png>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 14:18:21 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 14:18:21 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <BLU177-W5046FE96F08B44A3F70E63A7350@phx.gbl>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	, , ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	, <BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>, ,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>,
	<5370B259.3060606@palepurple.co.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>,
	<5370B621.4020601@palepurple.co.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W5046FE96F08B44A3F70E63A7350@phx.gbl>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901B9B@EXCHANGE1.ad.nottingham.ac.uk>

I'm pretty certain the problem is down to pasting in non UTF code.

So how best to make sure it doesn't get through to the PHP? Old Flash isn't so good at this sort of thing. I already have event handlers looking of r a large change in size of the text, to detect pastes. I can do something there in the editor to strip out non-UTF8 I think?

J

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Dave Burnett
Sent: 12 May 2014 14:15
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss


It's a ' (RIGHT SINGLE QUOTATION MARK<http://www.fileformat.info/info/unicode/char/2019/index.htm> - U+2019) character which has been encoded as CP-1252<http://en.wikipedia.org/wiki/Windows-1252> instead of UTF-8<http://en.wikipedia.org/wiki/UTF-8>.

http://stackoverflow.com/questions/2477452/a-showing-on-page-instead-of

(and I assume the tilded version is the left single quote).



________________________________
From: Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>
To: xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>
Date: Mon, 12 May 2014 13:23:40 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
This looks suspicious: that's not escaping, right? That looks to me like errant characters creeping in from somewhere? Paste from word, etc?

[cid:image001.png at 01CF6DED.076A2FB0]

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 12:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss



On 12/05/14 12:41, Julian Tenney wrote:
> What beats me is why the file should break halfway through a tag name.


What I've seen is that the PHP (or whatever) can't read the XML file as it has the wrong encoding specified, or it just somehow contains characters that are invalid.

see e.g.

http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-character-error

https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/


David.

>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David
> Goodwin
> Sent: 12 May 2014 12:37
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Is it possible that it's a unicode ` character?
>
>
> I've seen a number of UTF-8 related issues with the data.xml etc files.
>
> David.
>



--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________ Xerte-dev mailing list Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/95596b75/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 41424 bytes
Desc: image001.png
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/95596b75/attachment-0001.png>

From david at palepurple.co.uk  Mon May 12 14:34:58 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Mon, 12 May 2014 14:34:58 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>	<BLU177-W422D21842359208837B91DA7350@phx.gbl>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>	<5370B259.3060606@palepurple.co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>	<5370B621.4020601@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <5370CE02.7070006@palepurple.co.uk>

Can you attach / share the data.xml file somehow?

I know of a PHP library which is quite good at cleaning up such stuff ....

David.


On 12/05/14 13:23, Julian Tenney wrote:
> This looks suspicious: that?s not escaping, right? That looks to me like
> errant characters creeping in from somewhere? Paste from word, etc?
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
> Sent: 12 May 2014 12:53
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> On 12/05/14 12:41, Julian Tenney wrote:
>
>  > What beats me is why the file should break halfway through a tag name.
>
> What I've seen is that the PHP (or whatever) can't read the XML file as
> it has the wrong encoding specified, or it just somehow contains
> characters that are invalid.
>
> see e.g.
>
> http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-character-error
>
> https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/
>
> David.
>
>  >
>
>  > -----Original Message-----
>
>  > From: xerte-dev-bounces at lists.nottingham.ac.uk
> <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
>
>  > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David
>
>  > Goodwin
>
>  > Sent: 12 May 2014 12:37
>
>  > To: For Xerte technical developers
>
>  > Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
>  >
>
>  > Is it possible that it's a unicode ` character?
>
>  >
>
>  >
>
>  > I've seen a number of UTF-8 related issues with the data.xml etc files.
>
>  >
>
>  > David.
>
>  >
>
> --
>
> Pale Purple Ltd
>
> PHP Web application development and support
>
> http://www.palepurple.co.uk
>
> @PalePurpleLtd
>
> 07792 380669 / 0845 0046746
>
> _______________________________________________
>
> Xerte-dev mailing list
>
> Xerte-dev at lists.nottingham.ac.uk <mailto:Xerte-dev at lists.nottingham.ac.uk>
>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From Julian.Tenney at nottingham.ac.uk  Mon May 12 14:36:23 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 14:36:23 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <5370CE02.7070006@palepurple.co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B621.4020601@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
	<5370CE02.7070006@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901BCB@EXCHANGE1.ad.nottingham.ac.uk>

I asked if I could share the files, but not had a response: I'm sure it's OK, but I need to check,

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 14:35
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Can you attach / share the data.xml file somehow?

I know of a PHP library which is quite good at cleaning up such stuff ....

David.


On 12/05/14 13:23, Julian Tenney wrote:
> This looks suspicious: that's not escaping, right? That looks to me 
> like errant characters creeping in from somewhere? Paste from word, etc?
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David 
> Goodwin
> Sent: 12 May 2014 12:53
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> On 12/05/14 12:41, Julian Tenney wrote:
>
>  > What beats me is why the file should break halfway through a tag name.
>
> What I've seen is that the PHP (or whatever) can't read the XML file 
> as it has the wrong encoding specified, or it just somehow contains 
> characters that are invalid.
>
> see e.g.
>
> http://stackoverflow.com/questions/14463573/php-simplexml-load-file-in
> valid-character-error
>
> https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-non
> e-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/
>
> David.
>
>  >
>
>  > -----Original Message-----
>
>  > From: xerte-dev-bounces at lists.nottingham.ac.uk
> <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
>
>  > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of 
> David
>
>  > Goodwin
>
>  > Sent: 12 May 2014 12:37
>
>  > To: For Xerte technical developers
>
>  > Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
>  >
>
>  > Is it possible that it's a unicode ` character?
>
>  >
>
>  >
>
>  > I've seen a number of UTF-8 related issues with the data.xml etc files.
>
>  >
>
>  > David.
>
>  >
>
> --
>
> Pale Purple Ltd
>
> PHP Web application development and support
>
> http://www.palepurple.co.uk
>
> @PalePurpleLtd
>
> 07792 380669 / 0845 0046746
>
> _______________________________________________
>
> Xerte-dev mailing list
>
> Xerte-dev at lists.nottingham.ac.uk 
> <mailto:Xerte-dev at lists.nottingham.ac.uk>
>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From xerte at pgogywebstuff.com  Mon May 12 14:58:04 2014
From: xerte at pgogywebstuff.com (Pat L (pgogy))
Date: Mon, 12 May 2014 14:58:04 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901BCB@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B621.4020601@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
	<5370CE02.7070006@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901BCB@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <CAO2PinN9HxNp-3_HWQQ1cJE4r5gynUZxbc6dTSk0ajb=7Z_szQ@mail.gmail.com>

i thought there was a valid XML test before saving?


On Mon, May 12, 2014 at 2:36 PM, Julian Tenney <
Julian.Tenney at nottingham.ac.uk> wrote:

> I asked if I could share the files, but not had a response: I'm sure it's
> OK, but I need to check,
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:
> xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
> Sent: 12 May 2014 14:35
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Can you attach / share the data.xml file somehow?
>
> I know of a PHP library which is quite good at cleaning up such stuff ....
>
> David.
>
>
> On 12/05/14 13:23, Julian Tenney wrote:
> > This looks suspicious: that's not escaping, right? That looks to me
> > like errant characters creeping in from somewhere? Paste from word, etc?
> >
> > -----Original Message-----
> > From: xerte-dev-bounces at lists.nottingham.ac.uk
> > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David
> > Goodwin
> > Sent: 12 May 2014 12:53
> > To: For Xerte technical developers
> > Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
> >
> > On 12/05/14 12:41, Julian Tenney wrote:
> >
> >  > What beats me is why the file should break halfway through a tag name.
> >
> > What I've seen is that the PHP (or whatever) can't read the XML file
> > as it has the wrong encoding specified, or it just somehow contains
> > characters that are invalid.
> >
> > see e.g.
> >
> > http://stackoverflow.com/questions/14463573/php-simplexml-load-file-in
> > valid-character-error
> >
> > https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-non
> > e-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/
> >
> > David.
> >
> >  >
> >
> >  > -----Original Message-----
> >
> >  > From: xerte-dev-bounces at lists.nottingham.ac.uk
> > <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
> >
> >  > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of
> > David
> >
> >  > Goodwin
> >
> >  > Sent: 12 May 2014 12:37
> >
> >  > To: For Xerte technical developers
> >
> >  > Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
> >
> >  >
> >
> >  > Is it possible that it's a unicode ` character?
> >
> >  >
> >
> >  >
> >
> >  > I've seen a number of UTF-8 related issues with the data.xml etc
> files.
> >
> >  >
> >
> >  > David.
> >
> >  >
> >
> > --
> >
> > Pale Purple Ltd
> >
> > PHP Web application development and support
> >
> > http://www.palepurple.co.uk
> >
> > @PalePurpleLtd
> >
> > 07792 380669 / 0845 0046746
> >
> > _______________________________________________
> >
> > Xerte-dev mailing list
> >
> > Xerte-dev at lists.nottingham.ac.uk
> > <mailto:Xerte-dev at lists.nottingham.ac.uk>
> >
> > http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> >
> >
> >
> > _______________________________________________
> > Xerte-dev mailing list
> > Xerte-dev at lists.nottingham.ac.uk
> > http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> >
>
> --
> Pale Purple Ltd
>
> PHP Web application development and support
>
> http://www.palepurple.co.uk
> @PalePurpleLtd
> 07792 380669 / 0845 0046746
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an attachment
> may still contain software viruses which could damage your computer
> system, you are advised to perform your own checks. Email communications
> with the University of Nottingham may be monitored as permitted by UK
> legislation.
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/2440d6b5/attachment.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 15:01:09 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 15:01:09 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <CAO2PinN9HxNp-3_HWQQ1cJE4r5gynUZxbc6dTSk0ajb=7Z_szQ@mail.gmail.com>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B621.4020601@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
	<5370CE02.7070006@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901BCB@EXCHANGE1.ad.nottingham.ac.uk>
	<CAO2PinN9HxNp-3_HWQQ1cJE4r5gynUZxbc6dTSk0ajb=7Z_szQ@mail.gmail.com>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C16@EXCHANGE1.ad.nottingham.ac.uk>

The xml can still be valid and hold non UTF data, right? That seems to be the problem, and then Flash chokes on it.

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Pat L (pgogy)
Sent: 12 May 2014 14:58
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

i thought there was a valid XML test before saving?

On Mon, May 12, 2014 at 2:36 PM, Julian Tenney <Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>> wrote:
I asked if I could share the files, but not had a response: I'm sure it's OK, but I need to check,

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of David Goodwin
Sent: 12 May 2014 14:35
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Can you attach / share the data.xml file somehow?

I know of a PHP library which is quite good at cleaning up such stuff ....

David.


On 12/05/14 13:23, Julian Tenney wrote:
> This looks suspicious: that's not escaping, right? That looks to me
> like errant characters creeping in from somewhere? Paste from word, etc?
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of David
> Goodwin
> Sent: 12 May 2014 12:53
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> On 12/05/14 12:41, Julian Tenney wrote:
>
>  > What beats me is why the file should break halfway through a tag name.
>
> What I've seen is that the PHP (or whatever) can't read the XML file
> as it has the wrong encoding specified, or it just somehow contains
> characters that are invalid.
>
> see e.g.
>
> http://stackoverflow.com/questions/14463573/php-simplexml-load-file-in
> valid-character-error
>
> https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-non
> e-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/
>
> David.
>
>  >
>
>  > -----Original Message-----
>
>  > From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
> <mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>>
>
>  > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of
> David
>
>  > Goodwin
>
>  > Sent: 12 May 2014 12:37
>
>  > To: For Xerte technical developers
>
>  > Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
>  >
>
>  > Is it possible that it's a unicode ` character?
>
>  >
>
>  >
>
>  > I've seen a number of UTF-8 related issues with the data.xml etc files.
>
>  >
>
>  > David.
>
>  >
>
> --
>
> Pale Purple Ltd
>
> PHP Web application development and support
>
> http://www.palepurple.co.uk
>
> @PalePurpleLtd
>
> 07792 380669 / 0845 0046746
>
> _______________________________________________
>
> Xerte-dev mailing list
>
> Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
> <mailto:Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>>
>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/9a9180ff/attachment-0001.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 15:03:52 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 15:03:52 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B621.4020601@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
	<5370CE02.7070006@palepurple.co.uk> 
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C20@EXCHANGE1.ad.nottingham.ac.uk>

It's fine: they are attached,

-----Original Message-----
From: Tenney Julian 
Sent: 12 May 2014 14:36
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] Re: FW: Further Xerte Data Loss

I asked if I could share the files, but not had a response: I'm sure it's OK, but I need to check,

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 14:35
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Can you attach / share the data.xml file somehow?

I know of a PHP library which is quite good at cleaning up such stuff ....

David.


On 12/05/14 13:23, Julian Tenney wrote:
> This looks suspicious: that's not escaping, right? That looks to me 
> like errant characters creeping in from somewhere? Paste from word, etc?
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David 
> Goodwin
> Sent: 12 May 2014 12:53
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> On 12/05/14 12:41, Julian Tenney wrote:
>
>  > What beats me is why the file should break halfway through a tag name.
>
> What I've seen is that the PHP (or whatever) can't read the XML file 
> as it has the wrong encoding specified, or it just somehow contains 
> characters that are invalid.
>
> see e.g.
>
> http://stackoverflow.com/questions/14463573/php-simplexml-load-file-in
> valid-character-error
>
> https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-non
> e-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/
>
> David.
>
>  >
>
>  > -----Original Message-----
>
>  > From: xerte-dev-bounces at lists.nottingham.ac.uk
> <mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
>
>  > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of 
> David
>
>  > Goodwin
>
>  > Sent: 12 May 2014 12:37
>
>  > To: For Xerte technical developers
>
>  > Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
>  >
>
>  > Is it possible that it's a unicode ` character?
>
>  >
>
>  >
>
>  > I've seen a number of UTF-8 related issues with the data.xml etc files.
>
>  >
>
>  > David.
>
>  >
>
> --
>
> Pale Purple Ltd
>
> PHP Web application development and support
>
> http://www.palepurple.co.uk
>
> @PalePurpleLtd
>
> 07792 380669 / 0845 0046746
>
> _______________________________________________
>
> Xerte-dev mailing list
>
> Xerte-dev at lists.nottingham.ac.uk
> <mailto:Xerte-dev at lists.nottingham.ac.uk>
>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
-------------- next part --------------
A non-text attachment was scrubbed...
Name: preview.zip
Type: application/x-zip-compressed
Size: 32521 bytes
Desc: preview.zip
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/bf664f73/attachment-0001.bin>

From J.J.Smith at gcu.ac.uk  Mon May 12 15:23:38 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Mon, 12 May 2014 15:23:38 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <CAO2PinN9HxNp-3_HWQQ1cJE4r5gynUZxbc6dTSk0ajb=7Z_szQ@mail.gmail.com>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B621.4020601@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901B20@EXCHANGE1.ad.nottingham.ac.uk>
	<5370CE02.7070006@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901BCB@EXCHANGE1.ad.nottingham.ac.uk>,
	<CAO2PinN9HxNp-3_HWQQ1cJE4r5gynUZxbc6dTSk0ajb=7Z_szQ@mail.gmail.com>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC10209@ITSEMBXCLUS.enterprise.gcal.ac.uk>

The only check on the incoming data is this:

if(strlen($_POST['filedata'])!=strlen($_POST['filesize']))

Since both are coming from the Flash then it's possible that they could still match and the data could be corrupt...

I'd say a quick check of whether the end of the data ="</learningObject>" would be better than nothing to catch truncated files...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Pat L (pgogy) [xerte at pgogywebstuff.com]
Sent: 12 May 2014 14:58
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

i thought there was a valid XML test before saving?


On Mon, May 12, 2014 at 2:36 PM, Julian Tenney <Julian.Tenney at nottingham.ac.uk<mailto:Julian.Tenney at nottingham.ac.uk>> wrote:
I asked if I could share the files, but not had a response: I'm sure it's OK, but I need to check,

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of David Goodwin
Sent: 12 May 2014 14:35
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Can you attach / share the data.xml file somehow?

I know of a PHP library which is quite good at cleaning up such stuff ....

David.


On 12/05/14 13:23, Julian Tenney wrote:
> This looks suspicious: that's not escaping, right? That looks to me
> like errant characters creeping in from somewhere? Paste from word, etc?
>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of David
> Goodwin
> Sent: 12 May 2014 12:53
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> On 12/05/14 12:41, Julian Tenney wrote:
>
>  > What beats me is why the file should break halfway through a tag name.
>
> What I've seen is that the PHP (or whatever) can't read the XML file
> as it has the wrong encoding specified, or it just somehow contains
> characters that are invalid.
>
> see e.g.
>
> http://stackoverflow.com/questions/14463573/php-simplexml-load-file-in
> valid-character-error
>
> https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-non
> e-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/
>
> David.
>
>  >
>
>  > -----Original Message-----
>
>  > From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>
> <mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>>
>
>  > [mailto:xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk>] On Behalf Of
> David
>
>  > Goodwin
>
>  > Sent: 12 May 2014 12:37
>
>  > To: For Xerte technical developers
>
>  > Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
>  >
>
>  > Is it possible that it's a unicode ` character?
>
>  >
>
>  >
>
>  > I've seen a number of UTF-8 related issues with the data.xml etc files.
>
>  >
>
>  > David.
>
>  >
>
> --
>
> Pale Purple Ltd
>
> PHP Web application development and support
>
> http://www.palepurple.co.uk
>
> @PalePurpleLtd
>
> 07792 380669 / 0845 0046746
>
> _______________________________________________
>
> Xerte-dev mailing list
>
> Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
> <mailto:Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>>
>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk<mailto:Xerte-dev at lists.nottingham.ac.uk>
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From Julian.Tenney at nottingham.ac.uk  Mon May 12 15:34:57 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 15:34:57 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <5370B621.4020601@palepurple.co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>
	<BLU177-W422D21842359208837B91DA7350@phx.gbl>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B259.3060606@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>
	<5370B621.4020601@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C6A@EXCHANGE1.ad.nottingham.ac.uk>

The second of those links seems to describe the issue - is there a flash solution I can use to strip the non UTF8 out as it gets pasted in? I can't find one...

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 12:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss



On 12/05/14 12:41, Julian Tenney wrote:
> What beats me is why the file should break halfway through a tag name.


What I've seen is that the PHP (or whatever) can't read the XML file as it has the wrong encoding specified, or it just somehow contains characters that are invalid.

see e.g.

http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-character-error

https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-8-content-aka-how-to-remove-invalid-characters-in-utf-8/


David.

>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk 
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David 
> Goodwin
> Sent: 12 May 2014 12:37
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Is it possible that it's a unicode ` character?
>
>
> I've seen a number of UTF-8 related issues with the data.xml etc files.
>
> David.
>



--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From ronm at mitchellmedia.co.uk  Mon May 12 12:47:55 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Mon, 12 May 2014 12:47:55 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <00ea01cf6dd8$04760280$0d620780$@co.uk>

Very frustrating replying and seeing loads of other replies sent after mine
hitting the list before my reply even reaches! :-(

 

But anyway just re-reading the message from you user and I wonder.

 

1. Any likelihood of connection problems being the cause?

 

2. Wondering if rather than connectors or tab nav whether the problem only
occurs with page types that have nested pages?

 

 

From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk] 
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

 

I think the fact that it's a connector page in this case might be a red
herring. I previously thought it was to do with the tabnav page types but
that may be a red herring too.

 

Yes I've had reports of this quite a few times now (different installations)
but haven't been able to nail it down or reliably replicate it except that
in Alistairs case it was dodgy connection and so possibly packet loss.

 

The advice I've been giving is pasted below but I agree it's worrying
whenever this happens even though I haven't ever been able to make it happen
myself.

 

Sorry not much help I guess :-(

 

Guidance example given to one of the HEA projects:

I think you need to provide some guidance to the students to be very careful
when authoring if any connection problems are noticed. Also be careful when
copying and pasting text into xerte pages or writing code/html. Lost or
corrupted pages with xerte are very rare but the occurrences we do know
about either happen due to poor connectivity or to code/tags breaking the
xml. The key message is do not click publish if any problems are noticed
because preview saves the latest changes to preview.xml but these changes
are not written to data.xml until publish is selected. So if a problem is
noticed not clicking publish can sometimes mean that we can recover from
data.xml even if preview.xml has been broken.

 

 

From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Further Xerte Data Loss

 

I guys, I need to get to the bottom of this issue: I have a fairly advanced
user who has been periodically losing data: the email below explains the
problem. A couple of weeks ago it happened and we lost 40 slides, now it has
happened again. In the file I looked at, the broken file ended with the
characters

 

<cMen

 

i.e. an incomplete tag, which I believe should be <cMenuConnector
params="etc">

 

Does this ring any bells with anyone? Anyone experienced it before
elsewhere? I don't think the menu connectors are used much, which is why we
haven't come across it before?

 

I have a feeling it is something to do with escaping characters as data is
being saved, or as it is being read back to the editor. Anyone have any
other bright ideas how we can troubleshoot it? See below.

 

Thanks a lot,

 

Julian 

 

 

I've experienced further data loss again this morning, this time on a
completely different Xerte project (
<http://www.nottingham.ac.uk/toolkits/play_10189>
http://www.nottingham.ac.uk/toolkits/play_10189). 

 

I initially produced this resource using the Flash Xerte engine, but the
academic I'm working with was keen for me to transfer the data over into
HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5
engine and then set about reproducing the slides - no problem.

 

Everything was absolutely fine (after working on the project for most of
Friday) until around 10:30am this morning, when I lost around 10-15 slides
(including my review quiz with 20 MCQs). I'm now beginning to strongly
suspect that the Xerte Menu Connector templates might be having something to
do with this persistent problem? 

 

I have been using multiple connector menus in both my Grasslands and Meat
Science packages. In both I have a dominant 'Main Menu' at the start, which
then 'connects' via tabs to several sub-menus (of the same Connector
template type) dotted throughout the resource. In each of the sub-menus I
link back to the Main Menu.

 

My last actions before the slides were lost today, happened after I included
a further sub-menu connector (at around slide 70) and then linked this to
this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then
published the changes. 

 

With that being said, I've attempted to recreate the problem using a dummy
file, but I have so far failed to do it.

 

This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/a758cf40/attachment-0001.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 15:46:57 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 15:46:57 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <00ea01cf6dd8$04760280$0d620780$@co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>

1.       Possibly. He is remote to Nottingham, though has a Nottingham username.

2.       That doesn't trip any switches with me - what are you thinking?

Since I found non-UTF chars in the file, I have a strong suspicion that that is the cause: even though he is pasting into notepad from word (most of the time), some are getting through. I have suggested actually saving the file in notepad as UTF 8 to force it to happen.

But this is something I would really like to see fixed. Pasting from word is a fact of life unfortunately.

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Very frustrating replying and seeing loads of other replies sent after mine hitting the list before my reply even reaches! :-(

But anyway just re-reading the message from you user and I wonder...

1. Any likelihood of connection problems being the cause?

2. Wondering if rather than connectors or tab nav whether the problem only occurs with page types that have nested pages?


From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss.

The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself.

Sorry not much help I guess :-(

Guidance example given to one of the HEA projects:
I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.


From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Further Xerte Data Loss


I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters



<cMen



i.e. an incomplete tag, which I believe should be <cMenuConnector params="etc">



Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don't think the menu connectors are used much, which is why we haven't come across it before?



I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below...



Thanks a lot,



Julian





I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).



I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.



Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?



I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.



My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.



With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.


This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/a3d2016e/attachment.html>

From ronm at mitchellmedia.co.uk  Mon May 12 13:28:53 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Mon, 12 May 2014 13:28:53 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901B13@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<BLU177-W9114C1E39C3E23AC5F733A7350@phx.gbl>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A91@EXCHANGE1.ad.nottingham.ac.uk>	<BLU177-W422D21842359208837B91DA7350@phx.gbl>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AB4@EXCHANGE1.ad.nottingham.ac.uk>	<5370B259.3060606@palepurple.co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901AC4@EXCHANGE1.ad.nottingham.ac.uk>	<5370B621.4020601@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901B13@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <010101cf6ddd$bd417f60$37c47e20$@co.uk>

In my experience when there are problems like this preview.xml breaks first
then data.xml but I guess it depends if the user is in the habit of hitting
publish as a matter of course in which case both get corrupted.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 13:17
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

On this occasion, the file has broken half way through a CDATA section, so
everything ought to have escaped OK.

Just to be certain: data.xml is the published file, right, and preview.xml
is the editing file?

In which case it is the published file that has broken, and the editing file
is complete with a <learningObject> tag. I'll attach them, I'm just asking
if he minds if I share the files.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 12 May 2014 12:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss



On 12/05/14 12:41, Julian Tenney wrote:
> What beats me is why the file should break halfway through a tag name.


What I've seen is that the PHP (or whatever) can't read the XML file as it
has the wrong encoding specified, or it just somehow contains characters
that are invalid.

see e.g.

http://stackoverflow.com/questions/14463573/php-simplexml-load-file-invalid-
character-error

https://chrismckee.co.uk/saving-user-content-to-xml-error-contains-none-utf-
8-content-aka-how-to-remove-invalid-characters-in-utf-8/


David.

>
> -----Original Message-----
> From: xerte-dev-bounces at lists.nottingham.ac.uk
> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David 
> Goodwin
> Sent: 12 May 2014 12:37
> To: For Xerte technical developers
> Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
>
> Is it possible that it's a unicode ` character?
>
>
> I've seen a number of UTF-8 related issues with the data.xml etc files.
>
> David.
>



--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.






From ronm at mitchellmedia.co.uk  Mon May 12 12:32:52 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Mon, 12 May 2014 12:32:52 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <00be01cf6dd5$ea5c68a0$bf1539e0$@co.uk>

I think the fact that it's a connector page in this case might be a red
herring. I previously thought it was to do with the tabnav page types but
that may be a red herring too.

 

Yes I've had reports of this quite a few times now (different installations)
but haven't been able to nail it down or reliably replicate it except that
in Alistairs case it was dodgy connection and so possibly packet loss.

 

The advice I've been giving is pasted below but I agree it's worrying
whenever this happens even though I haven't ever been able to make it happen
myself.

 

Sorry not much help I guess :-(

 

Guidance example given to one of the HEA projects:

I think you need to provide some guidance to the students to be very careful
when authoring if any connection problems are noticed. Also be careful when
copying and pasting text into xerte pages or writing code/html. Lost or
corrupted pages with xerte are very rare but the occurrences we do know
about either happen due to poor connectivity or to code/tags breaking the
xml. The key message is do not click publish if any problems are noticed
because preview saves the latest changes to preview.xml but these changes
are not written to data.xml until publish is selected. So if a problem is
noticed not clicking publish can sometimes mean that we can recover from
data.xml even if preview.xml has been broken.

 

 

From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Further Xerte Data Loss

 

I guys, I need to get to the bottom of this issue: I have a fairly advanced
user who has been periodically losing data: the email below explains the
problem. A couple of weeks ago it happened and we lost 40 slides, now it has
happened again. In the file I looked at, the broken file ended with the
characters

 

<cMen

 

i.e. an incomplete tag, which I believe should be <cMenuConnector
params="etc">

 

Does this ring any bells with anyone? Anyone experienced it before
elsewhere? I don't think the menu connectors are used much, which is why we
haven't come across it before?

 

I have a feeling it is something to do with escaping characters as data is
being saved, or as it is being read back to the editor. Anyone have any
other bright ideas how we can troubleshoot it? See below.

 

Thanks a lot,

 

Julian 

 

 

I've experienced further data loss again this morning, this time on a
completely different Xerte project (
<http://www.nottingham.ac.uk/toolkits/play_10189>
http://www.nottingham.ac.uk/toolkits/play_10189). 

 

I initially produced this resource using the Flash Xerte engine, but the
academic I'm working with was keen for me to transfer the data over into
HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5
engine and then set about reproducing the slides - no problem.

 

Everything was absolutely fine (after working on the project for most of
Friday) until around 10:30am this morning, when I lost around 10-15 slides
(including my review quiz with 20 MCQs). I'm now beginning to strongly
suspect that the Xerte Menu Connector templates might be having something to
do with this persistent problem? 

 

I have been using multiple connector menus in both my Grasslands and Meat
Science packages. In both I have a dominant 'Main Menu' at the start, which
then 'connects' via tabs to several sub-menus (of the same Connector
template type) dotted throughout the resource. In each of the sub-menus I
link back to the Main Menu.

 

My last actions before the slides were lost today, happened after I included
a further sub-menu connector (at around slide 70) and then linked this to
this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then
published the changes. 

 

With that being said, I've attempted to recreate the problem using a dummy
file, but I have so far failed to do it.

 

This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/8689e9c8/attachment.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 15:52:28 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 15:52:28 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <00be01cf6dd5$ea5c68a0$bf1539e0$@co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00be01cf6dd5$ea5c68a0$bf1539e0$@co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C8C@EXCHANGE1.ad.nottingham.ac.uk>

> I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

So do I.

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss.

The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself.

Sorry not much help I guess :-(

Guidance example given to one of the HEA projects:
I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.


From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Further Xerte Data Loss


I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters



<cMen



i.e. an incomplete tag, which I believe should be <cMenuConnector params="etc">



Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don't think the menu connectors are used much, which is why we haven't come across it before?



I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below...



Thanks a lot,



Julian





I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).



I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.



Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?



I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.



My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.



With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.


This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/03e0d7f5/attachment-0001.html>

From ronm at mitchellmedia.co.uk  Mon May 12 15:56:51 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Mon, 12 May 2014 15:56:51 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <00ea01cf6dd8$04760280$0d620780$@co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
Message-ID: <019e01cf6df2$69651860$3c2f4920$@co.uk>

Note: there has obviously been lots of subsequent discussion but my replies
below were sent at 12:33 and seemed to reach the list at 15:45! :-(

 

From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

 

Very frustrating replying and seeing loads of other replies sent after mine
hitting the list before my reply even reaches! :-(

 

But anyway just re-reading the message from you user and I wonder.

 

1. Any likelihood of connection problems being the cause?

 

2. Wondering if rather than connectors or tab nav whether the problem only
occurs with page types that have nested pages?

 

 

From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk] 
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

 

I think the fact that it's a connector page in this case might be a red
herring. I previously thought it was to do with the tabnav page types but
that may be a red herring too.

 

Yes I've had reports of this quite a few times now (different installations)
but haven't been able to nail it down or reliably replicate it except that
in Alistairs case it was dodgy connection and so possibly packet loss.

 

The advice I've been giving is pasted below but I agree it's worrying
whenever this happens even though I haven't ever been able to make it happen
myself.

 

Sorry not much help I guess :-(

 

Guidance example given to one of the HEA projects:

I think you need to provide some guidance to the students to be very careful
when authoring if any connection problems are noticed. Also be careful when
copying and pasting text into xerte pages or writing code/html. Lost or
corrupted pages with xerte are very rare but the occurrences we do know
about either happen due to poor connectivity or to code/tags breaking the
xml. The key message is do not click publish if any problems are noticed
because preview saves the latest changes to preview.xml but these changes
are not written to data.xml until publish is selected. So if a problem is
noticed not clicking publish can sometimes mean that we can recover from
data.xml even if preview.xml has been broken.

 

 

From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Further Xerte Data Loss

 

I guys, I need to get to the bottom of this issue: I have a fairly advanced
user who has been periodically losing data: the email below explains the
problem. A couple of weeks ago it happened and we lost 40 slides, now it has
happened again. In the file I looked at, the broken file ended with the
characters

 

<cMen

 

i.e. an incomplete tag, which I believe should be <cMenuConnector
params="etc">

 

Does this ring any bells with anyone? Anyone experienced it before
elsewhere? I don't think the menu connectors are used much, which is why we
haven't come across it before?

 

I have a feeling it is something to do with escaping characters as data is
being saved, or as it is being read back to the editor. Anyone have any
other bright ideas how we can troubleshoot it? See below.

 

Thanks a lot,

 

Julian 

 

 

I've experienced further data loss again this morning, this time on a
completely different Xerte project (
<http://www.nottingham.ac.uk/toolkits/play_10189>
http://www.nottingham.ac.uk/toolkits/play_10189). 

 

I initially produced this resource using the Flash Xerte engine, but the
academic I'm working with was keen for me to transfer the data over into
HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5
engine and then set about reproducing the slides - no problem.

 

Everything was absolutely fine (after working on the project for most of
Friday) until around 10:30am this morning, when I lost around 10-15 slides
(including my review quiz with 20 MCQs). I'm now beginning to strongly
suspect that the Xerte Menu Connector templates might be having something to
do with this persistent problem? 

 

I have been using multiple connector menus in both my Grasslands and Meat
Science packages. In both I have a dominant 'Main Menu' at the start, which
then 'connects' via tabs to several sub-menus (of the same Connector
template type) dotted throughout the resource. In each of the sub-menus I
link back to the Main Menu.

 

My last actions before the slides were lost today, happened after I included
a further sub-menu connector (at around slide 70) and then linked this to
this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then
published the changes. 

 

With that being said, I've attempted to recreate the problem using a dummy
file, but I have so far failed to do it.

 

This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

 

 

This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/71d66d6d/attachment.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 15:55:51 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 15:55:51 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <019e01cf6df2$69651860$3c2f4920$@co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
	<019e01cf6df2$69651860$3c2f4920$@co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C93@EXCHANGE1.ad.nottingham.ac.uk>

Are you on some sort of watch list, where all your mail has to get approved by someone somewhere...?

;-)

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 15:57
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Note: there has obviously been lots of subsequent discussion but my replies below were sent at 12:33 and seemed to reach the list at 15:45! :-(

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Very frustrating replying and seeing loads of other replies sent after mine hitting the list before my reply even reaches! :-(

But anyway just re-reading the message from you user and I wonder...

1. Any likelihood of connection problems being the cause?

2. Wondering if rather than connectors or tab nav whether the problem only occurs with page types that have nested pages?


From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss.

The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself.

Sorry not much help I guess :-(

Guidance example given to one of the HEA projects:
I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.


From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Further Xerte Data Loss


I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters



<cMen



i.e. an incomplete tag, which I believe should be <cMenuConnector params="etc">



Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don't think the menu connectors are used much, which is why we haven't come across it before?



I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below...



Thanks a lot,



Julian





I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).



I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.



Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?



I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.



My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.



With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.


This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/235f0e1d/attachment-0001.html>

From d_b_burnett at hotmail.com  Mon May 12 16:00:54 2014
From: d_b_burnett at hotmail.com (Dave Burnett)
Date: Mon, 12 May 2014 11:00:54 -0400
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C93@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>,
	<00ea01cf6dd8$04760280$0d620780$@co.uk>,
	<019e01cf6df2$69651860$3c2f4920$@co.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C93@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <BLU177-W44D016CE91694FE94972B2A7350@phx.gbl>

lol.
Or does your accessorizing include and ankle monitor?
From: Julian.Tenney at nottingham.ac.uk
To: xerte-dev at lists.nottingham.ac.uk
Date: Mon, 12 May 2014 15:55:51 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Are you on some sort of watch list, where all your mail has to get approved by someone somewhere?? ;-) From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 15:57
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss Note: there has obviously been lots of subsequent discussion but my replies below were sent at 12:33 and seemed to reach the list at 15:45! :-( From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss Very frustrating replying and seeing loads of other replies sent after mine hitting the list before my reply even reaches! :-( But anyway just re-reading the message from you user and I wonder? 1. Any likelihood of connection problems being the cause? 2. Wondering if rather than connectors or tab nav whether the problem only occurs with page types that have nested pages?  From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk] 
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too. Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss. The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself. Sorry not much help I guess :-( Guidance example given to one of the HEA projects:I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.  From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Further Xerte Data Loss I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters <cMen i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?> Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before? I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below? Thanks a lot, Julian   I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).  I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem. Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?  I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu. My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.  With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it. This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
  This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.
 

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.
This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/6a647fb3/attachment.html>

From ronm at mitchellmedia.co.uk  Mon May 12 16:04:46 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Mon, 12 May 2014 16:04:46 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C93@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>	<00ea01cf6dd8$04760280$0d620780$@co.uk>	<019e01cf6df2$69651860$3c2f4920$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C93@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <01ca01cf6df3$845b7320$8d125960$@co.uk>

Yeah the Nottingham Uni mailing list police! ;-)

 

Never happens on any other lists or systems :-(

 

From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 15:56
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

 

Are you on some sort of watch list, where all your mail has to get approved
by someone somewhere.?

 

;-)

 

From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 15:57
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

 

Note: there has obviously been lots of subsequent discussion but my replies
below were sent at 12:33 and seemed to reach the list at 15:45! :-(

 

From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

 

Very frustrating replying and seeing loads of other replies sent after mine
hitting the list before my reply even reaches! :-(

 

But anyway just re-reading the message from you user and I wonder.

 

1. Any likelihood of connection problems being the cause?

 

2. Wondering if rather than connectors or tab nav whether the problem only
occurs with page types that have nested pages?

 

 

From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk] 
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

 

I think the fact that it's a connector page in this case might be a red
herring. I previously thought it was to do with the tabnav page types but
that may be a red herring too.

 

Yes I've had reports of this quite a few times now (different installations)
but haven't been able to nail it down or reliably replicate it except that
in Alistairs case it was dodgy connection and so possibly packet loss.

 

The advice I've been giving is pasted below but I agree it's worrying
whenever this happens even though I haven't ever been able to make it happen
myself.

 

Sorry not much help I guess :-(

 

Guidance example given to one of the HEA projects:

I think you need to provide some guidance to the students to be very careful
when authoring if any connection problems are noticed. Also be careful when
copying and pasting text into xerte pages or writing code/html. Lost or
corrupted pages with xerte are very rare but the occurrences we do know
about either happen due to poor connectivity or to code/tags breaking the
xml. The key message is do not click publish if any problems are noticed
because preview saves the latest changes to preview.xml but these changes
are not written to data.xml until publish is selected. So if a problem is
noticed not clicking publish can sometimes mean that we can recover from
data.xml even if preview.xml has been broken.

 

 

From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Further Xerte Data Loss

 

I guys, I need to get to the bottom of this issue: I have a fairly advanced
user who has been periodically losing data: the email below explains the
problem. A couple of weeks ago it happened and we lost 40 slides, now it has
happened again. In the file I looked at, the broken file ended with the
characters

 

<cMen

 

i.e. an incomplete tag, which I believe should be <cMenuConnector
params="etc">

 

Does this ring any bells with anyone? Anyone experienced it before
elsewhere? I don't think the menu connectors are used much, which is why we
haven't come across it before?

 

I have a feeling it is something to do with escaping characters as data is
being saved, or as it is being read back to the editor. Anyone have any
other bright ideas how we can troubleshoot it? See below.

 

Thanks a lot,

 

Julian 

 

 

I've experienced further data loss again this morning, this time on a
completely different Xerte project (
<http://www.nottingham.ac.uk/toolkits/play_10189>
http://www.nottingham.ac.uk/toolkits/play_10189). 

 

I initially produced this resource using the Flash Xerte engine, but the
academic I'm working with was keen for me to transfer the data over into
HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5
engine and then set about reproducing the slides - no problem.

 

Everything was absolutely fine (after working on the project for most of
Friday) until around 10:30am this morning, when I lost around 10-15 slides
(including my review quiz with 20 MCQs). I'm now beginning to strongly
suspect that the Xerte Menu Connector templates might be having something to
do with this persistent problem? 

 

I have been using multiple connector menus in both my Grasslands and Meat
Science packages. In both I have a dominant 'Main Menu' at the start, which
then 'connects' via tabs to several sub-menus (of the same Connector
template type) dotted throughout the resource. In each of the sub-menus I
link back to the Main Menu.

 

My last actions before the slides were lost today, happened after I included
a further sub-menu connector (at around slide 70) and then linked this to
this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then
published the changes. 

 

With that being said, I've attempted to recreate the problem using a dummy
file, but I have so far failed to do it.

 

This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

 

 

This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

 

 

This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/bbfa48cb/attachment-0001.html>

From J.J.Smith at gcu.ac.uk  Mon May 12 16:01:15 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Mon, 12 May 2014 16:01:15 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C93@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
	<019e01cf6df2$69651860$3c2f4920$@co.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C93@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020A@ITSEMBXCLUS.enterprise.gcal.ac.uk>

NSA filtered me thinks ;-)

Tom and I are working on the new editor just now... should we look at building in some kind of "history"... I'd like to see a undo/redo buttons but should we have a history that persists to server?

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 12 May 2014 15:55
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Are you on some sort of watch list, where all your mail has to get approved by someone somewhere??

;-)

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 15:57
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Note: there has obviously been lots of subsequent discussion but my replies below were sent at 12:33 and seemed to reach the list at 15:45! :-(

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Very frustrating replying and seeing loads of other replies sent after mine hitting the list before my reply even reaches! :-(

But anyway just re-reading the message from you user and I wonder?

1. Any likelihood of connection problems being the cause?

2. Wondering if rather than connectors or tab nav whether the problem only occurs with page types that have nested pages?


From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss.

The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself.

Sorry not much help I guess :-(

Guidance example given to one of the HEA projects:
I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.


From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Further Xerte Data Loss


I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters



<cMen



i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?>



Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before?



I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below?



Thanks a lot,



Julian





I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).



I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.



Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?



I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.



My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.



With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.


This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number SC021474


From xerte at pgogywebstuff.com  Mon May 12 16:08:39 2014
From: xerte at pgogywebstuff.com (Pat L (pgogy))
Date: Mon, 12 May 2014 16:08:39 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <019e01cf6df2$69651860$3c2f4920$@co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
	<019e01cf6df2$69651860$3c2f4920$@co.uk>
Message-ID: <CAO2PinNnP3fyrkFb5GmJ3DW8S1Yu3oe9T07+DTJ0Ac_3E-XYPg@mail.gmail.com>

same here, and the uni signature off is just annoying on long threads


On Mon, May 12, 2014 at 3:56 PM, Ron Mitchell <ronm at mitchellmedia.co.uk>wrote:

> Note: there has obviously been lots of subsequent discussion but my
> replies below were sent at 12:33 and seemed to reach the list at 15:45! :-(
>
>
>
> *From:* xerte-dev-bounces at lists.nottingham.ac.uk [mailto:
> xerte-dev-bounces at lists.nottingham.ac.uk] *On Behalf Of *Ron Mitchell
> *Sent:* 12 May 2014 12:48
>
> *To:* 'For Xerte technical developers'
> *Subject:* [Xerte-dev] Re: FW: Further Xerte Data Loss
>
>
>
> Very frustrating replying and seeing loads of other replies sent after
> mine hitting the list before my reply even reaches! :-(
>
>
>
> But anyway just re-reading the message from you user and I wonder?
>
>
>
> 1. Any likelihood of connection problems being the cause?
>
>
>
> 2. Wondering if rather than connectors or tab nav whether the problem only
> occurs with page types that have nested pages?
>
>
>
>
>
> *From:* Ron Mitchell [mailto:ronm at mitchellmedia.co.uk<ronm at mitchellmedia.co.uk>]
>
> *Sent:* 12 May 2014 12:33
> *To:* 'For Xerte technical developers'
> *Subject:* RE: [Xerte-dev] FW: Further Xerte Data Loss
>
>
>
> I think the fact that it's a connector page in this case might be a red
> herring. I previously thought it was to do with the tabnav page types but
> that may be a red herring too.
>
>
>
> Yes I've had reports of this quite a few times now (different
> installations) but haven't been able to nail it down or reliably replicate
> it except that in Alistairs case it was dodgy connection and so possibly
> packet loss.
>
>
>
> The advice I've been giving is pasted below but I agree it's worrying
> whenever this happens even though I haven't ever been able to make it
> happen myself.
>
>
>
> Sorry not much help I guess :-(
>
>
>
> Guidance example given to one of the HEA projects:
>
> I think you need to provide some guidance to the students to be very
> careful when authoring if any connection problems are noticed. Also be
> careful when copying and pasting text into xerte pages or writing
> code/html. Lost or corrupted pages with xerte are very rare but the
> occurrences we do know about either happen due to poor connectivity or to
> code/tags breaking the xml. The key message is do not click publish if any
> problems are noticed because preview saves the latest changes to
> preview.xml but these changes are not written to data.xml until publish is
> selected. So if a problem is noticed not clicking publish can sometimes
> mean that we can recover from data.xml even if preview.xml has been broken.
>
>
>
>
>
> *From:* xerte-dev-bounces at lists.nottingham.ac.uk [
> mailto:xerte-dev-bounces at lists.nottingham.ac.uk<xerte-dev-bounces at lists.nottingham.ac.uk>]
> *On Behalf Of *Julian Tenney
> *Sent:* 12 May 2014 12:08
> *To:* For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
> *Subject:* [Xerte-dev] FW: Further Xerte Data Loss
>
>
>
> I guys, I need to get to the bottom of this issue: I have a fairly
> advanced user who has been periodically losing data: the email below
> explains the problem. A couple of weeks ago it happened and we lost 40
> slides, now it has happened again. In the file I looked at, the broken file
> ended with the characters
>
>
>
> <cMen
>
>
>
> i.e. an incomplete tag, which I believe should be <cMenuConnector
> params=?etc?>
>
>
>
> Does this ring any bells with anyone? Anyone experienced it before
> elsewhere? I don?t think the menu connectors are used much, which is why we
> haven?t come across it before?
>
>
>
> I have a feeling it is something to do with escaping characters as data is
> being saved, or as it is being read back to the editor. Anyone have any
> other bright ideas how we can troubleshoot it? See below?
>
>
>
> Thanks a lot,
>
>
>
> Julian
>
>
>
>
>
> I've experienced further data loss again this morning, this time on a
> completely different Xerte project (
> http://www.nottingham.ac.uk/toolkits/play_10189).
>
>
>
> I initially produced this resource using the Flash Xerte engine, but the
> academic I'm working with was keen for me to transfer the data over into
> HTML5. I made a direct duplicate copy of the Flash project, ticked the
> HTML5 engine and then set about reproducing the slides - no problem.
>
>
>
> Everything was absolutely fine (after working on the project for most of
> Friday) until around 10:30am this morning, when I lost around 10-15 slides
> (including my review quiz with 20 MCQs). I'm now beginning to strongly
> suspect that the Xerte Menu Connector templates might be having something
> to do with this persistent problem?
>
>
>
> I have been using multiple connector menus in both my Grasslands and Meat
> Science packages. In both I have a dominant 'Main Menu' at the start, which
> then 'connects' via tabs to several sub-menus (of the same Connector
> template type) dotted throughout the resource. In each of the sub-menus I
> link back to the Main Menu.
>
>
>
> My last actions before the slides were lost today, happened after I
> included a further sub-menu connector (at around slide 70) and then linked
> this to this from the 'Main Menu' (slide 2) via the drop-down selector
> menu. I then published the changes.
>
>
>
> With that being said, I've attempted to recreate the problem using a dummy
> file, but I have so far failed to do it.
>
>
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
>
>
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/a3c1da0d/attachment-0001.html>

From ronm at mitchellmedia.co.uk  Mon May 12 16:13:49 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Mon, 12 May 2014 16:13:49 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>	<00ea01cf6dd8$04760280$0d620780$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <01d501cf6df4$c7cb7000$57625000$@co.uk>

The occasions I know of it's mostly been impossible to troubleshoot due to lack of detailed info or too late data already lost.

 

Alistair had regular problems and while we were trying to nail it down (over weeks) too many inconsistencies e.g. different projects, some apparently no copy and paste involved but only seemed to happen with dodgy connection.

 

The reason I mentioned nested pages is that seems to be common whenever this has happened but may just be coincidental.

 

I've also seen those ??? characters before but I've seen broken xml without any of that and I think I've seen working LO's which haven't broken and yet had those chars.

 

Is there a way of reliably generating those sorts of chars to past in to prove these are the cause?

 

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 15:47
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

 

1.      Possibly. He is remote to Nottingham, though has a Nottingham username.

2.      That doesn?t trip any switches with me ? what are you thinking?

 

Since I found non-UTF chars in the file, I have a strong suspicion that that is the cause: even though he is pasting into notepad from word (most of the time), some are getting through. I have suggested actually saving the file in notepad as UTF 8 to force it to happen.

 

But this is something I would really like to see fixed. Pasting from word is a fact of life unfortunately.

 

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

 

Very frustrating replying and seeing loads of other replies sent after mine hitting the list before my reply even reaches! :-(

 

But anyway just re-reading the message from you user and I wonder?

 

1. Any likelihood of connection problems being the cause?

 

2. Wondering if rather than connectors or tab nav whether the problem only occurs with page types that have nested pages?

 

 

From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk] 
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

 

I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

 

Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss.

 

The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself.

 

Sorry not much help I guess :-(

 

Guidance example given to one of the HEA projects:

I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.

 

 

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Further Xerte Data Loss

 

I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters

 

<cMen

 

i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?>

 

Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before?

 

I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below?

 

Thanks a lot,

 

Julian 

 

 

I've experienced further data loss again this morning, this time on a completely different Xerte project ( <http://www.nottingham.ac.uk/toolkits/play_10189> http://www.nottingham.ac.uk/toolkits/play_10189). 

 

I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.

 

Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem? 

 

I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.

 

My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes. 

 

With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.

 

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

 

 

This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140512/a962c53c/attachment-0001.html>

From Julian.Tenney at nottingham.ac.uk  Mon May 12 16:17:30 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Mon, 12 May 2014 16:17:30 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020A@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
	<019e01cf6df2$69651860$3c2f4920$@co.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C93@EXCHANGE1.ad.nottingham.ac.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020A@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1A901CB7@EXCHANGE1.ad.nottingham.ac.uk>

History is a good idea. 

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 12 May 2014 16:01
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

NSA filtered me thinks ;-)

Tom and I are working on the new editor just now... should we look at building in some kind of "history"... I'd like to see a undo/redo buttons but should we have a history that persists to server?

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 12 May 2014 15:55
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Are you on some sort of watch list, where all your mail has to get approved by someone somewhere...?

;-)

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 15:57
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Note: there has obviously been lots of subsequent discussion but my replies below were sent at 12:33 and seemed to reach the list at 15:45! :-(

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Very frustrating replying and seeing loads of other replies sent after mine hitting the list before my reply even reaches! :-(

But anyway just re-reading the message from you user and I wonder...

1. Any likelihood of connection problems being the cause?

2. Wondering if rather than connectors or tab nav whether the problem only occurs with page types that have nested pages?


From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss.

The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself.

Sorry not much help I guess :-(

Guidance example given to one of the HEA projects:
I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.


From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Further Xerte Data Loss


I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters



<cMen



i.e. an incomplete tag, which I believe should be <cMenuConnector params="etc">



Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don't think the menu connectors are used much, which is why we haven't come across it before?



I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below...



Thanks a lot,



Julian





I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).



I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.



Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?



I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.



My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.



With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.


This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From david at palepurple.co.uk  Mon May 12 16:54:09 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Mon, 12 May 2014 16:54:09 +0100
Subject: [Xerte-dev] Various pull requests on github (sqlite support,
 security fixes, export zip fixes).
Message-ID: <5370EEA1.2080803@palepurple.co.uk>

Hi,

See :

https://github.com/thexerteproject/xerteonlinetoolkits/pull/65

  * changes XOT to break it's dependency on mysql_ (outside of 'setup')
  * removes the requirement to have mysql - it is defaulting to sqlite - 
this ought to make it easier for people to test.
  * removes any need to have a 'setup/' installer (to a large part). 
Chucking the files into a web server and browsing to the install should 
trigger it to create an empty database in the system temp dir.


See :

https://github.com/thexerteproject/xerteonlinetoolkits/pull/66

  * When I browse away from the front page, or try and reload it, 
Firefox get's stuck in a refresh loop.


and :

See :

https://github.com/thexerteproject/xerteonlinetoolkits/pull/57

  * Improve the storing of extra_flags against a template.
  * Add new ZIP export module (use php's ZIP extension if possible, 
otherwise fall back to the PHP userspace library).
  * this should resolve memory limit issues with exporting zip files.


thanks
David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From J.J.Smith at gcu.ac.uk  Mon May 12 16:27:24 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Mon, 12 May 2014 16:27:24 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <01d501cf6df4$c7cb7000$57625000$@co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>,
	<01d501cf6df4$c7cb7000$57625000$@co.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020B@ITSEMBXCLUS.enterprise.gcal.ac.uk>

AFAIK Ron they are just the characters that Word stick in when you type "word" and you get the special opening/closing quotes etc...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell [ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 16:13
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

The occasions I know of it's mostly been impossible to troubleshoot due to lack of detailed info or too late data already lost.

Alistair had regular problems and while we were trying to nail it down (over weeks) too many inconsistencies e.g. different projects, some apparently no copy and paste involved but only seemed to happen with dodgy connection.

The reason I mentioned nested pages is that seems to be common whenever this has happened but may just be coincidental.

I've also seen those ??? characters before but I've seen broken xml without any of that and I think I've seen working LO's which haven't broken and yet had those chars.

Is there a way of reliably generating those sorts of chars to past in to prove these are the cause?

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 15:47
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss


1.      Possibly. He is remote to Nottingham, though has a Nottingham username.

2.      That doesn?t trip any switches with me ? what are you thinking?

Since I found non-UTF chars in the file, I have a strong suspicion that that is the cause: even though he is pasting into notepad from word (most of the time), some are getting through. I have suggested actually saving the file in notepad as UTF 8 to force it to happen.

But this is something I would really like to see fixed. Pasting from word is a fact of life unfortunately.

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Very frustrating replying and seeing loads of other replies sent after mine hitting the list before my reply even reaches! :-(

But anyway just re-reading the message from you user and I wonder?

1. Any likelihood of connection problems being the cause?

2. Wondering if rather than connectors or tab nav whether the problem only occurs with page types that have nested pages?


From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss.

The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself.

Sorry not much help I guess :-(

Guidance example given to one of the HEA projects:
I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.


From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Further Xerte Data Loss


I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters



<cMen



i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?>



Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before?



I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below?



Thanks a lot,



Julian





I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).



I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.



Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?



I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.



My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.



With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.


This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number SC021474


From reijnders at tor.nl  Tue May 13 10:24:32 2014
From: reijnders at tor.nl (Tom Reijnders)
Date: Tue, 13 May 2014 11:24:32 +0200
Subject: [Xerte-dev] Re: Various pull requests on github (sqlite support,
	security fixes, export zip fixes).
In-Reply-To: <5370EEA1.2080803@palepurple.co.uk>
References: <5370EEA1.2080803@palepurple.co.uk>
Message-ID: <2b5b42f5-cf4c-49de-ba90-ecf62ae0e76f@email.android.com>

Thank you David! I'll try to take care of it tonight.

Tom

On 12 mei 2014 17:54:09 CEST, David Goodwin <david at palepurple.co.uk> wrote:
>Hi,
>
>See :
>
>https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>
>  * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>* removes the requirement to have mysql - it is defaulting to sqlite - 
>this ought to make it easier for people to test.
>  * removes any need to have a 'setup/' installer (to a large part). 
>Chucking the files into a web server and browsing to the install should
>
>trigger it to create an empty database in the system temp dir.
>
>
>See :
>
>https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>
>  * When I browse away from the front page, or try and reload it, 
>Firefox get's stuck in a refresh loop.
>
>
>and :
>
>See :
>
>https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>
>  * Improve the storing of extra_flags against a template.
>  * Add new ZIP export module (use php's ZIP extension if possible, 
>otherwise fall back to the PHP userspace library).
>  * this should resolve memory limit issues with exporting zip files.
>
>
>thanks
>David.
>
>-- 
>Pale Purple Ltd
>
>PHP Web application development and support
>
>http://www.palepurple.co.uk
>@PalePurpleLtd
>07792 380669 / 0845 0046746
>
>_______________________________________________
>Xerte-dev mailing list
>Xerte-dev at lists.nottingham.ac.uk
>http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>This message and any attachment are intended solely for the addressee
>and may contain confidential information. If you have received this
>message in error, please send it back to me, and immediately delete it.
>Please do not use, copy or disclose the information contained in this
>message or in any attachment.  Any views or opinions expressed by the
>author of this email do not necessarily reflect the views of the
>University of Nottingham.
>
>This message has been checked for viruses but the contents of an
>attachment
>may still contain software viruses which could damage your computer
>system, you are advised to perform your own checks. Email
>communications with the University of Nottingham may be monitored as
>permitted by UK legislation.

-- 
Verzonden van mijn Android telefoon met K-9 Mail.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/cc346859/attachment-0001.html>

From reijnders at tor.nl  Tue May 13 21:02:51 2014
From: reijnders at tor.nl (Tom Reijnders)
Date: Tue, 13 May 2014 22:02:51 +0200
Subject: [Xerte-dev] Re: Various pull requests on github (sqlite support,
 security fixes, export zip fixes).
In-Reply-To: <2b5b42f5-cf4c-49de-ba90-ecf62ae0e76f@email.android.com>
References: <5370EEA1.2080803@palepurple.co.uk>
	<2b5b42f5-cf4c-49de-ba90-ecf62ae0e76f@email.android.com>
Message-ID: <53727A6B.7030107@tor.nl>

Done.

Why does session_handler still use mysql_connect? I just happen to see 
that one.

Tom

Tom Reijnders schreef op 13-5-2014 11:24:
> Thank you David! I'll try to take care of it tonight.
>
> Tom
>
> On 12 mei 2014 17:54:09 CEST, David Goodwin <david at palepurple.co.uk> 
> wrote:
>
>     Hi,
>
>     See :
>
>     https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>
>        * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>        * removes the requirement to have mysql - it is defaulting to sqlite -
>     this ought to make it easier for people to test.
>        * removes any need to have a 'setup/' installer (to a large part).
>     Chucking the files into a web server and browsing to the install should
>     trigger it to create an empty database in the system temp dir.
>
>
>     See :
>
>     https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>
>        * When I browse away from the front page, or try and reload it,
>     Firefox get's stuck in a refresh loop.
>
>
>     and :
>
>     See :
>
>     https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>
>        * Improve the storing of extra_flags against a template.
>        * Add new ZIP export module (use php's ZIP extension if possible,
>     otherwise fall back to the PHP userspace library).
>        * this should resolve memory limit issues with exporting zip files.
>
>
>     thanks
>     David.
>
>
> -- 
> Verzonden van mijn Android telefoon met K-9 Mail.
>
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this 
> message in error, please send it back to me, and immediately delete 
> it.   Please do not use, copy or disclose the information contained in 
> this message or in any attachment.  Any views or opinions expressed by 
> the author of this email do not necessarily reflect the views of the 
> University of Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as 
> permitted by UK legislation.
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

-- 
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/15aa40ae/attachment.html>

From xerte at pgogywebstuff.com  Tue May 13 21:40:36 2014
From: xerte at pgogywebstuff.com (Pat L (pgogy))
Date: Tue, 13 May 2014 21:40:36 +0100
Subject: [Xerte-dev] Re: Various pull requests on github (sqlite support,
 security fixes, export zip fixes).
In-Reply-To: <53727A6B.7030107@tor.nl>
References: <5370EEA1.2080803@palepurple.co.uk>
	<2b5b42f5-cf4c-49de-ba90-ecf62ae0e76f@email.android.com>
	<53727A6B.7030107@tor.nl>
Message-ID: <CAO2PinP7bSMnkYLxWyTsUpuR831MuqOAqTmKQN=v2GOcNcFJyA@mail.gmail.com>

Is the session handler used at all?


On Tue, May 13, 2014 at 9:02 PM, Tom Reijnders <reijnders at tor.nl> wrote:

>  Done.
>
> Why does session_handler still use mysql_connect? I just happen to see
> that one.
>
> Tom
>
> Tom Reijnders schreef op 13-5-2014 11:24:
>
> Thank you David! I'll try to take care of it tonight.
>
> Tom
>
> On 12 mei 2014 17:54:09 CEST, David Goodwin <david at palepurple.co.uk><david at palepurple.co.uk>wrote:
>>
>> Hi,
>>
>> See :
>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>>
>>   * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>>   * removes the requirement to have mysql - it is defaulting to sqlite -
>> this ought to make it easier for people to test.
>>   * removes any need to have a 'setup/' installer (to a large part).
>> Chucking the files into a web server and browsing to the install should
>> trigger it to create an empty database in the system temp dir.
>>
>>
>> See :
>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>>
>>   * When I browse away from the front page, or try and reload it,
>> Firefox get's stuck in a refresh loop.
>>
>>
>> and :
>>
>> See :
>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>>
>>   * Improve the storing of extra_flags against a template.
>>   * Add new ZIP export module (use php's ZIP extension if possible,
>> otherwise fall back to the PHP userspace library).
>>   * this should resolve memory limit issues with exporting zip files.
>>
>>
>> thanks
>> David.
>>
>>
> --
> Verzonden van mijn Android telefoon met K-9 Mail.
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
>
> _______________________________________________
> Xerte-dev mailing listXerte-dev at lists.nottingham.ac.ukhttp://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
> --
> --
>
> Tom Reijnders
> TOR Informatica
> Chopinlaan 27
> 5242HM Rosmalen
> Tel: 073 5226191
> Fax: 073 5226196
>
>
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/6ccc974b/attachment.html>

From reijnders at tor.nl  Tue May 13 22:12:00 2014
From: reijnders at tor.nl (Tom Reijnders)
Date: Tue, 13 May 2014 23:12:00 +0200
Subject: [Xerte-dev] Re: Various pull requests on github (sqlite support,
 security fixes, export zip fixes).
In-Reply-To: <CAO2PinP7bSMnkYLxWyTsUpuR831MuqOAqTmKQN=v2GOcNcFJyA@mail.gmail.com>
References: <5370EEA1.2080803@palepurple.co.uk>	<2b5b42f5-cf4c-49de-ba90-ecf62ae0e76f@email.android.com>	<53727A6B.7030107@tor.nl>
	<CAO2PinP7bSMnkYLxWyTsUpuR831MuqOAqTmKQN=v2GOcNcFJyA@mail.gmail.com>
Message-ID: <53728AA0.5020109@tor.nl>

Ah, good point. Don't know.

Pat L (pgogy) schreef op 13-5-2014 22:40:
> Is the session handler used at all?
>
>
> On Tue, May 13, 2014 at 9:02 PM, Tom Reijnders <reijnders at tor.nl 
> <mailto:reijnders at tor.nl>> wrote:
>
>     Done.
>
>     Why does session_handler still use mysql_connect? I just happen to
>     see that one.
>
>     Tom
>
>     Tom Reijnders schreef op 13-5-2014 11:24:
>>     Thank you David! I'll try to take care of it tonight.
>>
>>     Tom
>>
>>     On 12 mei 2014 17:54:09 CEST, David Goodwin
>>     <david at palepurple.co.uk> <mailto:david at palepurple.co.uk> wrote:
>>
>>         Hi,
>>
>>         See :
>>
>>         https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>>
>>            * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>>            * removes the requirement to have mysql - it is defaulting to sqlite -
>>         this ought to make it easier for people to test.
>>            * removes any need to have a 'setup/' installer (to a large part).
>>         Chucking the files into a web server and browsing to the install should
>>         trigger it to create an empty database in the system temp dir.
>>
>>
>>         See :
>>
>>         https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>>
>>            * When I browse away from the front page, or try and reload it,
>>         Firefox get's stuck in a refresh loop.
>>
>>
>>         and :
>>
>>         See :
>>
>>         https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>>
>>            * Improve the storing of extra_flags against a template.
>>            * Add new ZIP export module (use php's ZIP extension if possible,
>>         otherwise fall back to the PHP userspace library).
>>            * this should resolve memory limit issues with exporting zip files.
>>
>>
>>         thanks
>>         David.
>>
>>
>>     -- 
>>     Verzonden van mijn Android telefoon met K-9 Mail.
>>
>>     This message and any attachment are intended solely for the
>>     addressee and may contain confidential information. If you have
>>     received this message in error, please send it back to me, and
>>     immediately delete it.   Please do not use, copy or disclose the
>>     information contained in this message or in any attachment.  Any
>>     views or opinions expressed by the author of this email do not
>>     necessarily reflect the views of the University of Nottingham.
>>
>>     This message has been checked for viruses but the contents of an
>>     attachment may still contain software viruses which could damage
>>     your computer system, you are advised to perform your own checks.
>>     Email communications with the University of Nottingham may be
>>     monitored as permitted by UK legislation.
>>
>>
>>
>>
>>     _______________________________________________
>>     Xerte-dev mailing list
>>     Xerte-dev at lists.nottingham.ac.uk  <mailto:Xerte-dev at lists.nottingham.ac.uk>
>>     http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>     -- 
>     --
>
>     Tom Reijnders
>     TOR Informatica
>     Chopinlaan 27
>     5242HM Rosmalen
>     Tel: 073 5226191
>     Fax: 073 5226196
>
>
>     This message and any attachment are intended solely for the
>     addressee and may contain confidential information. If you have
>     received this message in error, please send it back to me, and
>     immediately delete it.   Please do not use, copy or disclose the
>     information contained in this message or in any attachment.  Any
>     views or opinions expressed by the author of this email do not
>     necessarily reflect the views of the University of Nottingham.
>
>     This message has been checked for viruses but the contents of an
>     attachment may still contain software viruses which could damage
>     your computer system, you are advised to perform your own checks.
>     Email communications with the University of Nottingham may be
>     monitored as permitted by UK legislation.
>
>
>
>     _______________________________________________
>     Xerte-dev mailing list
>     Xerte-dev at lists.nottingham.ac.uk
>     <mailto:Xerte-dev at lists.nottingham.ac.uk>
>     http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

-- 
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/8169cd86/attachment-0001.html>

From david at palepurple.co.uk  Tue May 13 22:12:27 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Tue, 13 May 2014 22:12:27 +0100
Subject: [Xerte-dev] session_handler.php / template saving / file upload
	mime checking
Message-ID: <53728ABB.8090007@palepurple.co.uk>



Pat L (pgogy) wrote:
>
> Is the session handler used at all?
>
>



It seems to just remain as a commented out part of the config.php file 
(I think).

I just did a 'ack-grep mysql_ --php' to try and find files which 
contained mysql isms.


I'm currently working on improving the uploading of files (av scanning, 
mime type checking) and saving of the templates (ensure the XML is valid)

 From a template saving point of view, can someone explain what the 
difference is between :

modules/xerte/upload/save.php

and

modules/site/upload/save.php

They're almost identical, and it would make sense for the one to just 
require_once(the.other). If so, which way around?



thanks
David.


> Pat L (pgogy) <mailto:xerte at pgogywebstuff.com>
> 13 May 2014 21:40
> Is the session handler used at all?
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> Tom Reijnders <mailto:reijnders at tor.nl>
> 13 May 2014 21:02
> Done.
>
> Why does session_handler still use mysql_connect? I just happen to see 
> that one.
>
> Tom
>
> Tom Reijnders schreef op 13-5-2014 11:24:
>
> -- 
> --
>
> Tom Reijnders
> TOR Informatica
> Chopinlaan 27
> 5242HM Rosmalen
> Tel: 073 5226191
> Fax: 073 5226196
>
>
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this 
> message in error, please send it back to me, and immediately delete 
> it.   Please do not use, copy or disclose the information contained in 
> this message or in any attachment.  Any views or opinions expressed by 
> the author of this email do not necessarily reflect the views of the 
> University of Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as 
> permitted by UK legislation.
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> Tom Reijnders <mailto:reijnders at tor.nl>
> 13 May 2014 10:24
> Thank you David! I'll try to take care of it tonight.
>
> Tom
>
>
> -- 
> Verzonden van mijn Android telefoon met K-9 Mail.
>
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this 
> message in error, please send it back to me, and immediately delete 
> it.   Please do not use, copy or disclose the information contained in 
> this message or in any attachment.  Any views or opinions expressed by 
> the author of this email do not necessarily reflect the views of the 
> University of Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as 
> permitted by UK legislation.
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
> David Goodwin <mailto:david at palepurple.co.uk>
> 12 May 2014 16:54
> Hi,
>
> See :
>
> https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>
>  * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>  * removes the requirement to have mysql - it is defaulting to sqlite 
> - this ought to make it easier for people to test.
>  * removes any need to have a 'setup/' installer (to a large part). 
> Chucking the files into a web server and browsing to the install 
> should trigger it to create an empty database in the system temp dir.
>
>
> See :
>
> https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>
>  * When I browse away from the front page, or try and reload it, 
> Firefox get's stuck in a refresh loop.
>
>
> and :
>
> See :
>
> https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>
>  * Improve the storing of extra_flags against a template.
>  * Add new ZIP export module (use php's ZIP extension if possible, 
> otherwise fall back to the PHP userspace library).
>  * this should resolve memory limit issues with exporting zip files.
>
>
> thanks
> David.
>

-- 
--
Pale Purple Ltd
http://www.palepurple.co.uk @PalePurpleLtd
0845 0046746 / 07792 380669
"Bespoke web and mobile application development"
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/b3269e7e/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: compose-unknown-contact.jpg
Type: image/jpeg
Size: 770 bytes
Desc: not available
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/b3269e7e/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: postbox-contact.jpg
Type: image/jpeg
Size: 1169 bytes
Desc: not available
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/b3269e7e/attachment-0001.jpg>

From xerte at pgogywebstuff.com  Tue May 13 23:22:36 2014
From: xerte at pgogywebstuff.com (Pat L (pgogy))
Date: Tue, 13 May 2014 23:22:36 +0100
Subject: [Xerte-dev] Re: session_handler.php / template saving / file upload
	mime checking
In-Reply-To: <53728ABB.8090007@palepurple.co.uk>
References: <53728ABB.8090007@palepurple.co.uk>
Message-ID: <CAO2PinMYdUX1vRoqp+SPBVkQ2gCknFxro4DZtpy+QugE7KTfRg@mail.gmail.com>

Not sure if module inheritance ever checked on save.php existing
It should do if cleaned up.
I would say keep modules/xerte
then see how we can make site work without it


On Tue, May 13, 2014 at 10:12 PM, David Goodwin <david at palepurple.co.uk>wrote:

>
>
> Pat L (pgogy) wrote:
>
>
> Is the session handler used at all?
>
>
>
>
>
> It seems to just remain as a commented out part of the config.php file (I
> think).
>
> I just did a 'ack-grep mysql_ --php' to try and find files which contained
> mysql isms.
>
>
> I'm currently working on improving the uploading of files (av scanning,
> mime type checking) and saving of the templates (ensure the XML is valid)
>
> >From a template saving point of view, can someone explain what the
> difference is between :
>
> modules/xerte/upload/save.php
>
> and
>
> modules/site/upload/save.php
>
> They're almost identical, and it would make sense for the one to just
> require_once(the.other). If so, which way around?
>
>
>
> thanks
> David.
>
>
>  Pat L (pgogy) <xerte at pgogywebstuff.com>
>  13 May 2014 21:40
> Is the session handler used at all?
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>  Tom Reijnders <reijnders at tor.nl>
>  13 May 2014 21:02
> Done.
>
> Why does session_handler still use mysql_connect? I just happen to see
> that one.
>
> Tom
>
> Tom Reijnders schreef op 13-5-2014 11:24:
>
> --
> --
>
> Tom Reijnders
> TOR Informatica
> Chopinlaan 27
> 5242HM Rosmalen
> Tel: 073 5226191
> Fax: 073 5226196
>
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>  Tom Reijnders <reijnders at tor.nl>
>  13 May 2014 10:24
> Thank you David! I'll try to take care of it tonight.
>
> Tom
>
>
> --
> Verzonden van mijn Android telefoon met K-9 Mail.
>
> This message and any attachment are intended solely for the addressee and
> may contain confidential information. If you have received this message in
> error, please send it back to me, and immediately delete it.   Please do
> not use, copy or disclose the information contained in this message or in
> any attachment.  Any views or opinions expressed by the author of this
> email do not necessarily reflect the views of the University of Nottingham.
>
> This message has been checked for viruses but the contents of an
> attachment may still contain software viruses which could damage your
> computer system, you are advised to perform your own checks. Email
> communications with the University of Nottingham may be monitored as
> permitted by UK legislation.
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>  David Goodwin <david at palepurple.co.uk>
>  12 May 2014 16:54
> Hi,
>
> See :
>
> https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>
>  * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>  * removes the requirement to have mysql - it is defaulting to sqlite -
> this ought to make it easier for people to test.
>  * removes any need to have a 'setup/' installer (to a large part).
> Chucking the files into a web server and browsing to the install should
> trigger it to create an empty database in the system temp dir.
>
>
> See :
>
> https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>
>  * When I browse away from the front page, or try and reload it, Firefox
> get's stuck in a refresh loop.
>
>
> and :
>
> See :
>
> https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>
>  * Improve the storing of extra_flags against a template.
>  * Add new ZIP export module (use php's ZIP extension if possible,
> otherwise fall back to the PHP userspace library).
>  * this should resolve memory limit issues with exporting zip files.
>
>
> thanks
> David.
>
>
> --
> --
> Pale Purple Ltd
> http://www.palepurple.co.uk @PalePurpleLtd
> 0845 0046746 / 07792 380669
> "Bespoke web and mobile application development"
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/ef967260/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: postbox-contact.jpg
Type: image/jpeg
Size: 1169 bytes
Desc: not available
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/ef967260/attachment-0002.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: compose-unknown-contact.jpg
Type: image/jpeg
Size: 770 bytes
Desc: not available
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140513/ef967260/attachment-0003.jpg>

From reijnders at tor.nl  Wed May 14 08:47:13 2014
From: reijnders at tor.nl (Tom Reijnders)
Date: Wed, 14 May 2014 09:47:13 +0200
Subject: [Xerte-dev] Re: Various pull requests on github (sqlite support,
 security fixes, export zip fixes).
In-Reply-To: <5370EEA1.2080803@palepurple.co.uk>
References: <5370EEA1.2080803@palepurple.co.uk>
Message-ID: <53731F81.2070405@tor.nl>

David,

After the merge my database_library.php still uses mysql_ calls. Is that 
correct, or did something go wrong during the merge?
It's easy enough to update now, but I was just wondering.

Tom

David Goodwin schreef op 12-5-2014 17:54:
> Hi,
>
> See :
>
> https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>
>  * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>  * removes the requirement to have mysql - it is defaulting to sqlite 
> - this ought to make it easier for people to test.
>  * removes any need to have a 'setup/' installer (to a large part). 
> Chucking the files into a web server and browsing to the install 
> should trigger it to create an empty database in the system temp dir.
>
>
> See :
>
> https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>
>  * When I browse away from the front page, or try and reload it, 
> Firefox get's stuck in a refresh loop.
>
>
> and :
>
> See :
>
> https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>
>  * Improve the storing of extra_flags against a template.
>  * Add new ZIP export module (use php's ZIP extension if possible, 
> otherwise fall back to the PHP userspace library).
>  * this should resolve memory limit issues with exporting zip files.
>
>
> thanks
> David.
>

-- 
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196



From david at palepurple.co.uk  Wed May 14 09:02:54 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Wed, 14 May 2014 09:02:54 +0100
Subject: [Xerte-dev] Re: Various pull requests on github (sqlite support,
 security fixes, export zip fixes).
In-Reply-To: <53731F81.2070405@tor.nl>
References: <5370EEA1.2080803@palepurple.co.uk> <53731F81.2070405@tor.nl>
Message-ID: <5373232E.3050704@palepurple.co.uk>

Hi - Yes, database_library.php is the only place that has mysql_ calls.

There should also be a database_library_sqlite.php.

It's not particularly elegant, but the code should use one or the other 
to show it switching between the two.

I wasn't sure if it could be rewritten to use PDO or not (in that, I 
wasn't sure if PDO would be an acceptable solution).

David.

Tom Reijnders wrote:
>
> David,
>
> After the merge my database_library.php still uses mysql_ calls. Is
> that correct, or did something go wrong during the merge?
> It's easy enough to update now, but I was just wondering.
>
> Tom
>
> David Goodwin schreef op 12-5-2014 17:54:
>>
>> Hi,
>>
>> See :
>>
>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>>
>> * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>> * removes the requirement to have mysql - it is defaulting to sqlite
>> - this ought to make it easier for people to test.
>> * removes any need to have a 'setup/' installer (to a large part).
>> Chucking the files into a web server and browsing to the install
>> should trigger it to create an empty database in the system temp dir.
>>
>>
>> See :
>>
>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>>
>> * When I browse away from the front page, or try and reload it,
>> Firefox get's stuck in a refresh loop.
>>
>>
>> and :
>>
>> See :
>>
>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>>
>> * Improve the storing of extra_flags against a template.
>> * Add new ZIP export module (use php's ZIP extension if possible,
>> otherwise fall back to the PHP userspace library).
>> * this should resolve memory limit issues with exporting zip files.
>>
>>
>> thanks
>> David.
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140514/260b5704/attachment.html>

From reijnders at tor.nl  Wed May 14 09:10:11 2014
From: reijnders at tor.nl (Tom Reijnders)
Date: Wed, 14 May 2014 10:10:11 +0200
Subject: [Xerte-dev] Re: Various pull requests on github (sqlite support,
 security fixes, export zip fixes).
In-Reply-To: <5373232E.3050704@palepurple.co.uk>
References: <5370EEA1.2080803@palepurple.co.uk> <53731F81.2070405@tor.nl>
	<5373232E.3050704@palepurple.co.uk>
Message-ID: <537324E3.9080607@tor.nl>

Ah. Thanks.

I am not sure either, buy I am in favour of using PDO.
Certainly if there is also a sqlite solution for people that don't care.
We can even add a 3rd file for mysqli....

Thanks for all this work. It was on my list, but simply didn't have the 
time.

Tom


David Goodwin schreef op 14-5-2014 10:02:
> Hi - Yes, database_library.php is the only place that has mysql_ calls.
>
> There should also be a database_library_sqlite.php.
>
> It's not particularly elegant, but the code should use one or the 
> other to show it switching between the two.
>
> I wasn't sure if it could be rewritten to use PDO or not (in that, I 
> wasn't sure if PDO would be an acceptable solution).
>
> David.
>
> Tom Reijnders wrote:
>>
>> David,
>>
>> After the merge my database_library.php still uses mysql_ calls. Is
>> that correct, or did something go wrong during the merge?
>> It's easy enough to update now, but I was just wondering.
>>
>> Tom
>>
>> David Goodwin schreef op 12-5-2014 17:54:
>>>
>>> Hi,
>>>
>>> See :
>>>
>>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>>>
>>> * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>>> * removes the requirement to have mysql - it is defaulting to sqlite
>>> - this ought to make it easier for people to test.
>>> * removes any need to have a 'setup/' installer (to a large part).
>>> Chucking the files into a web server and browsing to the install
>>> should trigger it to create an empty database in the system temp dir.
>>>
>>>
>>> See :
>>>
>>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>>>
>>> * When I browse away from the front page, or try and reload it,
>>> Firefox get's stuck in a refresh loop.
>>>
>>>
>>> and :
>>>
>>> See :
>>>
>>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>>>
>>> * Improve the storing of extra_flags against a template.
>>> * Add new ZIP export module (use php's ZIP extension if possible,
>>> otherwise fall back to the PHP userspace library).
>>> * this should resolve memory limit issues with exporting zip files.
>>>
>>>
>>> thanks
>>> David.
>>>
>>
>
> This message and any attachment are intended solely for the addressee 
> and may contain confidential information. If you have received this 
> message in error, please send it back to me, and immediately delete 
> it.   Please do not use, copy or disclose the information contained in 
> this message or in any attachment.  Any views or opinions expressed by 
> the author of this email do not necessarily reflect the views of the 
> University of Nottingham.
>
> This message has been checked for viruses but the contents of an 
> attachment may still contain software viruses which could damage your 
> computer system, you are advised to perform your own checks. Email 
> communications with the University of Nottingham may be monitored as 
> permitted by UK legislation.
>
>
>
>
> _______________________________________________
> Xerte-dev mailing list
> Xerte-dev at lists.nottingham.ac.uk
> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

-- 
--

Tom Reijnders
TOR Informatica
Chopinlaan 27
5242HM Rosmalen
Tel: 073 5226191
Fax: 073 5226196

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140514/f150845f/attachment.html>

From xerte at pgogywebstuff.com  Wed May 14 09:32:14 2014
From: xerte at pgogywebstuff.com (Pat L (pgogy))
Date: Wed, 14 May 2014 09:32:14 +0100
Subject: [Xerte-dev] Re: Various pull requests on github (sqlite support,
 security fixes, export zip fixes).
In-Reply-To: <53728AA0.5020109@tor.nl>
References: <5370EEA1.2080803@palepurple.co.uk>
	<2b5b42f5-cf4c-49de-ba90-ecf62ae0e76f@email.android.com>
	<53727A6B.7030107@tor.nl>
	<CAO2PinP7bSMnkYLxWyTsUpuR831MuqOAqTmKQN=v2GOcNcFJyA@mail.gmail.com>
	<53728AA0.5020109@tor.nl>
Message-ID: <CAO2PinOewrq+uvFdrKd+SURCBv8FBkynMfoSBFZPNzk_tM45jQ@mail.gmail.com>

I added it a week before I left nottingham as database sessions seemed to
last longer. However in the last week, Nottingham upgraded php and it
killed db sessions so I turned it off
On 13 May 2014 22:12, "Tom Reijnders" <reijnders at tor.nl> wrote:

>  Ah, good point. Don't know.
>
> Pat L (pgogy) schreef op 13-5-2014 22:40:
>
> Is the session handler used at all?
>
>
> On Tue, May 13, 2014 at 9:02 PM, Tom Reijnders <reijnders at tor.nl> wrote:
>
>>  Done.
>>
>> Why does session_handler still use mysql_connect? I just happen to see
>> that one.
>>
>> Tom
>>
>> Tom Reijnders schreef op 13-5-2014 11:24:
>>
>> Thank you David! I'll try to take care of it tonight.
>>
>> Tom
>>
>> On 12 mei 2014 17:54:09 CEST, David Goodwin <david at palepurple.co.uk><david at palepurple.co.uk>wrote:
>>>
>>> Hi,
>>>
>>> See :
>>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/65
>>>
>>>   * changes XOT to break it's dependency on mysql_ (outside of 'setup')
>>>   * removes the requirement to have mysql - it is defaulting to sqlite -
>>> this ought to make it easier for people to test.
>>>   * removes any need to have a 'setup/' installer (to a large part).
>>> Chucking the files into a web server and browsing to the install should
>>> trigger it to create an empty database in the system temp dir.
>>>
>>>
>>> See :
>>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/66
>>>
>>>   * When I browse away from the front page, or try and reload it,
>>> Firefox get's stuck in a refresh loop.
>>>
>>>
>>> and :
>>>
>>> See :
>>> https://github.com/thexerteproject/xerteonlinetoolkits/pull/57
>>>
>>>   * Improve the storing of extra_flags against a template.
>>>   * Add new ZIP export module (use php's ZIP extension if possible,
>>> otherwise fall back to the PHP userspace library).
>>>   * this should resolve memory limit issues with exporting zip files.
>>>
>>>
>>> thanks
>>> David.
>>>
>>>
>> --
>> Verzonden van mijn Android telefoon met K-9 Mail.
>>
>> This message and any attachment are intended solely for the addressee and
>> may contain confidential information. If you have received this message in
>> error, please send it back to me, and immediately delete it.   Please do
>> not use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>>
>>
>>
>> _______________________________________________
>> Xerte-dev mailing listXerte-dev at lists.nottingham.ac.ukhttp://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>>
>>
>> --
>> --
>>
>> Tom Reijnders
>> TOR Informatica
>> Chopinlaan 27
>> 5242HM Rosmalen
>> Tel: 073 5226191
>> Fax: 073 5226196
>>
>>
>>
>> This message and any attachment are intended solely for the addressee and
>> may contain confidential information. If you have received this message in
>> error, please send it back to me, and immediately delete it.   Please do
>> not use, copy or disclose the information contained in this message or in
>> any attachment.  Any views or opinions expressed by the author of this
>> email do not necessarily reflect the views of the University of Nottingham.
>>
>> This message has been checked for viruses but the contents of an
>> attachment may still contain software viruses which could damage your
>> computer system, you are advised to perform your own checks. Email
>> communications with the University of Nottingham may be monitored as
>> permitted by UK legislation.
>>
>>
>> _______________________________________________
>> Xerte-dev mailing list
>> Xerte-dev at lists.nottingham.ac.uk
>> http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>>
>>
>
>
> _______________________________________________
> Xerte-dev mailing listXerte-dev at lists.nottingham.ac.ukhttp://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
>
>
> --
> --
>
> Tom Reijnders
> TOR Informatica
> Chopinlaan 27
> 5242HM Rosmalen
> Tel: 073 5226191
> Fax: 073 5226196
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140514/b945b073/attachment-0001.html>

From david at palepurple.co.uk  Thu May 15 11:21:19 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Thu, 15 May 2014 11:21:19 +0100
Subject: [Xerte-dev] Video playback quality in html5 LOs.
Message-ID: <5374951F.7070803@palepurple.co.uk>

Hi,

We've been asked to improve the quality/behaviour of video playback 
within html5 LOs.


I assume :

We shouldn't think about including a flash fallback player (given 
they're using a html5 template).



Therefore :

We should make sure there's a webm and mp4 format video available for 
the video playback.

and

We should re-encode the mp4 video to h264 basic profile to ensure it 
plays on mobile devices.




Does the data.xml/preview.xml file have an XSD to validate against - and 
specify which attributes/tags are legal?


Possible solutions are :



1. We could embed additional stuff in the xml files -

<?xml version="1.0"?>
<learningObject language="en-GB" name="Learning Object Title" 
navigation="Linear" textSize="12" displayMode="default">
.....
   <textVideo linkID="PG1400145368211" name="video test" align="Left" 
url="path/to/file" tip="Enter a Tooltipx" reset="true"
url_webm="path/to/file.webm" url_ogg="path/to/ogg"

transcriptTabTxt="Transcript" transcriptTabTip="Transcript" 
transcriptTabWidth="90"><![CDATA[asdasdasd asdasd asda sdas vide ov ideo 
video]]></textVideo>

....

(note the url_webm, url_ogg attributes).


Cons - presumably would get lost on re-edit - we'd need to have a 
post-save filter in XOT to add it into the xml files all the time.


Cons - attributes may not be legal/allowable.

Pros - we could add those attributes in, only if the underlying file 
exists - i.e. it's been transcoded in the background.




2. We could alter the html5 playback JavaScript to cause it to check for 
the existence of whatever the textVideo url attribute is pointing to 
plus a set of extensions which would be used to indicate that the file 
has been transcoded.

e.g. <textVideo url="foo" ... />

JS would check for :

foo.webm
foo.mp4
foo.ogg

And if it finds any of them, it'll use them. (http head request - to 
check for existence).

Pros : no need to touch data.xml/preview.xml, url attribute always 
points to what the user uploaded.

Pros : a simple cron job (perhaps) could rummage through USER-FILES 
looking for video files which haven't been transcoded. If the cron job 
is never run then the original file will always be used, if the cron job 
has run then hopefully there are some better files to use.


Cons : "magical"



Any other ideas/thoughts?

I don't suppose it's really feasible to just tell the template author 
they should upload the correct encoding - when they're unlikely to be 
aware of the nuances of video encoding, or the tools available for it?



thanks,

David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From Julian.Tenney at nottingham.ac.uk  Thu May 15 11:25:20 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Thu, 15 May 2014 11:25:20 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <5374951F.7070803@palepurple.co.uk>
References: <5374951F.7070803@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>

When we went through all this, the most important thing was that the user only had to upload one file: I think that's important, so would look for the automagical solution.

Any automagical solution would need to be licensed appropriately for redistribution.

I do wonder if the quality thing is down to the quality of the video uploaded (silk purse and sows ear?).


-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 15 May 2014 11:21
To: For Xerte technical developers
Subject: [Xerte-dev] Video playback quality in html5 LOs.

Hi,

We've been asked to improve the quality/behaviour of video playback within html5 LOs.


I assume :

We shouldn't think about including a flash fallback player (given 
they're using a html5 template).



Therefore :

We should make sure there's a webm and mp4 format video available for 
the video playback.

and

We should re-encode the mp4 video to h264 basic profile to ensure it 
plays on mobile devices.




Does the data.xml/preview.xml file have an XSD to validate against - and 
specify which attributes/tags are legal?


Possible solutions are :



1. We could embed additional stuff in the xml files -

<?xml version="1.0"?>
<learningObject language="en-GB" name="Learning Object Title" 
navigation="Linear" textSize="12" displayMode="default">
.....
   <textVideo linkID="PG1400145368211" name="video test" align="Left" 
url="path/to/file" tip="Enter a Tooltipx" reset="true"
url_webm="path/to/file.webm" url_ogg="path/to/ogg"

transcriptTabTxt="Transcript" transcriptTabTip="Transcript" 
transcriptTabWidth="90"><![CDATA[asdasdasd asdasd asda sdas vide ov ideo 
video]]></textVideo>

....

(note the url_webm, url_ogg attributes).


Cons - presumably would get lost on re-edit - we'd need to have a 
post-save filter in XOT to add it into the xml files all the time.


Cons - attributes may not be legal/allowable.

Pros - we could add those attributes in, only if the underlying file 
exists - i.e. it's been transcoded in the background.




2. We could alter the html5 playback JavaScript to cause it to check for 
the existence of whatever the textVideo url attribute is pointing to 
plus a set of extensions which would be used to indicate that the file 
has been transcoded.

e.g. <textVideo url="foo" ... />

JS would check for :

foo.webm
foo.mp4
foo.ogg

And if it finds any of them, it'll use them. (http head request - to 
check for existence).

Pros : no need to touch data.xml/preview.xml, url attribute always 
points to what the user uploaded.

Pros : a simple cron job (perhaps) could rummage through USER-FILES 
looking for video files which haven't been transcoded. If the cron job 
is never run then the original file will always be used, if the cron job 
has run then hopefully there are some better files to use.


Cons : "magical"



Any other ideas/thoughts?

I don't suppose it's really feasible to just tell the template author 
they should upload the correct encoding - when they're unlikely to be 
aware of the nuances of video encoding, or the tools available for it?



thanks,

David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From ronm at mitchellmedia.co.uk  Thu May 15 11:57:03 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Thu, 15 May 2014 11:57:03 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>
References: <5374951F.7070803@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <005501cf702c$68fdd010$3af97030$@co.uk>

and surely the first point of call anyway is where possible to upload to a
streaming solution and embed from there.

I know that's not always possible and that there are page types which
currently only work with directly uploaded video's or direct link's to .mp4
but still I see to many examples where authors are using the easy option for
them by uploading direct rather than the best option for end users
especially on slower connections etc.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 15 May 2014 11:25
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.

When we went through all this, the most important thing was that the user
only had to upload one file: I think that's important, so would look for the
automagical solution.

Any automagical solution would need to be licensed appropriately for
redistribution.

I do wonder if the quality thing is down to the quality of the video
uploaded (silk purse and sows ear?).


-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 15 May 2014 11:21
To: For Xerte technical developers
Subject: [Xerte-dev] Video playback quality in html5 LOs.

Hi,

We've been asked to improve the quality/behaviour of video playback within
html5 LOs.


I assume :

We shouldn't think about including a flash fallback player (given they're
using a html5 template).



Therefore :

We should make sure there's a webm and mp4 format video available for 
the video playback.

and

We should re-encode the mp4 video to h264 basic profile to ensure it 
plays on mobile devices.




Does the data.xml/preview.xml file have an XSD to validate against - and 
specify which attributes/tags are legal?


Possible solutions are :



1. We could embed additional stuff in the xml files -

<?xml version="1.0"?>
<learningObject language="en-GB" name="Learning Object Title" 
navigation="Linear" textSize="12" displayMode="default">
.....
   <textVideo linkID="PG1400145368211" name="video test" align="Left" 
url="path/to/file" tip="Enter a Tooltipx" reset="true"
url_webm="path/to/file.webm" url_ogg="path/to/ogg"

transcriptTabTxt="Transcript" transcriptTabTip="Transcript" 
transcriptTabWidth="90"><![CDATA[asdasdasd asdasd asda sdas vide ov ideo 
video]]></textVideo>

....

(note the url_webm, url_ogg attributes).


Cons - presumably would get lost on re-edit - we'd need to have a 
post-save filter in XOT to add it into the xml files all the time.


Cons - attributes may not be legal/allowable.

Pros - we could add those attributes in, only if the underlying file 
exists - i.e. it's been transcoded in the background.




2. We could alter the html5 playback JavaScript to cause it to check for 
the existence of whatever the textVideo url attribute is pointing to 
plus a set of extensions which would be used to indicate that the file 
has been transcoded.

e.g. <textVideo url="foo" ... />

JS would check for :

foo.webm
foo.mp4
foo.ogg

And if it finds any of them, it'll use them. (http head request - to 
check for existence).

Pros : no need to touch data.xml/preview.xml, url attribute always 
points to what the user uploaded.

Pros : a simple cron job (perhaps) could rummage through USER-FILES 
looking for video files which haven't been transcoded. If the cron job 
is never run then the original file will always be used, if the cron job 
has run then hopefully there are some better files to use.


Cons : "magical"



Any other ideas/thoughts?

I don't suppose it's really feasible to just tell the template author 
they should upload the correct encoding - when they're unlikely to be 
aware of the nuances of video encoding, or the tools available for it?



thanks,

David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.






From david at palepurple.co.uk  Thu May 15 12:02:31 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Thu, 15 May 2014 12:02:31 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>
References: <5374951F.7070803@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <53749EC7.8020801@palepurple.co.uk>



On 15/05/14 11:25, Julian Tenney wrote:
> When we went through all this, the most important thing was that the user only had to upload one file: I think that's important, so would look for the automagical solution.
>
> Any automagical solution would need to be licensed appropriately for redistribution.
>
> I do wonder if the quality thing is down to the quality of the video uploaded (silk purse and sows ear?).
>
>


That's an interesting expression. I hadn't come across that before.


Is there a standard/definition for what can be in the LO's data.xml?




We had the issue that the user uploaded a video in flv format, which 
didn't playback. Transcoding it to .mp4 fixed the issue.


Can the editor restrict the file format(s) uploaded at all?

I presume the non-HTML5 variant of XOT required FLV videos?


David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From Julian.Tenney at nottingham.ac.uk  Thu May 15 12:56:50 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Thu, 15 May 2014 12:56:50 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <53749EC7.8020801@palepurple.co.uk>
References: <5374951F.7070803@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>
	<53749EC7.8020801@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE180@EXCHANGE1.ad.nottingham.ac.uk>

Yes, that's why an flv snuck through: the door is open to flv for those still using the flash runtime. But we should get to a point when Flash is dead to us soon.

If you upload high quality video, it will play back high quality as things stand. If you upload poor quality, it doesn't matter how you re-encode it, it will remain poor quality.

If you automagically encode it to different formats, you have to make some big assumptions and there are trade-offs between quality and bandwidth.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 15 May 2014 12:03
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.



On 15/05/14 11:25, Julian Tenney wrote:
> When we went through all this, the most important thing was that the user only had to upload one file: I think that's important, so would look for the automagical solution.
>
> Any automagical solution would need to be licensed appropriately for redistribution.
>
> I do wonder if the quality thing is down to the quality of the video uploaded (silk purse and sows ear?).
>
>


That's an interesting expression. I hadn't come across that before.


Is there a standard/definition for what can be in the LO's data.xml?




We had the issue that the user uploaded a video in flv format, which didn't playback. Transcoding it to .mp4 fixed the issue.


Can the editor restrict the file format(s) uploaded at all?

I presume the non-HTML5 variant of XOT required FLV videos?


David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From Julian.Tenney at nottingham.ac.uk  Thu May 15 12:58:23 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Thu, 15 May 2014 12:58:23 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <5374951F.7070803@palepurple.co.uk>
References: <5374951F.7070803@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE183@EXCHANGE1.ad.nottingham.ac.uk>

> We've been asked to improve the quality/behaviour of video playback within html5 LOs.

What's the actual problem to solve? High quality in -> high quality out in my experience. 


From david at palepurple.co.uk  Thu May 15 13:22:56 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Thu, 15 May 2014 13:22:56 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE180@EXCHANGE1.ad.nottingham.ac.uk>
References: <5374951F.7070803@palepurple.co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>	<53749EC7.8020801@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE180@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <5374B1A0.8070803@palepurple.co.uk>



On 15/05/14 12:56, Julian Tenney wrote:
> Yes, that's why an flv snuck through: the door is open to flv for those still using the flash runtime.But we should get to a point when Flash is dead to us soon.
>
> If you upload high quality video, it will play back high quality as things stand. If you upload poor quality, it doesn't matter how you re-encode it, it will remain poor quality.
>

Yes.

I was planning to upgrade to Hollywood OS next year for the infinite 
magnification and enhancement capabilities.

> If you automagically encode it to different formats, you have to make some big assumptions and there are trade-offs between quality and bandwidth.

Yes. But at least it'd work. I had the impression that the organisation 
produced .flv videos by default - so transcoding to .mp4 would be useful 
to them.

The existing Javascript code attempts to look for an .mp4 to go with an 
.flv, so we'll just write something simple to create a .mp4 if an .flv 
exists with no partner .mp4.



David.


-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From Julian.Tenney at nottingham.ac.uk  Thu May 15 13:38:41 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Thu, 15 May 2014 13:38:41 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <5374B1A0.8070803@palepurple.co.uk>
References: <5374951F.7070803@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>
	<53749EC7.8020801@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE180@EXCHANGE1.ad.nottingham.ac.uk>
	<5374B1A0.8070803@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE1C5@EXCHANGE1.ad.nottingham.ac.uk>

> The existing Javascript code attempts to look for an .mp4 to go with an .flv, so we'll just write something simple to create a .mp4 if an .flv exists with no partner .mp4.

Not sure this is something I'd want to add to the distribution - BTW, .flv will play in the html5 templates as it stands:

http://www.nottingham.ac.uk/toolkits/play_10274



-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 15 May 2014 13:23
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.



On 15/05/14 12:56, Julian Tenney wrote:
> Yes, that's why an flv snuck through: the door is open to flv for those still using the flash runtime.But we should get to a point when Flash is dead to us soon.
>
> If you upload high quality video, it will play back high quality as things stand. If you upload poor quality, it doesn't matter how you re-encode it, it will remain poor quality.
>

Yes.

I was planning to upgrade to Hollywood OS next year for the infinite magnification and enhancement capabilities.

> If you automagically encode it to different formats, you have to make some big assumptions and there are trade-offs between quality and bandwidth.

Yes. But at least it'd work. I had the impression that the organisation produced .flv videos by default - so transcoding to .mp4 would be useful to them.

The existing Javascript code attempts to look for an .mp4 to go with an .flv, so we'll just write something simple to create a .mp4 if an .flv exists with no partner .mp4.



David.


--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From david at palepurple.co.uk  Thu May 15 13:52:49 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Thu, 15 May 2014 13:52:49 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE1C5@EXCHANGE1.ad.nottingham.ac.uk>
References: <5374951F.7070803@palepurple.co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>	<53749EC7.8020801@palepurple.co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE180@EXCHANGE1.ad.nottingham.ac.uk>	<5374B1A0.8070803@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE1C5@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <5374B8A1.4020304@palepurple.co.uk>



On 15/05/14 13:38, Julian Tenney wrote:
>> The existing Javascript code attempts to look for an .mp4 to go with an .flv, so we'll just write something simple to create a .mp4 if an .flv exists with no partner .mp4.
>
> Not sure this is something I'd want to add to the distribution - BTW, .flv will play in the html5 templates as it stands:
>

That loads through flash though - which is problematic for mobile devices.

I am 99% sure that if you were to put a .mp4 of video.flv at :

http://www.nottingham.ac.uk/toolkits/USER-FILES/10274-cczjrt-Nottingham/media/

i.e.

http://www.nottingham.ac.uk/toolkits/USER-FILES/10274-cczjrt-Nottingham/media/video.mp4


then the .mp4 will be played through the js etc.


This isn't (as far as I can tell) new behaviour.

David.


-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From Julian.Tenney at nottingham.ac.uk  Thu May 15 14:03:24 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Thu, 15 May 2014 14:03:24 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <5374B8A1.4020304@palepurple.co.uk>
References: <5374951F.7070803@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>
	<53749EC7.8020801@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE180@EXCHANGE1.ad.nottingham.ac.uk>
	<5374B1A0.8070803@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE1C5@EXCHANGE1.ad.nottingham.ac.uk>
	<5374B8A1.4020304@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE209@EXCHANGE1.ad.nottingham.ac.uk>

Fair point, yes that will fallback to the flash player.

But if you care about mobile devices, why are you doing .flv?

(I do see what you're getting at ;-) )

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 15 May 2014 13:53
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.



On 15/05/14 13:38, Julian Tenney wrote:
>> The existing Javascript code attempts to look for an .mp4 to go with an .flv, so we'll just write something simple to create a .mp4 if an .flv exists with no partner .mp4.
>
> Not sure this is something I'd want to add to the distribution - BTW, .flv will play in the html5 templates as it stands:
>

That loads through flash though - which is problematic for mobile devices.

I am 99% sure that if you were to put a .mp4 of video.flv at :

http://www.nottingham.ac.uk/toolkits/USER-FILES/10274-cczjrt-Nottingham/media/

i.e.

http://www.nottingham.ac.uk/toolkits/USER-FILES/10274-cczjrt-Nottingham/media/video.mp4


then the .mp4 will be played through the js etc.


This isn't (as far as I can tell) new behaviour.

David.


-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From david at palepurple.co.uk  Thu May 15 14:15:55 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Thu, 15 May 2014 14:15:55 +0100
Subject: [Xerte-dev] Re: Video playback quality in html5 LOs.
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1AEDE209@EXCHANGE1.ad.nottingham.ac.uk>
References: <5374951F.7070803@palepurple.co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE0C7@EXCHANGE1.ad.nottingham.ac.uk>	<53749EC7.8020801@palepurple.co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE180@EXCHANGE1.ad.nottingham.ac.uk>	<5374B1A0.8070803@palepurple.co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE1C5@EXCHANGE1.ad.nottingham.ac.uk>	<5374B8A1.4020304@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE209@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <5374BE0B.8080006@palepurple.co.uk>



On 15/05/14 14:03, Julian Tenney wrote:
> Fair point, yes that will fallback to the flash player.
>
> But if you care about mobile devices, why are you doing .flv?

Ha!

Well, obviously (?!) I wasn't the one trying to create the LO.
Whoever was, had uploading an FLV file and then (if memory serves me 
correctly) found that it didn't render on an iPad.

(I think it was a case of them doing a straight migration from using 
flash-based LOs to the shiny new HTML5 variant...)


>
> (I do see what you're getting at ;-) )

:)


David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From david at palepurple.co.uk  Mon May 19 14:39:42 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Mon, 19 May 2014 14:39:42 +0100
Subject: [Xerte-dev] Re: PHP compatibility
In-Reply-To: <E714C4E8-0938-45AD-9762-106FB920B62B@palepurple.co.uk>
References: <007801cec98d$cab9d0b0$602d7210$@co.uk>	<a0901fca5d74f133f414ac6caf5819ed@server.tor.nl>	<00ea01cec99d$0bf9c5d0$23ed5170$@co.uk>	<CAO2PinO7yM8ROfw+_GWhBsPj5CZM1vGFsRKMzomndERbyEM7Hw@mail.gmail.com>
	<E714C4E8-0938-45AD-9762-106FB920B62B@palepurple.co.uk>
Message-ID: <537A099E.4020000@palepurple.co.uk>

>
> I added 'db_query', 'db_query_one' function calls ages ago (see
> https://github.com/thexerteproject/xerteonlinetoolkits/blob/develop/website_code/php/database_library.php )  which
> do try to allow for the migration away from direct use of the PHP mysql
> driver (i.e. you could swap to use mysqli / mysqlnd or PDO or whatever
> easily enough). However, the last time I looked there was plenty of the
> codebase which still had 'mysql_query()' calls (or other mysql_ calls).
>
>

Just for a follow up:


A recent merge into the develop branch should have removed all usages of 
'mysql' functions from everywhere except the 'setup' folder and the 
website_code/php/database_library.php script.

In order to fully drop the dependency on 'mysql' we just need to rewrite 
db_query and db_query_one in database_library.php. This has already been 
done with the database_library-sqlite.php script.

David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From Julian.Tenney at nottingham.ac.uk  Tue May 20 11:14:42 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Tue, 20 May 2014 11:14:42 +0100
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham
In-Reply-To: <965A188BE17FF6418686E864C242C00A4D244D1420@EXCHANGE3.ad.nottingham.ac.uk>
References: <965A188BE17FF6418686E864C242C00A4D244D140F@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE69E@EXCHANGE1.ad.nottingham.ac.uk>
	<965A188BE17FF6418686E864C242C00A4D244D1420@EXCHANGE3.ad.nottingham.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F4102@EXCHANGE1.ad.nottingham.ac.uk>

I can't find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query - I've noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an 'intranet site') and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I've noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don't think it's working - perhaps because it comes after the 'if IE' conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140520/73dc1605/attachment.html>

From ronm at mitchellmedia.co.uk  Tue May 20 11:34:29 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Tue, 20 May 2014 11:34:29 +0100
Subject: [Xerte-dev] Re: FW: Xerte and IE doc mode at Nottingham
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F4102@EXCHANGE1.ad.nottingham.ac.uk>
References: <965A188BE17FF6418686E864C242C00A4D244D140F@EXCHANGE3.ad.nottingham.ac.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE69E@EXCHANGE1.ad.nottingham.ac.uk>	<965A188BE17FF6418686E864C242C00A4D244D1420@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F4102@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <004701cf7417$1582fed0$4088fc70$@co.uk>

Don't know about if statement but that meta line is in at least the
following:

modules\site\preview.php

modules\site\play.php

modules\xerte\player_html5\rloObject.htm

 

HTH

Ron

 

From: xerte-dev-bounces at lists.nottingham.ac.uk
[mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 20 May 2014 11:15
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

 

I can't find the code to have a look a this: where on earth is it? Does this
seem sensible?

 

 

Quick query - I've noticed that Nottingham users viewing Xerte projects
on-site with IE are having problems because IE will run in compatibility
view (as Xerte is an 'intranet site') and document mode often reverts to old
IE7 standards, therefore breaking a lot of the HTML5 functionality.

 

I've noticed that Xerte already has the <meta http-equiv="X-UA-Compatible"
content="IE=edge"> tag which should correct this, but unfortunately I don't
think it's working - perhaps because it comes after the 'if IE' conditions.
Is there any way these could be reordered so that the IE=edge comes right at
the top, like on Moodle? Or alternatively can this be configured server side
via HTTP header? 

 

 

This message and any attachment are intended solely for the addressee and
may contain confidential information. If you have received this message in
error, please send it back to me, and immediately delete it.   Please do not
use, copy or disclose the information contained in this message or in any
attachment.  Any views or opinions expressed by the author of this email do
not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system,
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nottingham.ac.uk/pipermail/xerte-dev/attachments/20140520/cca447f4/attachment.html>

From J.J.Smith at gcu.ac.uk  Tue May 20 11:44:44 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Tue, 20 May 2014 11:44:44 +0100
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F4102@EXCHANGE1.ad.nottingham.ac.uk>
References: <965A188BE17FF6418686E864C242C00A4D244D140F@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE69E@EXCHANGE1.ad.nottingham.ac.uk>
	<965A188BE17FF6418686E864C242C00A4D244D1420@EXCHANGE3.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F4102@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023C@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Thought we were only supporting "modern" browsers? ;-)

Anyway it's here https://github.com/thexerteproject/xerteonlinetoolkits/blob/e18bc9a58b1d63e09c351aa814f88ed2cf7ec827/modules/xerte/player_html5/rloObject.htm#L12

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 11:14
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

I can?t find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query ? I?ve noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an ?intranet site?) and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I?ve noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don?t think it?s working ? perhaps because it comes after the ?if IE? conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number SC021474


From J.J.Smith at gcu.ac.uk  Tue May 20 11:51:36 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Tue, 20 May 2014 11:51:36 +0100
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023C@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <965A188BE17FF6418686E864C242C00A4D244D140F@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE69E@EXCHANGE1.ad.nottingham.ac.uk>
	<965A188BE17FF6418686E864C242C00A4D244D1420@EXCHANGE3.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F4102@EXCHANGE1.ad.nottingham.ac.uk>,
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023C@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023D@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Looking here http://stackoverflow.com/questions/3726357/why-does-ie9-switch-to-compatibility-mode-on-my-website it seems that IE9 can and often will revert to compatibility mode for Intranet sites because thats how its been setup to operate...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 11:44
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

Thought we were only supporting "modern" browsers? ;-)

Anyway it's here https://github.com/thexerteproject/xerteonlinetoolkits/blob/e18bc9a58b1d63e09c351aa814f88ed2cf7ec827/modules/xerte/player_html5/rloObject.htm#L12

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 11:14
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

I can?t find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query ? I?ve noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an ?intranet site?) and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I?ve noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don?t think it?s working ? perhaps because it comes after the ?if IE? conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474


From J.J.Smith at gcu.ac.uk  Tue May 20 12:01:55 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Tue, 20 May 2014 12:01:55 +0100
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023C@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <965A188BE17FF6418686E864C242C00A4D244D140F@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE69E@EXCHANGE1.ad.nottingham.ac.uk>
	<965A188BE17FF6418686E864C242C00A4D244D1420@EXCHANGE3.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F4102@EXCHANGE1.ad.nottingham.ac.uk>,
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023C@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023E@ITSEMBXCLUS.enterprise.gcal.ac.uk>

It's possible that its the script or utf-8 setting BEFORE the IE=edge setting thats causing it... I've seen some suggest that the IE=edge needs to come first so try changing this:

<head>
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->

        <meta charset="utf-8">
        <meta http-equiv="X-UA-Compatible" content="IE=edge">


to this:

<head>
        <meta http-equiv="X-UA-Compatible" content="IE=edge">
        <meta charset="utf-8">
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 11:44
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

Thought we were only supporting "modern" browsers? ;-)

Anyway it's here https://github.com/thexerteproject/xerteonlinetoolkits/blob/e18bc9a58b1d63e09c351aa814f88ed2cf7ec827/modules/xerte/player_html5/rloObject.htm#L12

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 11:14
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

I can?t find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query ? I?ve noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an ?intranet site?) and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I?ve noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don?t think it?s working ? perhaps because it comes after the ?if IE? conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474


From J.J.Smith at gcu.ac.uk  Tue May 20 12:40:40 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Tue, 20 May 2014 12:40:40 +0100
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023E@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <965A188BE17FF6418686E864C242C00A4D244D140F@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE69E@EXCHANGE1.ad.nottingham.ac.uk>
	<965A188BE17FF6418686E864C242C00A4D244D1420@EXCHANGE3.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F4102@EXCHANGE1.ad.nottingham.ac.uk>,
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023C@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023E@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023F@ITSEMBXCLUS.enterprise.gcal.ac.uk>

OK so the IE conditionals before the doctype might also cause issues but there is a workaround

http://stackoverflow.com/questions/2518256/override-intranet-compatibility-mode-ie8

The relevant section reads that we should change the code to start

<!--[if HTML5]><![endif]-->
<!doctype html>
<!--[if IEMobile 7 ]><html class="no-js iem7"><![endif]-->
<!--[if lt IE 7]><html class="no-js lt-ie9 lt-ie8 lt-ie7 ie6 ie5"><![endif]-->
<!--[if IE 7]><html class="no-js lt-ie9 lt-ie8 ie7"><![endif]-->
<!--[if IE 8]><html class="no-js lt-ie9 ie8"><![endif]-->
<!--[if IE 9]><html class="no-js ie9"><![endif]-->
<!--[if (gt IE 9)|(gt IEMobile 7)|!(IEMobile)|!(IE)]><!--><html class="no-js" lang="en"><!--<![endif]-->
<head>
<!--[if !HTML5]>
        <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1">
<![endif]-->
        <meta charset="utf-8">

        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 12:01
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

It's possible that its the script or utf-8 setting BEFORE the IE=edge setting thats causing it... I've seen some suggest that the IE=edge needs to come first so try changing this:

<head>
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->

        <meta charset="utf-8">
        <meta http-equiv="X-UA-Compatible" content="IE=edge">


to this:

<head>
        <meta http-equiv="X-UA-Compatible" content="IE=edge">
        <meta charset="utf-8">
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 11:44
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

Thought we were only supporting "modern" browsers? ;-)

Anyway it's here https://github.com/thexerteproject/xerteonlinetoolkits/blob/e18bc9a58b1d63e09c351aa814f88ed2cf7ec827/modules/xerte/player_html5/rloObject.htm#L12

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 11:14
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

I can?t find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query ? I?ve noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an ?intranet site?) and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I?ve noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don?t think it?s working ? perhaps because it comes after the ?if IE? conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474


From ronm at mitchellmedia.co.uk  Tue May 20 15:22:17 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Tue, 20 May 2014 15:22:17 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020B@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>	<00ea01cf6dd8$04760280$0d620780$@co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>,
	<01d501cf6df4$c7cb7000$57625000$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020B@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <00d001cf7436$e82be300$b883a900$@co.uk>

Hi all
you may have seen on the community site yet another report of lost work? I'm not sure what installation the user (Susie) is using but I just asked the following questions:

What installation are you using - who looks after it?
What version?
Did you lose the pages after making a change?
Had you pasted any text content into any of the pages e.g. from Word?
Had you added any HTML or custom code to any of the pages?
Were you editing the LO on a slow connection or did you experience any connection problems while editing?

It seems to me this is becoming and increasing occurrence but not sure we're any closer to nailing down the cause or causes.

A student on one of the HEA projects reported this recently and responded as follows:

?No text was copied and pasted,  so I doubt there would be HTML, plus no hyper linking was included....however pages were lost, but only when working or accessing the project at my house anywhere else it 'seemed' to be fine... But I can't be too sure?

This mirrors the only consistent cause Alistair and I were able to identify e.g. he was having connectivity problems and regularly getting corrupt LO's or lost pages. Since his connectivity was fixed I don't think this has happened but I'll check. All that said I've never once had this happen to me even when on trains or using mobile connections including the same codebase or even installations as Alistair.

Even if a new solution is put in place to validate before saving etc it seems to me that there will still be high risk for all those users and installations unlikely to be able to upgrade immediately and would still be good to try to nail down the cause of this.

Has anyone been able to replicate?

How many cases do we collectively know of?

Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 12 May 2014 16:27
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

AFAIK Ron they are just the characters that Word stick in when you type "word" and you get the special opening/closing quotes etc...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell [ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 16:13
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

The occasions I know of it's mostly been impossible to troubleshoot due to lack of detailed info or too late data already lost.

Alistair had regular problems and while we were trying to nail it down (over weeks) too many inconsistencies e.g. different projects, some apparently no copy and paste involved but only seemed to happen with dodgy connection.

The reason I mentioned nested pages is that seems to be common whenever this has happened but may just be coincidental.

I've also seen those ??? characters before but I've seen broken xml without any of that and I think I've seen working LO's which haven't broken and yet had those chars.

Is there a way of reliably generating those sorts of chars to past in to prove these are the cause?

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 15:47
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss


1.      Possibly. He is remote to Nottingham, though has a Nottingham username.

2.      That doesn?t trip any switches with me ? what are you thinking?

Since I found non-UTF chars in the file, I have a strong suspicion that that is the cause: even though he is pasting into notepad from word (most of the time), some are getting through. I have suggested actually saving the file in notepad as UTF 8 to force it to happen.

But this is something I would really like to see fixed. Pasting from word is a fact of life unfortunately.

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Very frustrating replying and seeing loads of other replies sent after mine hitting the list before my reply even reaches! :-(

But anyway just re-reading the message from you user and I wonder?

1. Any likelihood of connection problems being the cause?

2. Wondering if rather than connectors or tab nav whether the problem only occurs with page types that have nested pages?


From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss.

The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself.

Sorry not much help I guess :-(

Guidance example given to one of the HEA projects:
I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.


From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Further Xerte Data Loss


I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters



<cMen



i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?>



Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before?



I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below?



Thanks a lot,



Julian





I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).



I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.



Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?



I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.



My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.



With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.


This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.






From david at palepurple.co.uk  Tue May 20 15:46:48 2014
From: david at palepurple.co.uk (David Goodwin)
Date: Tue, 20 May 2014 15:46:48 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <00d001cf7436$e82be300$b883a900$@co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>	<00ea01cf6dd8$04760280$0d620780$@co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>,
	<01d501cf6df4$c7cb7000$57625000$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020B@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<00d001cf7436$e82be300$b883a900$@co.uk>
Message-ID: <537B6AD8.2010903@palepurple.co.uk>

<snip/>


>
> A student on one of the HEA projects reported this recently and
> responded as follows:
>
> ?No text was copied and pasted,  so I doubt there would be HTML, plus
> no hyper linking was included....however pages were lost, but only
> when working or accessing the project at my house anywhere else it
> 'seemed' to be fine... But I can't be too sure?
>
> This mirrors the only consistent cause Alistair and I were able to
> identify e.g. he was having connectivity problems and regularly
> getting corrupt LO's or lost pages. Since his connectivity was fixed
> I don't think this has happened but I'll check. All that said I've
> never once had this happen to me even when on trains or using mobile
> connections including the same codebase or even installations as
> Alistair.
>
> Even if a new solution is put in place to validate before saving etc
> it seems to me that there will still be high risk for all those users
> and installations unlikely to be able to upgrade immediately and
> would still be good to try to nail down the cause of this.
>

I've certainly seen corruption due to invalid characters being embedded 
into the LO. I can't replicate this, regardless of what sort of 
formatting I embed in, which is a pain.


One of the changes I've committed to github / a pull request is that
there's better validation on the XML. i.e. if it doesn't validate then
XOT won't overwrite what ever is in the
USER-FILES/$whatever/(data|preview).xml file(s).

(via plugins/xml_parsing_check.php when/if merged).


David.

-- 
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746


From Julian.Tenney at nottingham.ac.uk  Tue May 20 16:01:21 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Tue, 20 May 2014 16:01:21 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <537B6AD8.2010903@palepurple.co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>,
	<01d501cf6df4$c7cb7000$57625000$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020B@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<00d001cf7436$e82be300$b883a900$@co.uk>
	<537B6AD8.2010903@palepurple.co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F4361@EXCHANGE1.ad.nottingham.ac.uk>

> One of the changes I've committed to github / a pull request is that there's better validation on the XML. i.e. if it doesn't validate then XOT won't overwrite what ever is in the USER-FILES/$whatever/(data|preview).xml file(s).

Cheers.

It does seem the data loss issue has the remote user as the common factor.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 20 May 2014 15:47
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

<snip/>


>
> A student on one of the HEA projects reported this recently and 
> responded as follows:
>
> ?No text was copied and pasted,  so I doubt there would be HTML, plus 
> no hyper linking was included....however pages were lost, but only 
> when working or accessing the project at my house anywhere else it 
> 'seemed' to be fine... But I can't be too sure?
>
> This mirrors the only consistent cause Alistair and I were able to 
> identify e.g. he was having connectivity problems and regularly 
> getting corrupt LO's or lost pages. Since his connectivity was fixed I 
> don't think this has happened but I'll check. All that said I've never 
> once had this happen to me even when on trains or using mobile 
> connections including the same codebase or even installations as 
> Alistair.
>
> Even if a new solution is put in place to validate before saving etc 
> it seems to me that there will still be high risk for all those users 
> and installations unlikely to be able to upgrade immediately and would 
> still be good to try to nail down the cause of this.
>

I've certainly seen corruption due to invalid characters being embedded into the LO. I can't replicate this, regardless of what sort of formatting I embed in, which is a pain.


One of the changes I've committed to github / a pull request is that there's better validation on the XML. i.e. if it doesn't validate then XOT won't overwrite what ever is in the USER-FILES/$whatever/(data|preview).xml file(s).

(via plugins/xml_parsing_check.php when/if merged).


David.

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

From Julian.Tenney at nottingham.ac.uk  Tue May 20 16:05:41 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Tue, 20 May 2014 16:05:41 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <00d001cf7436$e82be300$b883a900$@co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>,
	<01d501cf6df4$c7cb7000$57625000$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020B@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<00d001cf7436$e82be300$b883a900$@co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F4369@EXCHANGE1.ad.nottingham.ac.uk>

> How many cases do we collectively know of?

Here perhaps half a dozen. Three of those are one user, a remote user, who pastes a lot of stuff from Word. He's savvy though, so I don't think it's a liveware thing. I have never had it happen to me, and cannot replicate it.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 20 May 2014 15:22
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Hi all
you may have seen on the community site yet another report of lost work? I'm not sure what installation the user (Susie) is using but I just asked the following questions:

What installation are you using - who looks after it?
What version?
Did you lose the pages after making a change?
Had you pasted any text content into any of the pages e.g. from Word?
Had you added any HTML or custom code to any of the pages?
Were you editing the LO on a slow connection or did you experience any connection problems while editing?

It seems to me this is becoming and increasing occurrence but not sure we're any closer to nailing down the cause or causes.

A student on one of the HEA projects reported this recently and responded as follows:

?No text was copied and pasted,  so I doubt there would be HTML, plus no hyper linking was included....however pages were lost, but only when working or accessing the project at my house anywhere else it 'seemed' to be fine... But I can't be too sure?

This mirrors the only consistent cause Alistair and I were able to identify e.g. he was having connectivity problems and regularly getting corrupt LO's or lost pages. Since his connectivity was fixed I don't think this has happened but I'll check. All that said I've never once had this happen to me even when on trains or using mobile connections including the same codebase or even installations as Alistair.

Even if a new solution is put in place to validate before saving etc it seems to me that there will still be high risk for all those users and installations unlikely to be able to upgrade immediately and would still be good to try to nail down the cause of this.

Has anyone been able to replicate?

How many cases do we collectively know of?

Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 12 May 2014 16:27
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

AFAIK Ron they are just the characters that Word stick in when you type "word" and you get the special opening/closing quotes etc...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell [ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 16:13
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

The occasions I know of it's mostly been impossible to troubleshoot due to lack of detailed info or too late data already lost.

Alistair had regular problems and while we were trying to nail it down (over weeks) too many inconsistencies e.g. different projects, some apparently no copy and paste involved but only seemed to happen with dodgy connection.

The reason I mentioned nested pages is that seems to be common whenever this has happened but may just be coincidental.

I've also seen those ??? characters before but I've seen broken xml without any of that and I think I've seen working LO's which haven't broken and yet had those chars.

Is there a way of reliably generating those sorts of chars to past in to prove these are the cause?

From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 15:47
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss


1.      Possibly. He is remote to Nottingham, though has a Nottingham username.

2.      That doesn?t trip any switches with me ? what are you thinking?

Since I found non-UTF chars in the file, I have a strong suspicion that that is the cause: even though he is pasting into notepad from word (most of the time), some are getting through. I have suggested actually saving the file in notepad as UTF 8 to force it to happen.

But this is something I would really like to see fixed. Pasting from word is a fact of life unfortunately.

From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 12 May 2014 12:48
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Very frustrating replying and seeing loads of other replies sent after mine hitting the list before my reply even reaches! :-(

But anyway just re-reading the message from you user and I wonder?

1. Any likelihood of connection problems being the cause?

2. Wondering if rather than connectors or tab nav whether the problem only occurs with page types that have nested pages?


From: Ron Mitchell [mailto:ronm at mitchellmedia.co.uk]
Sent: 12 May 2014 12:33
To: 'For Xerte technical developers'
Subject: RE: [Xerte-dev] FW: Further Xerte Data Loss

I think the fact that it's a connector page in this case might be a red herring. I previously thought it was to do with the tabnav page types but that may be a red herring too.

Yes I've had reports of this quite a few times now (different installations) but haven't been able to nail it down or reliably replicate it except that in Alistairs case it was dodgy connection and so possibly packet loss.

The advice I've been giving is pasted below but I agree it's worrying whenever this happens even though I haven't ever been able to make it happen myself.

Sorry not much help I guess :-(

Guidance example given to one of the HEA projects:
I think you need to provide some guidance to the students to be very careful when authoring if any connection problems are noticed. Also be careful when copying and pasting text into xerte pages or writing code/html. Lost or corrupted pages with xerte are very rare but the occurrences we do know about either happen due to poor connectivity or to code/tags breaking the xml. The key message is do not click publish if any problems are noticed because preview saves the latest changes to preview.xml but these changes are not written to data.xml until publish is selected. So if a problem is noticed not clicking publish can sometimes mean that we can recover from data.xml even if preview.xml has been broken.


From: xerte-dev-bounces at lists.nottingham.ac.uk<mailto:xerte-dev-bounces at lists.nottingham.ac.uk> [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 12 May 2014 12:08
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk<mailto:xerte-dev at lists.nottingham.ac.uk>)
Subject: [Xerte-dev] FW: Further Xerte Data Loss


I guys, I need to get to the bottom of this issue: I have a fairly advanced user who has been periodically losing data: the email below explains the problem. A couple of weeks ago it happened and we lost 40 slides, now it has happened again. In the file I looked at, the broken file ended with the characters



<cMen



i.e. an incomplete tag, which I believe should be <cMenuConnector params=?etc?>



Does this ring any bells with anyone? Anyone experienced it before elsewhere? I don?t think the menu connectors are used much, which is why we haven?t come across it before?



I have a feeling it is something to do with escaping characters as data is being saved, or as it is being read back to the editor. Anyone have any other bright ideas how we can troubleshoot it? See below?



Thanks a lot,



Julian





I've experienced further data loss again this morning, this time on a completely different Xerte project (http://www.nottingham.ac.uk/toolkits/play_10189).



I initially produced this resource using the Flash Xerte engine, but the academic I'm working with was keen for me to transfer the data over into HTML5. I made a direct duplicate copy of the Flash project, ticked the HTML5 engine and then set about reproducing the slides - no problem.



Everything was absolutely fine (after working on the project for most of Friday) until around 10:30am this morning, when I lost around 10-15 slides (including my review quiz with 20 MCQs). I'm now beginning to strongly suspect that the Xerte Menu Connector templates might be having something to do with this persistent problem?



I have been using multiple connector menus in both my Grasslands and Meat Science packages. In both I have a dominant 'Main Menu' at the start, which then 'connects' via tabs to several sub-menus (of the same Connector template type) dotted throughout the resource. In each of the sub-menus I link back to the Main Menu.



My last actions before the slides were lost today, happened after I included a further sub-menu connector (at around slide 70) and then linked this to this from the 'Main Menu' (slide 2) via the drop-down selector menu. I then published the changes.



With that being said, I've attempted to recreate the problem using a dummy file, but I have so far failed to do it.


This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.


Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

From Julian.Tenney at nottingham.ac.uk  Tue May 20 16:08:09 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Tue, 20 May 2014 16:08:09 +0100
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023F@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <965A188BE17FF6418686E864C242C00A4D244D140F@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1AEDE69E@EXCHANGE1.ad.nottingham.ac.uk>
	<965A188BE17FF6418686E864C242C00A4D244D1420@EXCHANGE3.ad.nottingham.ac.uk>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F4102@EXCHANGE1.ad.nottingham.ac.uk>,
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023C@ITSEMBXCLUS.enterprise.gcal.ac.uk>,
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023E@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1023F@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F4374@EXCHANGE1.ad.nottingham.ac.uk>

Do you want to make that change?

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 20 May 2014 12:41
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

OK so the IE conditionals before the doctype might also cause issues but there is a workaround

http://stackoverflow.com/questions/2518256/override-intranet-compatibility-mode-ie8

The relevant section reads that we should change the code to start

<!--[if HTML5]><![endif]-->
<!doctype html>
<!--[if IEMobile 7 ]><html class="no-js iem7"><![endif]--> <!--[if lt IE 7]><html class="no-js lt-ie9 lt-ie8 lt-ie7 ie6 ie5"><![endif]--> <!--[if IE 7]><html class="no-js lt-ie9 lt-ie8 ie7"><![endif]--> <!--[if IE 8]><html class="no-js lt-ie9 ie8"><![endif]--> <!--[if IE 9]><html class="no-js ie9"><![endif]--> <!--[if (gt IE 9)|(gt IEMobile 7)|!(IEMobile)|!(IE)]><!--><html class="no-js" lang="en"><!--<![endif]--> <head> <!--[if !HTML5]>
        <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1"> <![endif]-->
        <meta charset="utf-8">

        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 12:01
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

It's possible that its the script or utf-8 setting BEFORE the IE=edge setting thats causing it... I've seen some suggest that the IE=edge needs to come first so try changing this:

<head>
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->

        <meta charset="utf-8">
        <meta http-equiv="X-UA-Compatible" content="IE=edge">


to this:

<head>
        <meta http-equiv="X-UA-Compatible" content="IE=edge">
        <meta charset="utf-8">
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 11:44
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

Thought we were only supporting "modern" browsers? ;-)

Anyway it's here https://github.com/thexerteproject/xerteonlinetoolkits/blob/e18bc9a58b1d63e09c351aa814f88ed2cf7ec827/modules/xerte/player_html5/rloObject.htm#L12

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 11:14
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

I can't find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query - I've noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an 'intranet site') and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I've noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don't think it's working - perhaps because it comes after the 'if IE' conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From J.J.Smith at gcu.ac.uk  Tue May 20 16:24:03 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Tue, 20 May 2014 16:24:03 +0100
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham
Message-ID: <uqdhxvkxls7cfhvpg1a67qvn.1400599440658@email.android.com>

I'll make it but you might want to try testing it at your end first if you know users that are having issues. I've not seen any issues and don't have IE9 to test but i don't remember ever seeing IE issues so it could be your default IE settings are different from mine.

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII


Julian Tenney <Julian.Tenney at nottingham.ac.uk> wrote:


Do you want to make that change?

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 20 May 2014 12:41
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

OK so the IE conditionals before the doctype might also cause issues but there is a workaround

http://stackoverflow.com/questions/2518256/override-intranet-compatibility-mode-ie8

The relevant section reads that we should change the code to start

<!--[if HTML5]><![endif]-->
<!doctype html>
<!--[if IEMobile 7 ]><html class="no-js iem7"><![endif]--> <!--[if lt IE 7]><html class="no-js lt-ie9 lt-ie8 lt-ie7 ie6 ie5"><![endif]--> <!--[if IE 7]><html class="no-js lt-ie9 lt-ie8 ie7"><![endif]--> <!--[if IE 8]><html class="no-js lt-ie9 ie8"><![endif]--> <!--[if IE 9]><html class="no-js ie9"><![endif]--> <!--[if (gt IE 9)|(gt IEMobile 7)|!(IEMobile)|!(IE)]><!--><html class="no-js" lang="en"><!--<![endif]--> <head> <!--[if !HTML5]>
        <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1"> <![endif]-->
        <meta charset="utf-8">

        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 12:01
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

It's possible that its the script or utf-8 setting BEFORE the IE=edge setting thats causing it... I've seen some suggest that the IE=edge needs to come first so try changing this:

<head>
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->

        <meta charset="utf-8">
        <meta http-equiv="X-UA-Compatible" content="IE=edge">


to this:

<head>
        <meta http-equiv="X-UA-Compatible" content="IE=edge">
        <meta charset="utf-8">
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 11:44
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

Thought we were only supporting "modern" browsers? ;-)

Anyway it's here https://github.com/thexerteproject/xerteonlinetoolkits/blob/e18bc9a58b1d63e09c351aa814f88ed2cf7ec827/modules/xerte/player_html5/rloObject.htm#L12

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 11:14
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

I can't find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query - I've noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an 'intranet site') and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I've noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don't think it's working - perhaps because it comes after the 'if IE' conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From Julian.Tenney at nottingham.ac.uk  Tue May 20 16:45:11 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Tue, 20 May 2014 16:45:11 +0100
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham
In-Reply-To: <uqdhxvkxls7cfhvpg1a67qvn.1400599440658@email.android.com>
References: <uqdhxvkxls7cfhvpg1a67qvn.1400599440658@email.android.com>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F43C5@EXCHANGE1.ad.nottingham.ac.uk>

If you commit it I'll test it,

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 20 May 2014 16:24
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

I'll make it but you might want to try testing it at your end first if you know users that are having issues. I've not seen any issues and don't have IE9 to test but i don't remember ever seeing IE issues so it could be your default IE settings are different from mine.

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII


Julian Tenney <Julian.Tenney at nottingham.ac.uk> wrote:


Do you want to make that change?

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 20 May 2014 12:41
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

OK so the IE conditionals before the doctype might also cause issues but there is a workaround

http://stackoverflow.com/questions/2518256/override-intranet-compatibility-mode-ie8

The relevant section reads that we should change the code to start

<!--[if HTML5]><![endif]-->
<!doctype html>
<!--[if IEMobile 7 ]><html class="no-js iem7"><![endif]--> <!--[if lt IE 7]><html class="no-js lt-ie9 lt-ie8 lt-ie7 ie6 ie5"><![endif]--> <!--[if IE 7]><html class="no-js lt-ie9 lt-ie8 ie7"><![endif]--> <!--[if IE 8]><html class="no-js lt-ie9 ie8"><![endif]--> <!--[if IE 9]><html class="no-js ie9"><![endif]--> <!--[if (gt IE 9)|(gt IEMobile 7)|!(IEMobile)|!(IE)]><!--><html class="no-js" lang="en"><!--<![endif]--> <head> <!--[if !HTML5]>
        <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1"> <![endif]-->
        <meta charset="utf-8">

        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 12:01
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

It's possible that its the script or utf-8 setting BEFORE the IE=edge setting thats causing it... I've seen some suggest that the IE=edge needs to come first so try changing this:

<head>
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->

        <meta charset="utf-8">
        <meta http-equiv="X-UA-Compatible" content="IE=edge">


to this:

<head>
        <meta http-equiv="X-UA-Compatible" content="IE=edge">
        <meta charset="utf-8">
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 11:44
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

Thought we were only supporting "modern" browsers? ;-)

Anyway it's here https://github.com/thexerteproject/xerteonlinetoolkits/blob/e18bc9a58b1d63e09c351aa814f88ed2cf7ec827/modules/xerte/player_html5/rloObject.htm#L12

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 11:14
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

I can't find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query - I've noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an 'intranet site') and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I've noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don't think it's working - perhaps because it comes after the 'if IE' conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


From ronm at mitchellmedia.co.uk  Tue May 20 17:29:11 2014
From: ronm at mitchellmedia.co.uk (Ron Mitchell)
Date: Tue, 20 May 2014 17:29:11 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F4361@EXCHANGE1.ad.nottingham.ac.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>	<00ea01cf6dd8$04760280$0d620780$@co.uk>	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>,
	<01d501cf6df4$c7cb7000$57625000$@co.uk>	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020B@ITSEMBXCLUS.enterprise.gcal.ac.uk>	<00d001cf7436$e82be300$b883a900$@co.uk>	<537B6AD8.2010903@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F4361@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <012d01cf7448$a2cf84d0$e86e8e70$@co.uk>

Oh dear maybe this isn't adding to the uncertainty or maybe it's a clue - just had the following reply from Alistair...

"I have still experienced some loss since my Broadband was stabilised. I was working on the Bootstrap template using Chrome browser and lost a set of pages about 2 weeks ago."

That's the first time I'm aware of this happening with Bootstrap too but I guess that's also writing xml so perhaps not really any different?

Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 20 May 2014 16:01
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

> One of the changes I've committed to github / a pull request is that there's better validation on the XML. i.e. if it doesn't validate then XOT won't overwrite what ever is in the USER-FILES/$whatever/(data|preview).xml file(s).

Cheers.

It does seem the data loss issue has the remote user as the common factor.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 20 May 2014 15:47
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

<snip/>


>
> A student on one of the HEA projects reported this recently and 
> responded as follows:
>
> ?No text was copied and pasted,  so I doubt there would be HTML, plus 
> no hyper linking was included....however pages were lost, but only 
> when working or accessing the project at my house anywhere else it 
> 'seemed' to be fine... But I can't be too sure?
>
> This mirrors the only consistent cause Alistair and I were able to 
> identify e.g. he was having connectivity problems and regularly 
> getting corrupt LO's or lost pages. Since his connectivity was fixed I 
> don't think this has happened but I'll check. All that said I've never 
> once had this happen to me even when on trains or using mobile 
> connections including the same codebase or even installations as 
> Alistair.
>
> Even if a new solution is put in place to validate before saving etc 
> it seems to me that there will still be high risk for all those users 
> and installations unlikely to be able to upgrade immediately and would 
> still be good to try to nail down the cause of this.
>

I've certainly seen corruption due to invalid characters being embedded into the LO. I can't replicate this, regardless of what sort of formatting I embed in, which is a pain.


One of the changes I've committed to github / a pull request is that there's better validation on the XML. i.e. if it doesn't validate then XOT won't overwrite what ever is in the USER-FILES/$whatever/(data|preview).xml file(s).

(via plugins/xml_parsing_check.php when/if merged).


David.

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev



From Julian.Tenney at nottingham.ac.uk  Wed May 21 09:24:03 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Wed, 21 May 2014 09:24:03 +0100
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss
In-Reply-To: <012d01cf7448$a2cf84d0$e86e8e70$@co.uk>
References: <4B6965E54E385E419329B3AA1E1A46F054C72FD5E1@EXCHANGE3.ad.nottingham.ac.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901A70@EXCHANGE1.ad.nottingham.ac.uk>
	<00ea01cf6dd8$04760280$0d620780$@co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1A901C80@EXCHANGE1.ad.nottingham.ac.uk>,
	<01d501cf6df4$c7cb7000$57625000$@co.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1020B@ITSEMBXCLUS.enterprise.gcal.ac.uk>
	<00d001cf7436$e82be300$b883a900$@co.uk>	<537B6AD8.2010903@palepurple.co.uk>
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F4361@EXCHANGE1.ad.nottingham.ac.uk>
	<012d01cf7448$a2cf84d0$e86e8e70$@co.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F44DA@EXCHANGE1.ad.nottingham.ac.uk>

In Bootstrap, all text is CDATA (I think), so it's not duff chars.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Ron Mitchell
Sent: 20 May 2014 17:29
To: 'For Xerte technical developers'
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

Oh dear maybe this isn't adding to the uncertainty or maybe it's a clue - just had the following reply from Alistair...

"I have still experienced some loss since my Broadband was stabilised. I was working on the Bootstrap template using Chrome browser and lost a set of pages about 2 weeks ago."

That's the first time I'm aware of this happening with Bootstrap too but I guess that's also writing xml so perhaps not really any different?

Ron

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney
Sent: 20 May 2014 16:01
To: For Xerte technical developers
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

> One of the changes I've committed to github / a pull request is that there's better validation on the XML. i.e. if it doesn't validate then XOT won't overwrite what ever is in the USER-FILES/$whatever/(data|preview).xml file(s).

Cheers.

It does seem the data loss issue has the remote user as the common factor.

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 20 May 2014 15:47
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: FW: Further Xerte Data Loss

<snip/>


>
> A student on one of the HEA projects reported this recently and 
> responded as follows:
>
> ?No text was copied and pasted,  so I doubt there would be HTML, plus 
> no hyper linking was included....however pages were lost, but only 
> when working or accessing the project at my house anywhere else it 
> 'seemed' to be fine... But I can't be too sure?
>
> This mirrors the only consistent cause Alistair and I were able to 
> identify e.g. he was having connectivity problems and regularly 
> getting corrupt LO's or lost pages. Since his connectivity was fixed I 
> don't think this has happened but I'll check. All that said I've never 
> once had this happen to me even when on trains or using mobile 
> connections including the same codebase or even installations as 
> Alistair.
>
> Even if a new solution is put in place to validate before saving etc 
> it seems to me that there will still be high risk for all those users 
> and installations unlikely to be able to upgrade immediately and would 
> still be good to try to nail down the cause of this.
>

I've certainly seen corruption due to invalid characters being embedded into the LO. I can't replicate this, regardless of what sort of formatting I embed in, which is a pain.


One of the changes I've committed to github / a pull request is that there's better validation on the XML. i.e. if it doesn't validate then XOT won't overwrite what ever is in the USER-FILES/$whatever/(data|preview).xml file(s).

(via plugins/xml_parsing_check.php when/if merged).


David.

--
Pale Purple Ltd

PHP Web application development and support

http://www.palepurple.co.uk
@PalePurpleLtd
07792 380669 / 0845 0046746

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev


_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

From J.J.Smith at gcu.ac.uk  Wed May 21 16:24:46 2014
From: J.J.Smith at gcu.ac.uk (Smith, John)
Date: Wed, 21 May 2014 16:24:46 +0100
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham
In-Reply-To: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F43C5@EXCHANGE1.ad.nottingham.ac.uk>
References: <uqdhxvkxls7cfhvpg1a67qvn.1400599440658@email.android.com>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F43C5@EXCHANGE1.ad.nottingham.ac.uk>
Message-ID: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1024F@ITSEMBXCLUS.enterprise.gcal.ac.uk>

Meant to say, I committed this earlier...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University
Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 16:45
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

If you commit it I'll test it,

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 20 May 2014 16:24
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

I'll make it but you might want to try testing it at your end first if you know users that are having issues. I've not seen any issues and don't have IE9 to test but i don't remember ever seeing IE issues so it could be your default IE settings are different from mine.

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII


Julian Tenney <Julian.Tenney at nottingham.ac.uk> wrote:


Do you want to make that change?

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 20 May 2014 12:41
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

OK so the IE conditionals before the doctype might also cause issues but there is a workaround

http://stackoverflow.com/questions/2518256/override-intranet-compatibility-mode-ie8

The relevant section reads that we should change the code to start

<!--[if HTML5]><![endif]-->
<!doctype html>
<!--[if IEMobile 7 ]><html class="no-js iem7"><![endif]--> <!--[if lt IE 7]><html class="no-js lt-ie9 lt-ie8 lt-ie7 ie6 ie5"><![endif]--> <!--[if IE 7]><html class="no-js lt-ie9 lt-ie8 ie7"><![endif]--> <!--[if IE 8]><html class="no-js lt-ie9 ie8"><![endif]--> <!--[if IE 9]><html class="no-js ie9"><![endif]--> <!--[if (gt IE 9)|(gt IEMobile 7)|!(IEMobile)|!(IE)]><!--><html class="no-js" lang="en"><!--<![endif]--> <head> <!--[if !HTML5]>
        <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1"> <![endif]-->
        <meta charset="utf-8">

        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 12:01
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

It's possible that its the script or utf-8 setting BEFORE the IE=edge setting thats causing it... I've seen some suggest that the IE=edge needs to come first so try changing this:

<head>
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->

        <meta charset="utf-8">
        <meta http-equiv="X-UA-Compatible" content="IE=edge">


to this:

<head>
        <meta http-equiv="X-UA-Compatible" content="IE=edge">
        <meta charset="utf-8">
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 11:44
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

Thought we were only supporting "modern" browsers? ;-)

Anyway it's here https://github.com/thexerteproject/xerteonlinetoolkits/blob/e18bc9a58b1d63e09c351aa814f88ed2cf7ec827/modules/xerte/player_html5/rloObject.htm#L12

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 11:14
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

I can't find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query - I've noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an 'intranet site') and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I've noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don't think it's working - perhaps because it comes after the 'if IE' conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment
may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474


From Julian.Tenney at nottingham.ac.uk  Wed May 21 16:29:35 2014
From: Julian.Tenney at nottingham.ac.uk (Julian Tenney)
Date: Wed, 21 May 2014 16:29:35 +0100
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham
In-Reply-To: <EE0B2AFFDB88B34AA864E00CE98914C22BFBC1024F@ITSEMBXCLUS.enterprise.gcal.ac.uk>
References: <uqdhxvkxls7cfhvpg1a67qvn.1400599440658@email.android.com>,
	<12C67A1EEC419342AF5E59DA31562C3F0D1B0F43C5@EXCHANGE1.ad.nottingham.ac.uk>
	<EE0B2AFFDB88B34AA864E00CE98914C22BFBC1024F@ITSEMBXCLUS.enterprise.gcal.ac.uk>
Message-ID: <12C67A1EEC419342AF5E59DA31562C3F0D1B0F48F9@EXCHANGE1.ad.nottingham.ac.uk>

Cheers, just been a  bit busy today. I'll get Joel to test it,

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 21 May 2014 16:25
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

Meant to say, I committed this earlier...

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 16:45
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

If you commit it I'll test it,

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 20 May 2014 16:24
To: xerte-dev at lists.nottingham.ac.uk
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

I'll make it but you might want to try testing it at your end first if you know users that are having issues. I've not seen any issues and don't have IE9 to test but i don't remember ever seeing IE issues so it could be your default IE settings are different from mine.

Regards

John Smith
Learning Technologist
School of Health and Life Sciences

Sent from Samsung Galaxy SII


Julian Tenney <Julian.Tenney at nottingham.ac.uk> wrote:


Do you want to make that change?

-----Original Message-----
From: xerte-dev-bounces at lists.nottingham.ac.uk [mailto:xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John
Sent: 20 May 2014 12:41
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

OK so the IE conditionals before the doctype might also cause issues but there is a workaround

http://stackoverflow.com/questions/2518256/override-intranet-compatibility-mode-ie8

The relevant section reads that we should change the code to start

<!--[if HTML5]><![endif]-->
<!doctype html>
<!--[if IEMobile 7 ]><html class="no-js iem7"><![endif]--> <!--[if lt IE 7]><html class="no-js lt-ie9 lt-ie8 lt-ie7 ie6 ie5"><![endif]--> <!--[if IE 7]><html class="no-js lt-ie9 lt-ie8 ie7"><![endif]--> <!--[if IE 8]><html class="no-js lt-ie9 ie8"><![endif]--> <!--[if IE 9]><html class="no-js ie9"><![endif]--> <!--[if (gt IE 9)|(gt IEMobile 7)|!(IEMobile)|!(IE)]><!--><html class="no-js" lang="en"><!--<![endif]--> <head> <!--[if !HTML5]>
        <meta http-equiv="X-UA-Compatible" content="IE=edge,chrome=1"> <![endif]-->
        <meta charset="utf-8">

        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 12:01
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

It's possible that its the script or utf-8 setting BEFORE the IE=edge setting thats causing it... I've seen some suggest that the IE=edge needs to come first so try changing this:

<head>
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->

        <meta charset="utf-8">
        <meta http-equiv="X-UA-Compatible" content="IE=edge">


to this:

<head>
        <meta http-equiv="X-UA-Compatible" content="IE=edge">
        <meta charset="utf-8">
        <script>(function(H){H.className=H.className.replace(/\bno-js\b/,'js')})(document.documentElement)</script><!--http://bit.ly/17XnmNe-->


Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Smith, John [J.J.Smith at gcu.ac.uk]
Sent: 20 May 2014 11:44
To: For Xerte technical developers
Subject: [Xerte-dev] Re: Xerte and IE doc mode at Nottingham

Thought we were only supporting "modern" browsers? ;-)

Anyway it's here https://github.com/thexerteproject/xerteonlinetoolkits/blob/e18bc9a58b1d63e09c351aa814f88ed2cf7ec827/modules/xerte/player_html5/rloObject.htm#L12

Regards,

John Smith | Learning Technologist
Room A251, Govan Mbeki Building | School of Health & Life Sciences | Glasgow Caledonian University Cowcaddens Road | Glasgow | G4 0BA

Please address ALL support requests to hlsblt at gcu.ac.uk where one of the School Learning Technologists will pick up the job. This will ensure that all jobs are completed as promptly as possible.
________________________________________
From: xerte-dev-bounces at lists.nottingham.ac.uk [xerte-dev-bounces at lists.nottingham.ac.uk] On Behalf Of Julian Tenney [Julian.Tenney at nottingham.ac.uk]
Sent: 20 May 2014 11:14
To: For Xerte technical developers (xerte-dev at lists.nottingham.ac.uk)
Subject: [Xerte-dev] FW: Xerte and IE doc mode at Nottingham

I can't find the code to have a look a this: where on earth is it? Does this seem sensible?


Quick query - I've noticed that Nottingham users viewing Xerte projects on-site with IE are having problems because IE will run in compatibility view (as Xerte is an 'intranet site') and document mode often reverts to old IE7 standards, therefore breaking a lot of the HTML5 functionality.

I've noticed that Xerte already has the <meta http-equiv="X-UA-Compatible" content="IE=edge"> tag which should correct this, but unfortunately I don't think it's working - perhaps because it comes after the 'if IE' conditions. Is there any way these could be reordered so that the IE=edge comes right at the top, like on Moodle? Or alternatively can this be configured server side via HTTP header?



This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.




Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev
This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it.   Please do not use, copy or disclose the information contained in this message or in any attachment.  Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham.

This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.





Glasgow Caledonian University is a registered Scottish charity, number SC021474

_______________________________________________
Xerte-dev mailing list
Xerte-dev at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte-dev