[Xerte] toolkits installation feedback (+data.xml missing innew USER-FILES subdirs)

[Patrick Lockley]

It seems to work ok for the academic types (if anyone else would like to
comment on this let me know - the comment why not write a non-ldap
system will be ignored)

Those tables could be as big as they want to be, but I never need a
distinct ID in them, as the distinct ID comes from somewhere else. I
thought sql types called this a superkey? Maybe mysql no like. Maybe it
would make things faster - it won't hurt to put one in if someone
fancies experimenting.

I agree a more flexible sitedetails would be nice, but at the moment the
data that comes back and populates the xerte_toolkits_site object - It
would be much nicer if there was scope to do something more dynamic -
I'd probably go for a mysql_fetch_object if I did it now.

If you send me a link I can track it in fiddler - ideally the traffic
from when you click create to when the window pops up.

-----Original Message-----
From: xerte-bounces at lists.nottingham.ac.uk
[mailto:xerte-bounces at lists.nottingham.ac.uk] On Behalf Of David Goodwin
Sent: 15 September 2009 14:20
To: Xerte discussion list
Subject: Re: [Xerte] toolkits installation feedback (+data.xml missing
innew USER-FILES subdirs)

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Patrick Lockley wrote:
> I can change the suggestions - it took me an age using a new system
> (same code) to work out what it meant. I don't really get LDAP.
> 

I don't use it a huge amount ... so don't worry :)

> Templaterights / templatesyndication don't really need a primary key -
> it's linked to templatedetails on the template_id column. If this is
> naughty mysql then please accept my apologies.

Ah ok. Well, normally you'd want a key/index there to speed up searches.
It's possibly naughty... but if there are only ever a few rows in the
table, it's a non-issue.

> 
> I don't get point C, sorry, could you clarify it for me.
> 

<tangent>
Rather than a table which looks like :

sitedetails:
  bind_dn -> value
  admin -> value
  admin_pass -> value
  site_id -> value
  site_url -> value
  apache -> value
  ....

It may be better to have e.g. :

sitedetails :
  site_id
  name
  value

So your code would do e.g. 'select * from sitedetails where site_id = 1'
to get all name/value pairs back.

The benefit of this is that if you decide to have a new sitedetail field
you just add a new row into the database - rather than having to update
the schema of the table.

The only possible downside is that the fields would need to be of the
same storage type (e.g. char(255)); at the moment you have some that are
also text - so you could either have a seperate table for them (e.g.
sitedetails_text) or there is a software pattern which solves the
problem whose name I've conveniently forgotten.

</tangent>

> The new template code is in
website_code/php/templates/new_template.php
> 
> What does fiddler show you?
> 


Fiddler2.com implies it's only available for Windows; I don't use
Windows :-/ I can use wireshark or e.g. the burp suite - do you want all
traffic from login onwards?

Thanks for the quick reply,

David.
- --
 David Goodwin                          Pale Purple Limited
 Office: 0845 0046746                   Mobile: 07792380669
 http://www.palepurple.co.uk            Company No: 5580814
 'Business Web Application Development and Training in PHP'
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkqvlHUACgkQ/ISo3RF5V6b6YACfYmwLuUVpZh8wW9c+HyY6rjio
EBoAoMRU1XEUa6HSCiU917ZfTRkzZnpP
=Y922
-----END PGP SIGNATURE-----
_______________________________________________
Xerte mailing list
Xerte at lists.nottingham.ac.uk
http://lists.nottingham.ac.uk/mailman/listinfo/xerte