On 7/22/05, graeme <[EMAIL PROTECTED]> wrote:
> I've been pottering around in the code and came across the following
> changes that I feel need to be done, most are pedantic observations:
> 
> *version.php*

Fixed, thanks for noticing that. :)

> *Controller.class.php*
> 
> This is a singleton class BUT the implementation is poorly done:

Yes, there's no reason for them to be singletons either. It's on my
list of things to do, refactor Controller into a proper abstract
_factory_ class. I created a new ticket for it, #37.

> *WebRequest.class.php*
> loadParameters() method
> 
> The code to parse the PATH_INFO should be conditional on AG_URL_FORMAT
> (set up in settings.ini). You may want to extract all of the PATH_INFO
> code to a private methods to make the code a little cleaner.

Yes, agreed... created ticket #38


> *SessionStorage.class.php*
> initialize() method
> This (I think) links into anther problem which I've yet to fully grasp
> but I think that the session name needs to be more than just Agavi. I'd
> suggest Agavi + ApplicationName (not certain where/if an app name is set)
> The problem that I saw was when I ran two different M3 applications on
> the same machine - not tested it with Agavi - but the second app would
> fail if an object was saved with the user object of the first app, this
> is because they are sharing the same session info and in the second app
> the class is not auto loaded.

Certainly couldnt be any harm in making it less ambiguous, eh? :)

ticket #39 and #40.


-Mike
_______________________________________________
agavi-dev mailing list
[email protected]
http://labworkz.com/cgi-bin/mailman/listinfo/agavi-dev

Reply via email to