[dev] [commits] Horde branch Horde_LoginTasks-Refactor created. b70b84343d05538bad914a0738ae25e63ab31806

Gunnar Wrobel p at rdus.de
Wed Mar 3 13:55:49 UTC 2010


Hi Michael,

thanks for the feedback.

Quoting Michael M Slusarz <slusarz at horde.org>:

> Quoting Gunnar Wrobel <p at rdus.de>:
>
>>    There are two tests that are marked as incomplete: I currently  
>> still fail to understand the LoginTasks_Tasklist logic that uses  
>> the private _addTask variable. Depending on the task order some  
>> tasks will get displayed or not. But I don't see the rationale  
>> behind it. But the area should see no modifications during  
>> refactoring.
>
> I assume you are talking about the _addFlag variable (rather than  
> the addTag function).

Correct :)

> The _addFlag variable is used to determine if a tasks page needs to  
> be generated before a confirmation pages.

I did not have enough time to try to understand the task list handling  
in depth now. But as I did not intend to change anything within that  
class I think I can safely postpone this to later.

> For example, if you have the following tasks:
>
> DISPLAY_CONFIRM_NO
> DISPLAY_CONFIRM_YES
> DISPLAY_NONE
> DISPLAY_NOTICE
> DISPLAY_CONFIRM_YES
> DISPLAY_NONE
>
> There should be 3 confirmation screens generated: 1 with CONFIRM_NO  
> and CONFIRM_YES tasks,

As CONFIRM_NO != CONFIRM_YES the check "$v['task']->display !=  
$display" in needDisplay() should disallow displaying both on the same  
screen or do I misunderstand that section?

> 1 with NOTICE task, and 1 with CONFIRM_YES task.

I'm done with the core part of the refactoring. Having the abstract  
Horde_LoginTasks_Backend class should be sufficient for me to plug the  
system into something other than Horde. Can you give me some feedback  
if you think the modifications are okay and if I can merge?

Two additional items...

One question concerning ONCE tasks: Why is the _once value set when  
the task is being added to the tasklist? Shouldn't it be set once the  
task was actually executed? I didn't test this but I assume the user  
could login, visit the confirmation page, kill his session, login  
again and thus avoid execution of the task.

The Horde_Util::getFormData() in runTasks() is still inconvenient as  
it accesses $_POST directly so I guess I will add a workaround there  
later.


Cheers,

Gunnar

>
> michael
>
> -- 
> ___________________________________
> Michael Slusarz [slusarz at horde.org]
>
>
> -- 
> Horde developers mailing list - Join the hunt: http://horde.org/bounties/
> Frequently Asked Questions: http://horde.org/faq/
> To unsubscribe, mail: dev-unsubscribe at lists.horde.org
>







More information about the dev mailing list