Re: [commits] Horde branch master updated. a599f7e01b65871add15284defd4ba128c7b985f

Quoting Jan Schneider <[email protected]>:

>> commit a599f7e01b65871add15284defd4ba128c7b985f
>> Author: Michael J Rubinsky <[email protected]>
>> Date:   Thu May 15 16:40:56 2014 -0400
>>
>>    Add Horde_Core_History to work around chicken/egg problem.
>>
>>    Bug: 13105
>>
>> framework/Core/lib/Horde/Core/Factory/History.php |    3 +-
>> framework/Core/lib/Horde/Core/History.php         |  163  
>> +++++++++++++++++++++
>> framework/Core/package.xml                        |    6 +-
>> 3 files changed, 169 insertions(+), 3 deletions(-)
>> create mode 100644 framework/Core/lib/Horde/Core/History.php
>>
>> http://github.com/horde/horde/commit/a599f7e01b65871add15284defd4ba128c7b985f
>
> Re the @todo: Horde_Core_History *does* match Horde_History  
> typehints because it extends it. Or are you talking about different  
> typehints?

Exactly. I *had* to extend it for just that reason, but that means I  
couldn't rely on the __call() magic method to automatically pass  
non-decorated method calls to the composed Horde_History object.  
Though, now that I am explaining this it occurs to me that the reason  
I needed to compose the history object was because I originally  
thought I would need to create a brand new history object when the  
authenticated state changed. I later realized all I needed to do was  
ensure the 'who' parameter was passed to the log method. So, tl;dr, I  
probably don't need to actually wrap another history object anymore  
and can just override the log method. I'll revisit this next time I'm  
in front of my dev box.

-- 
mike
The Horde Project
http://www.horde.org
https://www.facebook.com/hordeproject
https://www.twitter.com/hordeproject

-- 
dev mailing list
Frequently Asked Questions: http://wiki.horde.org/FAQ
To unsubscribe, mail: [email protected]