[dev] Splitting the share driver

[Michael Rubinsky]

Quoting Duck <duck at obala.net>:

> On Monday 07 August 2006 16:51, Gunnar Wrobel wrote:
>> So why is there a need for this random string?
>
> Honestly I don't know. Maybe is needed by the module architecture.

For one, the id is numeric, the share_name is text.  It would have  
been impossible - or at least very difficult - to use ids as the share  
identifiers for the turba share implementation where we rely on the  
username being part of the share identifier.

> I took a little look at Ansel code. The navigation uses IDs. But they are
> changed to names when retrieve shares. So moving to IDs should not be hard to
> implement and will bring performance and reduction of some not needed code.

I believe that the share_name is needed for some of the Horde_Share  
methods, like Horde_Share::exists() so we would have to (and do)  
convert back to the share name for these instances anyway.  Although  
there are probably places in the Ansel code where we can get away with  
moving more towards the ids.

Thanks,
mike

--
The Horde Project (www.horde.org)
mrubinsk at horde.org