[dev] new optimized sql Share driver

[Jan Schneider]

Zitat von Duck <duck at obala.net>:

> On Monday 14 of January 2008 05:53:52 Chuck Hagenbuch wrote:
>> Quoting Duck <duck at obala.net>:
>> > I started to work on a new share driver today as I having problems
>> > with memory
>> > and execution time inside Horde_Share. Fist I must mention that my my
>> > driver, for now, do not support session cache, share names and
>> > hierarchies (no sub shares for ansel yet) as I was concentrating on the
>> > performance of the data retrieval.
>>
>> This sounds potentially promising; I hope to look at it in a few days.
>> Unfortunately I got pretty backed up this weekend and have a lot of
>> patches that I'm behind on.
>>
>> -chuck
>
> I go foward fixing names and attributes so is fully functional in Genie. I
> will test it with other shares groups.
>
> I added it as ticket #6109.
>
> I must add someting here. Looking at share code we list shares using share
> name as list index. I purpose to change this to use directly the share ID.
> This will speed up geting the share in the manipulation process even in
> DataTree driver and save a little bit of list memory.

That's not possible since other share implementation like Kolab don't  
have IDs.

Jan.

-- 
Do you need professional PHP or Horde consulting?
http://horde.org/consulting/