s4:"samdb_search_count" - introduce a "mem_ctx" parameter

[Matthias Dieter Wallnöfer]

Ah, sorry,

didn't know about this rule.

Greets,
Matthias

Andrew Bartlett wrote:
> On Mon, 2010-10-25 at 19:43 +0200, Matthias Dieter Wallnöfer wrote:
>    
>> The branch, master has been updated
>>         via  6fb64b9 s4:"samdb_search_count" - introduce a "mem_ctx" parameter
>>         via  fd7943b ldb:gendb_* calls: support a "NULL" resultset parameter
>>        from  8bc2b54 s4-test: Extend DRS-msDSIntId test to verify Configuration NC replica also
>>
>> http://gitweb.samba.org/?p=samba.git;a=shortlog;h=master
>>
>>
>> - Log -----------------------------------------------------------------
>> commit 6fb64b9c7a281c2d148238390fccc08dce962f92
>> Author: Matthias Dieter Wallnöfer<mdw at samba.org>
>> Date:   Mon Oct 25 18:14:02 2010 +0200
>>
>>      s4:"samdb_search_count" - introduce a "mem_ctx" parameter
>>
>>      All other "samdb_search_*" calls do have one - why "samdb_search_count" doesn't?
>>
>>      Autobuild-User: Matthias Dieter Wallnöfer<mdw at samba.org>
>>      Autobuild-Date: Mon Oct 25 17:42:33 UTC 2010 on sn-devel-104
>>      
> Matthias,
>
> While I do appreciate your clean-up work, I would encourage you to ask
> such questions on IRC or the mailing list, rather than just guessing and
> committing.
>
> In this case, the reason there is no mem_ctx parameter is that no memory
> is returned, and so there is no reason that the caller should supply a
> memory context.
>
> While there are other functions in Samba that take a memory context and
> do not return memory, that should not be taken as precedent - in fact,
> these should be tidied up so that they do not return memory.  As the
> behaviour changes can be subtle, I ask that you allow me to review such
> patches for you.
>
> Thanks,
>
> Andrew Bartlett
>
>