On Tue, Jul 22, 2014 at 03:32:18PM +0200, Pavel Březina wrote:
On 07/22/2014 10:43 AM, Jakub Hrozek wrote:
>On Tue, Jul 22, 2014 at 09:33:01AM +0200, Jakub Hrozek wrote:
>>On Mon, Jul 21, 2014 at 02:10:34PM +0200, Pavel Březina wrote:
>>>On 07/21/2014 12:07 PM, Jakub Hrozek wrote:
>>>>On Fri, Jul 18, 2014 at 01:15:25PM +0200, Pavel Březina wrote:
>>>>>https://fedorahosted.org/sssd/ticket/2385
>>>>
>>>>The code looks good, but needs to be rebased on top of PavelR's
patches
>>>>I just pushed (sorry, first comes, first served..)
>>>
>>>The problem is that the first patch is basically the same as Pavel wrote. It
>>>just contains a comment explaining why ENOENT error is acceptable and a
>>>little bit different debug message. Feel free to pick whatever version you
>>>think is better, rebasing is very simple.
>>
>>Ah, I didn't realize this is the same place as well.
>>
>>>
>>>The second patch applies correctly.
>>
>>OK, I will review it.
>
>0001: I kinda like the comment, feel free to re-send with just the
>comment
Done.
>0002: Can you drop state->num_missing_sids? Appears to be unused, the
>array is NULL-terminated anyway
It is used. There is a little bit of confusion because some functions
returns NULL terminated arrays others not so we have to keep both. Looking
into the code I think it should be possible to use only NULL terminated
arrays.
Ah, you're right, I was confused because at one spot num_missing_sids is
used as input and at others as output, sorry.
ACK to both, then.
Shall I create a refactoring ticket?
You can, maybe it would be better to do the change in master right
away while you still know the code?