[SSSD] [PATCHES] Alternative objectClass for group maps

Jakub Hrozek jhrozek at redhat.com
Mon Sep 15 08:22:03 UTC 2014


On Thu, Sep 11, 2014 at 04:20:25PM +0200, Jakub Hrozek wrote:
> On Thu, Sep 11, 2014 at 12:24:45PM +0200, Michal Židek wrote:
> > The function is there to replace long "if" conditions. There
> > were 2 in the code. One used strncasecmp and one strcasecmp.
> >                                 ^
> > Where no length is needed to be specified, you can pass 0
> > to this function and the len parameter will be calculated
> > automatically based on the length of objectclass.
> 
> OK, this makes sense.
> 
> I tested IPA backend with a non-trivial hierarchy:
> 
> user - nonposixgr1
>      - posixgr1 - nonposixgr2 - posixgr2
> 
> AD with the same hierarchy and pure LDAP with RFC2307. With AD I tested
> also the matchin_rule_in_chain option doesn't break.
> 
> So ACK from me. I'd like to wait for our tests to finish before pushing,
> though.

The tests finally finished and found no regressions, so I went ahead and
pushed these patches:

* master:
     * 6f91c61426c8cfbfec52d5e77ae4650007694e69
     * 7ba70236daccb48432350147d0560b3302518cee 
* sssd-1-11:
     * cfa74fcb5f6ba23f41a9ddaa76c3ebae6156da86
     * 9e99c000a4e2647328e71b4db272b4b73a7189c5 



More information about the sssd-devel mailing list