On 04/15/2013 12:44 PM, Jakub Hrozek wrote:
On Fri, Apr 12, 2013 at 08:35:23PM +0200, Michal Židek wrote:
> On 04/12/2013 04:57 PM, Jakub Hrozek wrote:
>> On Thu, Apr 11, 2013 at 11:35:52AM +0200, Michal Židek wrote:
>>> New patch attached.
>>>
>>> Thanks
>>> Michal
>>>
>>> + if (IS_SUBDOMAIN(dinfo)) {
>>> + name_lower = sss_tc_utf8_str_tolower(tmp_ctx, parsed_name);
>>> + domain_lower = sss_tc_utf8_str_tolower(tmp_ctx, parsed_domain);
>>> + if (!name_lower || !domain_lower) {
>>> + DEBUG(SSSDBG_CRIT_FAILURE, ("Out of memory\n"));
>>> + ret = ENOMEM;
>>> + goto done;
>>> + }
>>
>> One last nitpick -- while this code is technically correct now, it might not
>> be in future, we should also check the case_sensitive flag in dinfo and
>> only lowercase based on the value of the flag.
>
> Ok. I have also slightly changed the code flow in update_filter()
> function to make it more readable and added new parameter to force
> case sensibility (for autofs maps). New patch attached.
>
> Thanks
> Michal
>
It seems that fqdns do not work with this patch. Also
Yes. Sorry, the last change I did broke fqdns for users from native domains.
get_next_domain(dom, true) can maybe be used in the domain loop.
Great. I could remove the ugly part with "main_domain". Thanks for
catching this.
I am also sending the patch with the annoying message to this thread (it
was already acked in another thread, but depends on this patch with
subdomains).
Thanks
Michal