On Tue, 2011-12-13 at 14:44 +0100, Jakub Hrozek wrote: Patch 0001: Nack I don't like that we're assuming the length of the string here. I'd much prefer that sss_utf8_tolower() and sss_tc_utf8_tolower() take a length argument and pass those verbatim to the libunistring and glib2 lowercasing functions. We shouldn't assume that the string is NULL-terminated (and it's possible, though unlikely, that we might at some point want to work on a substring anyway).
Also, I don't think it's safe to use talloc_strdup() in sss_tc_utf8_tolower(). It would be safer here to use talloc_memdup(). Patch 0002: Ack. Patch 0003: Nack Correct me if I'm wrong, but aren't you leaking the old pd->user if you replace it with the primary name? Patch 0004: Ack Patch 0005: Ack Patch 0006: Ack Patch 0007: Ack patch 0008: This should probably be using sss_utf8_case_eq() instead of strcasecmp(). Patch 0009: Ack.
signature.asc
Description: This is a digitally signed message part
_______________________________________________ sssd-devel mailing list sssd-devel@lists.fedorahosted.org https://fedorahosted.org/mailman/listinfo/sssd-devel
