Hi, On 01/30/2014 02:54 PM, Hans de Goede wrote: > Hi, > > On 01/30/2014 02:26 PM, Julien Cristau wrote: >> On Thu, Jan 30, 2014 at 09:43:37 +0100, Hans de Goede wrote: >> >>> Hi, >>> >>> On 01/30/2014 12:51 AM, Peter Hutterer wrote: >>>> Introduced in fecc7eb1cf66db64728ee2d68cd9443df7e70879 and reverts most of >>>> that but it's helpfully mixed with other stuff. >>>> >>>> InputAttributes are not const, they're strdup'd everywhere but the test >>>> code >>>> and freed properly. Revert the const char changes and fix the test up >>>> instead. >>>> >>>> Signed-off-by: Peter Hutterer <[email protected]> >>> >>> Please use XNFstrdup instead of strdup (or add explicit error checking) >>> with that changed: >>> >> AFAICT the only strdups Peter adds are in the test code. I don't think >> it matters there. > > Look closer, he also adds strdup calls to the new XkbInitRules function in > xkb/xkbInit.c .
Oops correction I should look closer myself, that is in 2/7 not 3/7 . Anyways I still believe that using XNFstrdup instead would be better. I don't see why sloppy coding (lacking error checking is sloppy coding) would be ok in test cases. Regards, Hans _______________________________________________ [email protected]: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
