Yes- I have reverted the earlier fix, and added the DIY version, along with a number of other fixes from the bug tracker.
allan On Tue, Sep 17, 2013 at 7:12 PM, Olaf Meeuwissen <olaf.meeuwissen at avasys.jp> wrote: > > Sergey Vlasov writes: > >> On Mon, Aug 26, 2013 at 12:41:19PM +0900, Olaf Meeuwissen wrote: >>> m. allan noah writes: >>> >>> > That is one option. The other is to do own own upcasing, [...] >>> >>> Sure, but why bother if you can use standard library API? >> >> setlocale() has a process-wide effect - if a multithreaded process >> uses the SANE API on one thread and has other threads running at the >> same time, using setlocale() even temporarily can interfere with >> operations performed by other threads. > > Hadn't thought of that, sorry. > > Indeed, there is no mention of the change being thread-local so this > could affect a multi-threaded frontend. In that case, it might be a > better idea to roll our own upcasing. > > Allan, please revert dd4f26a805ed7be8e30b3cd655de891141a9530d and apply > the patch you suggested in [1]. > > [1] > http://lists.alioth.debian.org/pipermail/sane-devel/2013-August/031641.html > > Hope this helps, > -- > Olaf Meeuwissen, LPIC-2 FLOSS Engineer -- AVASYS CORPORATION > FSF Associate Member #1962 Help support software freedom > http://www.fsf.org/jf?referrer=1962 -- "The truth is an offense, but not a sin"