On Fri, 2015 Sep 25 19:49-0700, Paul Eggert wrote: > > Thanks, given all that history let's rewrite it so that the compiler > can decide what '\n' maps to, that way it'll work even in EBCDIC > environments that agree with IANA instead of IBM. I installed the > attached patch, which should fix the bugs you mentioned.
I'm happy to report that test-c-ctype in Git ff1ef114 now passes with both signed and unsigned EBCDIC chars on z/OS. Thank you for chasing this down! I will investigate further some of the issues that have been uncovered in this thread, and return to this list with my findings in a few days. --Daniel -- Daniel Richard G. || sk...@iskunk.org My ASCII-art .sig got a bad case of Times New Roman.
