@RichFelker genuinely curious why recent musl has 8-bit locale support? Easy to remove, but a bit surprised it's not C.UTF-8 only anymore.
@bofh453 If you want a trivial way to compile it out, in src/internal/locale_impl.h define MB_CUR_MAX as 4.
-
-
@richfelker heh. that's actually exactly what I've been doing.Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@richfelker by the way, musl's fnmatch is by far the nicest/smallest I've seen, I use it in an NTapi toolkit. -
@bofh453 Actually I want to improve it - the 3 stages are not needed; it can be done with just 2 and that improves big-O in some cases.
End of conversation
New conversation -
-
-
@bofh453 However note that this also fixed various issues with build scripts using grep/sed/etc. on invalid (or non-UTF-8) text files. -
@richfelker ahh. makes sense. I've run into that before, and my sol'n was to loosen utf8 validation in regex, which is a bad idea in a libc. -
@bofh453 Note that the byte-based C locale only takes effect if you explicitly set LC_* vars to "C" or prog doesn't call setlocale(). -
@bofh453 In the latter case, functions that depend on character encoding can't be expected to work portably.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.