|
Message-ID: <20140703071318.GA10117@brightrain.aerifal.cx>
Date: Thu, 3 Jul 2014 03:13:18 -0400
From: Rich Felker <dalias@...c.org>
To: musl@...ts.openwall.com
Subject: Possible bytelocale patch
As mentioned in the LC_CTYPE conformance thread, I may hold off on
committing these changes, but I figure it's best to go ahead and post
the patch. As-is, the patch introduced a conformance regression; I had
hoped mapping high bytes to a high PUA range would be acceptable, but
since musl returns true for iswprint() for PUA codepoints, this makes
isprint() and iswprint() inconsistent and thus non-conforming. As
there are fewer than 128 permanent non-characters in Unicode, I think
at least iswprint would need to have locale-specific behavior to fix
this problem.
Other than that, the patch works, it seems mostly complete (let me
know if anything is missing), and it presumably makes happy the people
who want to grep binary files (and all of the conservative corporate
titans who get the final word in the Austin Group)...
Rich
View attachment "locale_part2.diff" of type "text/plain" (11777 bytes)
Powered by blists - more mailing lists
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.