On Tue, Mar 19, 2013 at 09:44:37PM -0500, Vladimir Támara Patiño wrote: > Attached --includes patch for spanish speaking countries.
I don't think it makes sense to manually update the en_US.UTF-8 file. Rather, I think we should have a script which generates en_US.UTF-8 from unicode.org XML data files (see http://www.unicode.org/Public/6.2.0/ucdxml/). As a first step, such a script would generate an en_US.UTF-8 file containing the same character definition blocks as the current file, possibly containing corrections within these blocks. In a later step we could think about adding more blocks if doing so would be useful. A suitable scripting language would be Perl since Perl exists in the base system and supports XML parsing. Your diff is adding ctype definitions for character sets which our libc doesn't support, such as euc-jp. I don't see any reason to add such files as long as libc doesn't support the corresponding encodings. I don't even see much reason to add support for additional encodings unless you can provide a convincing reason to do so. UTF-8 should cover most, if not all, needs.
