~mdw / sgt / charset / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 9b7e7a9) | patch
sbcsgen.pl was giving different results on different machines in the case
authorjacob <jacob@cda61777-01e9-0310-a592-d414129be87e>
Wed, 26 Apr 2006 23:01:06 +0000 (23:01 +0000)
committerjacob <jacob@cda61777-01e9-0310-a592-d414129be87e>
Wed, 26 Apr 2006 23:01:06 +0000 (23:01 +0000)
commit4fd00a6fa00809e1179cf8f7bcdd33cc7bbfbb29
treef423c35530fdf0ee555c05a92a189cb23df62349tree | snapshot (tar.gz zip)
parent9b7e7a9270e9749c6c86bd70ea98377d0d0a9003commit | diff
sbcsgen.pl was giving different results on different machines in the case
where two SBCS code points mapped to a single Unicode point.
Changed so that by default it favours the lower SBCS code point.
On ixion, this highlighted ambiguities in CS_MAC_THAI, CS_MAC_SYMBOL, and
CS_VISCII. Guessed at a preference for the first two and added "sortpriority"
directives. (No idea about VISCII.)

git-svn-id: svn://svn.tartarus.org/sgt/charset@6641 cda61777-01e9-0310-a592-d414129be87e
sbcs.dat diff | blob | blame | history
sbcsgen.pl diff | blob | blame | history
Simon Tatham's `charset' library; import of svn://svn.tartarus.org/sgt/charset