perl-unicode

[big5-*.ucm] please revise if possible

2002-04-20 03:23:17
On Saturday, April 20, 2002, at 05:21 , Autrijus Tang wrote:
On Sat, Apr 20, 2002 at 08:00:04AM +0900, Dan Kogai wrote:
Is this okay?  I think this is due to the edition difference.  If so,
please submit a fixed version of TW.t

Yes, as attached, thanks.

HenBao!

FYI here is what ucmlint is saying.
perl5.7.3 -Mblib bin/ucmlint -e ucm/big5-*.ucm
ucm/big5-eten.ucm: no error found
ucm/big5-hkscs.ucm: no error found
Is:
ucm/big5-eten.ucm: 16 errors found
ucm/big5-hkscs.ucm: 1319 errors found

That's right. The old ucm files was lacking these codepoints.

If possible please mark those that don't round-trip with 1| or 3|. euc-jp and Mac* do use this feature. you can grep thru for this. ucmlint can also help. try it without an argument first then with '-e' to make sure. Should be trivial for big5-eten but big5-hkscs may be hard ... or maybe not. just grep thru 'dupe encode map' and sed thru. It seems like CJK Compatibility parts (\x{F900}-\x{FA00}) are "new"....

Dan the Encode Maintainer