On Thu, 28 Mar 2002, Dan Kogai wrote:
Dan,
I am ALMOST ready to release version 1.00 of Encode. But with a last
minute, welcome addition of Jungshik Shin to the team, I am tempted to
wait until he is done with Johab.
So Jungshik, how long will you take to submit Johab.pm ? If you think
it takes longer (FYI we are ALREADY in the code-freeze phase), tell me
so and I will release 1.00 without Johab.
I looked at HZ.pm and realized that in case of JOHAB, I can't
use the approach used for HZ.pm (for UTF-8 -> JOHAB, it's easy to do it
the way HZ encoder is written, but JOHAB -> UTF-8 cannot be done that
way). If I really want algorithimic converter for JOHAB, I have to do
what's done for UTF-8, but I guess it'll take me long because
I'm not familiar with mixing C and Perl.
Therefore, I just made johab.enc (which format is needed, ucm
or enc? It seems like compile can deal with both of them. Perl
devel. snapshot I downloaded don't have any ucm file for CJK encodings
except for EUC-JP). I'm going to attach it along with some other fixes
for iso-2022-kr(it uses SO/SI to shift between - invoke on GL - KS X
1001 and US-ASCII/KS X 1003 and 'ESC$)C' is only used to declare that G0
is US-ASCII/KS X 1003 and G1 is KS X 1001), euc-kr.enc/ksc5601.enc (two
characters are added to KS X 1001 in December 1998. cp949.enc already has
them at corresponding positions, but euc-kr.enc and ksc5601.enc didn't).
I can assure you that the mapping table for johab is correct,
but I haven't tested it yet. I'm trying to squeeze out disk space to
install the devel. version of Perl. Could you apply the patch
and see if it compiles all right? It's up to you whether to
include this or not in 1.0.
FYI the alias is already fixed.
Thank you !
Jungshik
kr.patch
Description: a patch to add johab and fix a few KR-related issues
