On Mon, 13 Aug 2001 10:35:08 +0700
Martin_Hosken(_at_)sil(_dot_)org wrote:
PS2. Any chance that the Text::Unicode::Normalize module (if it is going
into the core) can have its dependency on Lingua::KO::Hangul::Util removed?
We certainly can disperse Lingua::KO::Hangul::Util,
e.g. decomposeHangul and composeHangul to Unicode::Normalize,
getHangulName to Unicode::UCD,
parseHangulName to charnames.pm,
and we have some copies of common constants for Hangul.
Considering the particularity of Hangul Syllables,
- they comprise *11,172* characters.
cf. Unicode 3.0 Summary
(see http://www.unicode.org/unicode/standard/versions/)
Category V 2.1 V 3.0
Alphabetics, Symbols 6511 10236
CJK Ideographs 21204 27786
Hangul Syllables 11172 11172
- their properties can (and virtually *must*) be derived
using well-defined algorithms.
- maybe unfamiliar (not only to western people)
in comparison with LATIN, GREEK, HAN, etc. scripts.
would it be better to gather those functions into one module?
-----
Regards, SADAHIRO Tomoyuki
