Script to create cp949.dat of go-charset from the CP949.TXT.

gen-cp949_dat.py

#!/usr/bin/env python

# CP949 specs;
# http://ko.wikipedia.org/wiki/%EC%BD%94%EB%93%9C_%ED%8E%98%EC%9D%B4%EC%A7%80_949
# http://msdn.microsoft.com/ko-kr/goglobal/cc305154.aspx

import re

ptn = re.compile(r'0x([A-F0-9]*)\s*0x([A-F0-9]*)')

chunk_begin = 0
chunk_buf = ""
chunk_cnt = 0

code_cnt = 0
data_buf = ""

def makeChunkData(begin, chunk):
    cl = len(chunk)
    return chr((begin & 0xff00) >> 8) + chr(begin & 0xff) + \
           chr((cl & 0xff00) >> 8) + chr(cl & 0xff) + chunk

last_native = 0


# XXX: download CP949.TXT from;
# http://www.unicode.org/Public/MAPPINGS/VENDORS/MICSFT/WINDOWS/CP949.TXT
for line in open("CP949.TXT"):
    if line.startswith("#"):
        continue
    rst = ptn.findall(line)
    if len(rst) != 1:
        continue

    curr_native, curr_unicode = int(rst[0][0], 16), int(rst[0][1], 16)
    if curr_native == curr_unicode:
        continue

    code_cnt += 1

    if chunk_begin == 0:
        chunk_begin = curr_native

    if (curr_native - 1 != last_native) and (last_native != 0):
        data_buf += makeChunkData(chunk_begin, chunk_buf)
        chunk_cnt += 1
        chunk_begin = curr_native
        chunk_buf = ""

    chunk_buf += unichr(curr_unicode).encode('utf-8')
    last_native = curr_native

data_buf += makeChunkData(chunk_begin, chunk_buf)
chunk_cnt += 1

wp = open('cp949.dat', 'w')
wp.write(chr((code_cnt & 0xff00) >> 8))
wp.write(chr(code_cnt & 0xff))
wp.write(chr((chunk_cnt & 0xff00) >> 8))
wp.write(chr(chunk_cnt & 0xff))
wp.write(data_buf)
wp.close()

print "chunk_cnt =", chunk_cnt, "code_cnt =", code_cnt