[KoCo-CVS] [Commit] KoreanCodecs/tools generate_codec_mapping.py
Brought to you by:
perky
Menu
▾
▴
[KoCo-CVS] [Commit] KoreanCodecs/tools generate_codec_mapping.py
|
From: Hye-Shik C. <pe...@us...> - 2003-01-12 23:12:49
|
perky 03/01/12 15:12:49
Added: tools generate_codec_mapping.py
Log:
Move src/tablegen.py to tools/generate_codec_mapping.py
Revision Changes Path
1.1 KoreanCodecs/tools/generate_codec_mapping.py
Index: generate_codec_mapping.py
===================================================================
#
# generate_codec_mapping.py - $Revision: 1.1 $
#
# Code Table Generator
#
# Author: Hye-Shik Chang <pe...@Fr...>
# Date : $Date: 2003/01/12 23:12:48 $
#
#
# This file is part of KoreanCodecs.
#
# KoreanCodecs is free software; you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as published
# by the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# KoreanCodecs is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public License
# along with KoreanCodecs; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
#
import time
UNICODE_INVALID = "UNIINV,"
COPYRIGHT_HEADER = """\
/*
* This file is part of KoreanCodecs.
*
* KoreanCodecs is free software; you can redistribute it and/or modify
* it under the terms of the GNU Lesser General Public License as published
* by the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* KoreanCodecs is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public License
* along with KoreanCodecs; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
*
* Generated by generate_codec_mapping.py on %s
* $Id: generate_codec_mapping.py,v 1.1 2003/01/12 23:12:48 perky Exp $
*/
""" % time.asctime(time.gmtime())
def tohex(s):
return "\\x%02x\\x%02x" % tuple(map(ord, s))
def decodemapgen(fo, prefix, bottom, top, m, print_region=1, print_index=1):
fmap = {}
for oco, dco in m.items():
fmap.setdefault(ord(oco[0]), {})
fmap[ord(oco[0])][ord(oco[1])] = dco
firstkeys = fmap.keys()
firstkeys.sort()
if print_region:
print >> fo, """\
#define %s_bottom %d
#define %s_top %d
""" % (prefix, bottom, prefix, top)
for fk in firstkeys:
seckeys = fmap[fk].keys()
seckeys.sort()
print >> fo, "static const Py_UNICODE %s_%02X[%d] = { /* %02X::%02X-%02X */" \
% (prefix, fk, top-bottom+1, fk, bottom, top)
seckeys = range(bottom, top+1)
while seckeys:
dp = seckeys[:8]
del seckeys[:8]
print >> fo, " ", ' '.join([
fmap[fk].has_key(i) and ("0x%04x," % ord(fmap[fk][i])) or UNICODE_INVALID
for i in dp
])
print >> fo, "};"
print >> fo
if print_index:
decmapindex(fo, prefix, fmap)
def decmapindex(fo, prefix, fmap):
print >> fo, "static const Py_UNICODE *%s_map[128] = {" % (prefix)
for i in range(128, 256):
if fmap.has_key(i):
print >> fo, " %s_%02X, /* 0x%02X */" % (prefix, i, i)
else:
print >> fo, " 0, /* 0x%02X */" % i
print >> fo, "};"
print >> fo
def encodemapgen(fo, prefix, m, threshold):
ecodes = m.keys()
ecodes.sort()
eblocks = [[-99999, -99999, {}]]
for i in ecodes:
if eblocks[-1][1] + threshold < ord(i):
eblocks.append([ord(i), ord(i), {ord(i):m[i]}])
else:
eblocks[-1][1] = ord(i)
eblocks[-1][2][ord(i)] = m[i]
blkcount = 0
for blk in eblocks[1:]:
print >> fo, "static const DBYTECHAR %s_page%d[%d] = { /* 0x%04x - 0x%04x */" % \
(prefix, blkcount, blk[1]-blk[0]+1, blk[0], blk[1])
blkcount += 1
obl = range(blk[0], blk[1]+1)
while obl:
dp = obl[:8]
del obl[:8]
print >> fo, " ", " ".join([
blk[2].has_key(ok) and
'0x%02x%02x,' % tuple(map(ord, blk[2][ok])) or "NOCHAR,"
for ok in dp
])
print >> fo, "};"
print >> fo
blkcount = 0
print >> fo, "#define _%s(uni) ( \\" % prefix
for blk in eblocks[1:]:
print >> fo, " uni >= 0x%04x && uni <= 0x%04x ? %s_page%d[uni-0x%04x] : \\" % (
blk[0], blk[1], prefix, blkcount, blk[0] )
blkcount += 1
print >> fo, " NOCHAR \\"
print >> fo, ")"
print >> fo
def hintgen(fo, prefix, m):
k = range(256)
print >> fo, "static const char %s_hint[256] = {" % prefix
while k:
n = k[:16]
del k[:16]
print >> fo, " ", " ".join(['%d,' % m[i] for i in n])
print >> fo, "};"
print >> fo
ksc5601 = open("_koco_ksc5601.h", "w")
print >> ksc5601, COPYRIGHT_HEADER
# johab_ideograph ksc5601_hangul ksc5601_ideograph ksc5601_misc uhc.py
from korean.mappings import ksc5601_hangul, ksc5601_ideograph, ksc5601_misc
ksc5601_decoding = {}
ksc5601_decoding.update(ksc5601_hangul.decoding_map)
ksc5601_decoding.update(ksc5601_ideograph.decoding_map)
ksc5601_decoding.update(ksc5601_misc.decoding_map)
decodemapgen(ksc5601, "ksc5601_decode", 0xa1, 0xfe, ksc5601_decoding)
print >> ksc5601
ksc5601_encoding = {}
ksc5601_encoding.update(ksc5601_hangul.encoding_map)
ksc5601_encoding.update(ksc5601_ideograph.encoding_map)
ksc5601_encoding.update(ksc5601_misc.encoding_map)
del ksc5601
del ksc5601_decoding, ksc5601_hangul, ksc5601_ideograph, ksc5601_misc
uhctable = open("_koco_uhc.h", "w")
print >> uhctable, COPYRIGHT_HEADER
print >> uhctable, """\
#define uhc_page0_bottom 0x41
#define uhc_page0_top 0xfe
#define uhc_page1_bottom 0x41
#define uhc_page1_top 0xa0
"""
from korean.mappings import uhc
uhcpage0 = {}
uhcpage1 = {}
fmap = {}
# Hmm, I need dictionary comprehension....
for code, uni in uhc.decoding_map.items():
if code[0] <= '\xa0': # page 0
uhcpage0[code] = uni
else:
uhcpage1[code] = uni
fmap[ord(code[0])] = None # just for index
decodemapgen(uhctable, "uhc_decode", 0x41, 0xfe, uhcpage0, 0, 0)
decodemapgen(uhctable, "uhc_decode", 0x41, 0xa0, uhcpage1, 0, 0)
decmapindex(uhctable, "uhc_decode", fmap)
hintarray = []
for i in range(256):
if chr(i).isalpha() or 0x81 <= i <= 0xA0:
hintarray.append(1)
else:
hintarray.append(0)
hintgen(uhctable, "uhc_decode", hintarray)
encmapfile = open("_koco_wansungenc.h", "w")
print >> encmapfile, COPYRIGHT_HEADER
ksc5601_encoding.update(uhc.encoding_map)
encodemapgen(encmapfile, "wansung_encode", ksc5601_encoding, 512)
#
# $Id: generate_codec_mapping.py,v 1.1 2003/01/12 23:12:48 perky Exp $
#
# -*- End-Of-File -*-
|
