glibc/localedata/unicode-gen/gen_translit_combining.py

446 lines
19 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

#!/usr/bin/python3
# -*- coding: utf-8 -*-
#
# Generate a translit_combining file from a UnicodeData file.
# Copyright (C) 2015-2017 Free Software Foundation, Inc.
# This file is part of the GNU C Library.
#
# The GNU C Library 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 2.1 of the License, or (at your option) any later version.
#
# The GNU C Library 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 the GNU C Library; if not, see
# <http://www.gnu.org/licenses/>.
'''
Generate a translit_combining file from UnicodeData.txt
To see how this script is used, call it with the “-h” option:
$ ./gen_translit_combining -h
… prints usage message …
'''
import argparse
import time
import unicode_utils
def read_input_file(filename):
'''Reads the original glibc translit_combining file to get the
original head and tail.
We want to replace only the part of the file between
“translit_start” and “translit_end”
'''
head = tail = ''
with open(filename, mode='r') as translit_file:
for line in translit_file:
head = head + line
if line.startswith('translit_start'):
break
for line in translit_file:
if line.startswith('translit_end'):
tail = line
break
for line in translit_file:
tail = tail + line
return (head, tail)
def output_head(translit_file, unicode_version, head=''):
'''Write the header of the output file, i.e. the part of the file
before the “translit_start” line.
'''
if ARGS.input_file and head:
translit_file.write(head)
else:
translit_file.write('escape_char /\n')
translit_file.write('comment_char %\n')
translit_file.write(unicode_utils.COMMENT_HEADER)
translit_file.write('\n')
translit_file.write('% Transliterations that remove all ')
translit_file.write('combining characters (accents,\n')
translit_file.write('% pronounciation marks, etc.).\n')
translit_file.write('% Generated automatically from UnicodeData.txt '
+ 'by gen_translit_combining.py '
+ 'on {:s} '.format(time.strftime('%Y-%m-%d'))
+ 'for Unicode {:s}.\n'.format(unicode_version))
translit_file.write('\n')
translit_file.write('LC_CTYPE\n')
translit_file.write('\n')
translit_file.write('translit_start\n')
def output_tail(translit_file, tail=''):
'''Write the tail of the output file'''
if ARGS.input_file and tail:
translit_file.write(tail)
else:
translit_file.write('translit_end\n')
translit_file.write('\n')
translit_file.write('END LC_CTYPE\n')
def is_combining_remove(code_point):
'''Check whether this is a combining character which should be listed
in the section of the translit_combining file where combining
characters are replaced by empty strings.
We ignore combining characters from many scripts here because
the original translit_combining file didnt do this for the
combining characters from these scripts either and I am not
sure yet whether this would be useful to do for all combining
characters or not. For the moment I think it is better to keep
close to the spirit of the original file.
'''
if not unicode_utils.is_combining(code_point):
return False
name = unicode_utils.UNICODE_ATTRIBUTES[code_point]['name']
for substring in ('DEVANAGARI',
'BENGALI',
'CYRILLIC',
'SYRIAC',
'THAANA',
'NKO',
'GURMUKHI',
'TAMIL',
'GUJARATI',
'ORIYA',
'TELUGU',
'KANNADA',
'MALAYALAM',
'SINHALA',
'THAI',
'LAO',
'TIBETAN',
'MYANMAR',
'ETHIOPIC',
'TAGALOG',
'HANUNOO',
'BUHID',
'TAGBANWA',
'KHMER',
'MONGOLIAN',
'LIMBU',
'NEW TAI LUE',
'BUGINESE',
'BALINESE',
'SUNDANESE',
'LEPCHA',
'IDEOGRAPHIC',
'HANGUL',
'SYLOTI',
'SAURASHTRA',
'KAYAH',
'REJANG',
'CHAM',
'VARIATION SELECTOR',
'KHAROSHTHI',
'MUSICAL SYMBOL',
'SAMARITAN',
'MANDAIC',
'TAI THAM',
'BATAK',
'VEDIC',
'COPTIC',
'TIFINAGH',
'BAMUM',
'JAVANESE',
'TAI VIET',
'MEETEI',
'MANICHAEAN',
'BRAHMI',
'KAITHI',
'CHAKMA',
'MAHAJANI',
'SHARADA',
'KHOJKI',
'KHUDAWADI',
'GRANTHA',
'TIRHUTA',
'SIDDHAM',
'MODI VOWEL',
'MODI SIGN',
'TAKRI',
'BASSA VAH',
'PAHAWH HMONG',
'MIAO',
'DUPLOYAN',
'MENDE KIKAKUI',
'AHOM',
'SIGNWRITING'
):
if substring in name:
return False
return True
def canonical_decompose(code_point):
'''http://www.unicode.org/reports/tr44/#Character_Decomposition_Mappings
In some instances a canonical mapping or a compatibility mapping
may consist of a single character. For a canonical mapping, this
indicates that the character is a canonical equivalent of another
single character. For a compatibility mapping, this indicates that
the character is a compatibility equivalent of another single
character.
A canonical mapping may also consist of a pair of characters, but
is never longer than two characters. When a canonical mapping
consists of a pair of characters, the first character may itself
be a character with a decomposition mapping, but the second
character never has a decomposition mapping.
We ignore the canonical decomposition for code points
matching certain substrings because the original translit_combining
file didnt include these types of characters either. I am unsure
about the usefulness of including them and want to keep close
to the spirit of the original file for the moment.
'''
name = unicode_utils.UNICODE_ATTRIBUTES[code_point]['name']
for substring in ('MUSICAL SYMBOL',
'CJK COMPATIBILITY IDEOGRAPH',
'BALINESE',
'KAITHI LETTER',
'CHAKMA VOWEL',
'GRANTHA VOWEL',
'TIRHUTA VOWEL',
'SIDDHAM VOWEL'):
if substring in name:
return []
decomposition = unicode_utils.UNICODE_ATTRIBUTES[
code_point]['decomposition']
if decomposition and not decomposition.startswith('<'):
decomposed_code_points = [int(x, 16) for x in decomposition.split(' ')]
if decomposed_code_points:
cd0 = canonical_decompose(decomposed_code_points[0])
if cd0:
decomposed_code_points = cd0 + decomposed_code_points[1:]
return decomposed_code_points
else:
return []
def special_decompose(code_point_list):
'''
Decompositions which are not canonical or which are not in
UnicodeData.txt at all but some of these were used in the original
translit_combining file in glibc and they seemed to make sense.
I want to keep the update of translit_combining close to the
spirit of the original file, therefore I added these special
decomposition rules here.
'''
special_decompose_dict = {
# Ø U+00D8 is already handled in translit_neutral. But
# translit_combining is usually included after translit_neutral
# and Ǿ U+01FE LATIN CAPITAL LETTER O WITH STROKE AND ACUTE
# has a canonical decomposition to Ø U+00D8 and we want to
# further decompose this to U+004F.
(0x00D8,): [0x004F], # Ø → O
# ø U+00F8 is already handled in translit_neutral. But
# translit_combining is usually included after translit_neutral
# and ǿ U+01FF LATIN SMALL LETTER O WITH STROKE AND ACUTE
# has a canonical decomposition to ø U+00F8 and we want to
# further decompose this to U+006F.
(0x00F8,): [0x006F], # ø → o
# æ U+00E6 is already in translit_compat because ligatures
# are handled in translit_compat. But ǣ U+01E3 has a
# canonical decomposition to U+00E6, U+0304 and we want to
# further decompose this to “ae”.
(0x00E6,): [0x0061, 0x0065], # æ → ae
# Æ U+00C6 is already in translit_compat because ligatures
# are handled in translit_compat. But Ǣ U+01E2 has a
# canonical decomposition to U+00C6, U+0304 and we want to
# further decompose this to “AE”
(0x00C6,): [0x0041, 0x0045], # Æ → AE
# U+05F2 HEBREW LIGATURE YIDDISH DOUBLE YOD is already in
# translit_compat because ligatures are handled in translit_compat.
# But U+FB1F has a canonical decomposition to U+05F2 and
# we want to further decompose this to U+05D9, U+05D9.
(0x05F2,): [0x05D9, 0x05D9], # ײ → יי
# 0x2002 has a <compat> decomposition to 0x0020 in UnicodeData.txt
# But U+2000 EN QUAD has a canonical decomposition U+2002
# and we want to further decompose this to U+0020.
(0x2002,): [0x0020], # EN SPACE → SPACE
# 0x2003 has a <compat> decomposition to 0x0020 in UnicodeData.txt
# But U+2001 EM QUAD has a canonical decomposition to U+2003
# and we want to further decompose this to U+0020.
(0x2003,): [0x0020], # EM SPACE → SPACE
# U+2260 ≠ has the canonical decomposition U+003D U+0338
# (= followed by ̸). After stripping the combining characters,
# the result is only = which reverses the meaning.
# Therefore, we add a special rules here for such mathematical
# negations:
(0x21AE,): [0x0021, 0x003C, 0x002D, 0x003E], # ↮ → !<->
(0x21CD,): [0x0021, 0x003C, 0x003D], # ⇍ → !<=
(0x21CE,): [0x0021, 0x003C, 0x003D, 0x003E], # ⇎ → !<=>
(0x21CF,): [0x0021, 0x003D, 0x003E], # ⇏ → !=>
(0x2204,): [0x0021, 0x2203], # ∄ → !∃
(0x2209,): [0x0021, 0x2208], # ∉ → !∈
(0x220C,): [0x0021, 0x220B], # ∌ → !∋
(0x2224,): [0x0021, 0x2223], # ∤ → !
(0x2226,): [0x0021, 0x2225], # ∦ → !∥
(0x2241,): [0x0021, 0x007E], # ≁ → !~
(0x2244,): [0x0021, 0x007E, 0x002D], # ≄ → !~-
(0x2247,): [0x0021, 0x007E, 0x003D], # ≇ → !~=
(0x2249,): [0x0021, 0x007E, 0x007E], # ≉ → !~~
(0x2260,): [0x0021, 0x003D], # ≠ → !=
(0x2262,): [0x0021, 0x003D, 0x003D], # ≢ → !==
(0x226D,): [0x0021, 0x224D], # ≭ → !≍
(0x226E,): [0x0021, 0x003C], # ≮ → !<
(0x226F,): [0x0021, 0x003E], # ≯ → !>
(0x2270,): [0x0021, 0x003C, 0x003D], # ≰ → !<=
(0x2271,): [0x0021, 0x003E, 0x003D], # ≱ → !>=
(0x2274,): [0x0021, 0x003C, 0x007E], # ≴ → !<~
(0x2275,): [0x0021, 0x003E, 0x007E], # ≵ → !>~
(0x2278,): [0x0021, 0x003C, 0x003E], # ≸ → !<>
(0x2279,): [0x0021, 0x003E, 0x003C], # ≹ → !><
(0x2280,): [0x0021, 0x227A], # ⊀ → !≺
(0x2281,): [0x0021, 0x227B], # ⊁ → !≻
(0x2284,): [0x0021, 0x2282], # ⊄ → !⊂
(0x2285,): [0x0021, 0x2283], # ⊅ → !⊃
(0x2288,): [0x0021, 0x2282, 0x003D], # ⊈ → !⊂=
(0x2289,): [0x0021, 0x2283, 0x003D], # ⊉ → !⊃=
(0x22AC,): [0x0021, 0x22A2], # ⊬ → !⊢
(0x22AD,): [0x0021, 0x22A8], # ⊭ → !⊨
(0x22AE,): [0x0021, 0x22A9], # ⊮ → !⊩
(0x22AF,): [0x0021, 0x22AB], # ⊯ → !⊫
(0x22E0,): [0x0021, 0x227C], # ⋠ → !≼
(0x22E1,): [0x0021, 0x227D], # ⋡ → !≽
(0x22E2,): [0x0021, 0x2291], # ⋢ → !⊑
(0x22E3,): [0x0021, 0x2292], # ⋣ → !⊒
(0x22EA,): [0x0021, 0x22B2], # ⋪ → !⊲
(0x22EB,): [0x0021, 0x22B3], # ⋫ → !⊳
(0x22EC,): [0x0021, 0x22B4], # ⋬ → !⊴
(0x22ED,): [0x0021, 0x22B5], # ⋭ → !⊵
(0x2ADC,): [0x0021, 0x2ADD], # ⫝̸ → !⫝
# Special rule for 〈 U+3008 is added
# because 〉 U+2329 has the canonical decomposition U+3008
# and we want to further decompose this to > U+003C.
(0x3008,): [0x003C], # 〈 → <
# Special rule for 〉 U+3009 is added
# because 〉 U+232A has the canonical decomposition U+3009
# and we want to further decompose this to < U+003E.
(0x3009,): [0x003E], # 〉→ >
}
if tuple(code_point_list) in special_decompose_dict:
return special_decompose_dict[tuple(code_point_list)]
else:
return code_point_list
def output_combining_remove(translit_file):
'''Write the section of the translit_combining file where combining
characters are replaced by empty strings.
'''
translit_file.write('\n')
for code_point in sorted(unicode_utils.UNICODE_ATTRIBUTES):
name = unicode_utils.UNICODE_ATTRIBUTES[code_point]['name']
if is_combining_remove(code_point):
translit_file.write('% {:s}\n'.format(name))
translit_file.write('{:s} ""\n'.format(
unicode_utils.ucs_symbol(code_point)))
translit_file.write('\n')
def output_decompositions(translit_file):
'''Write the section of the translit_combining file where characters
characters are decomposed and combining characters stripped from
the decompositions.
'''
for code_point in sorted(unicode_utils.UNICODE_ATTRIBUTES):
if special_decompose([code_point]) != [code_point]:
decomposed_code_points = [special_decompose([code_point])]
else:
decomposed_code_points = [canonical_decompose(code_point)]
if decomposed_code_points[0]:
while True:
special_decomposed_code_points = special_decompose(
decomposed_code_points[-1])
if (special_decomposed_code_points
!= decomposed_code_points[-1]):
decomposed_code_points.append(
special_decomposed_code_points)
continue
special_decomposed_code_points = []
for decomposed_code_point in decomposed_code_points[-1]:
special_decomposed_code_points += special_decompose(
[decomposed_code_point])
if (special_decomposed_code_points
== decomposed_code_points[-1]):
break
decomposed_code_points.append(
special_decomposed_code_points)
for index in range(0, len(decomposed_code_points)):
decomposed_code_points[index] = [
x for x in decomposed_code_points[index]
if not is_combining_remove(x)]
if decomposed_code_points[0]:
translit_file.write('% {:s}\n'.format(
unicode_utils.UNICODE_ATTRIBUTES[code_point]['name']))
translit_file.write('{:s} '.format(
unicode_utils.ucs_symbol(code_point)))
for index in range(0, len(decomposed_code_points)):
if index > 0:
translit_file.write(';')
if len(decomposed_code_points[index]) > 1:
translit_file.write('"')
for decomposed_code_point in decomposed_code_points[index]:
translit_file.write('{:s}'.format(
unicode_utils.ucs_symbol(decomposed_code_point)))
if len(decomposed_code_points[index]) > 1:
translit_file.write('"')
translit_file.write('\n')
translit_file.write('\n')
def output_transliteration(translit_file):
'''Write the new transliteration to the output file'''
output_combining_remove(translit_file)
output_decompositions(translit_file)
if __name__ == "__main__":
PARSER = argparse.ArgumentParser(
description='''
Generate a translit_combining file from UnicodeData.txt.
''')
PARSER.add_argument(
'-u', '--unicode_data_file',
nargs='?',
type=str,
default='UnicodeData.txt',
help=('The UnicodeData.txt file to read, '
+ 'default: %(default)s'))
PARSER.add_argument(
'-i', '--input_file',
nargs='?',
type=str,
help=''' The original glibc/localedata/locales/translit_combining
file.''')
PARSER.add_argument(
'-o', '--output_file',
nargs='?',
type=str,
default='translit_combining.new',
help='''The new translit_combining file, default: %(default)s. If the
original glibc/localedata/locales/translit_combining file has
been given as an option, the header up to the
“translit_start” line and the tail from the “translit_end”
line to the end of the file will be copied unchanged into the
output file. ''')
PARSER.add_argument(
'--unicode_version',
nargs='?',
required=True,
type=str,
help='The Unicode version of the input files used.')
ARGS = PARSER.parse_args()
unicode_utils.fill_attributes(ARGS.unicode_data_file)
HEAD = TAIL = ''
if ARGS.input_file:
(HEAD, TAIL) = read_input_file(ARGS.input_file)
with open(ARGS.output_file, mode='w') as TRANSLIT_FILE:
output_head(TRANSLIT_FILE, ARGS.unicode_version, head=HEAD)
output_transliteration(TRANSLIT_FILE)
output_tail(TRANSLIT_FILE, tail=TAIL)