localedata/unicode-gen/gen_translit_cjk_compat.py

   1 #!/usr/bin/python3
   2 # -*- coding: utf-8 -*-
   3 #
   4 # Generate a translit_cjk_compat file from a UnicodeData file.
   5 # Copyright (C) 2015-2016 Free Software Foundation, Inc.
   6 # This file is part of the GNU C Library.
   7 #
   8 # The GNU C Library is free software; you can redistribute it and/or
   9 # modify it under the terms of the GNU Lesser General Public
  10 # License as published by the Free Software Foundation; either
  11 # version 2.1 of the License, or (at your option) any later version.
  12 #
  13 # The GNU C Library is distributed in the hope that it will be useful,
  14 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  15 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  16 # Lesser General Public License for more details.
  17 #
  18 # You should have received a copy of the GNU Lesser General Public
  19 # License along with the GNU C Library; if not, see
  20 # <http://www.gnu.org/licenses/>.
  21
  22 '''
  23 Generate a translit_cjk_compat file from UnicodeData.txt
  24
  25 To see how this script is used, call it with the “-h” option:
  26
  27     $ ./gen_translit_cjk_compat -h
  28     … prints usage message …
  29 '''
  30
  31 import argparse
  32 import time
  33 import sys
  34 import unicode_utils
  35
  36 def read_input_file(filename):
  37     '''Reads the original glibc translit_cjk_compat file to get the
  38     original head and tail.
  39
  40     We want to replace only the part of the file between
  41     “translit_start” and “translit_end”
  42     '''
  43     head = tail = ''
  44     with open(filename, mode='r') as translit_file:
  45         for line in translit_file:
  46             head = head + line
  47             if line.startswith('translit_start'):
  48                 break
  49         for line in translit_file:
  50             if line.startswith('translit_end'):
  51                 tail = line
  52                 break
  53         for line in translit_file:
  54             tail = tail + line
  55     return (head, tail)
  56
  57 def output_head(translit_file, unicode_version, head=''):
  58     '''Write the header of the output file, i.e. the part of the file
  59     before the “translit_start” line.
  60     '''
  61     if ARGS.input_file and head:
  62         translit_file.write(head)
  63     else:
  64         translit_file.write('escape_char /\n')
  65         translit_file.write('comment_char %\n')
  66         translit_file.write('\n')
  67         translit_file.write('% Transliterations of CJK compatibility ')
  68         translit_file.write('characters.\n')
  69         translit_file.write('% Generated automatically from UnicodeData.txt '
  70                             + 'by gen_translit_cjk_compat.py '
  71                             + 'on {:s} '.format(time.strftime('%Y-%m-%d'))
  72                             + 'for Unicode {:s}.\n'.format(unicode_version))
  73         translit_file.write('\n')
  74         translit_file.write('LC_CTYPE\n')
  75         translit_file.write('\n')
  76         translit_file.write('translit_start\n')
  77
  78 def output_tail(translit_file, tail=''):
  79     '''Write the tail of the output file'''
  80     if ARGS.input_file and tail:
  81         translit_file.write(tail)
  82     else:
  83         translit_file.write('translit_end\n')
  84         translit_file.write('\n')
  85         translit_file.write('END LC_CTYPE\n')
  86
  87 def special_decompose(code_point_list):
  88     '''
  89     Decompositions which are not in UnicodeData.txt at all but which
  90     were used in the original translit_cjk_compat file in glibc and
  91     which seem to make sense.  I want to keep the update of
  92     translit_cjk_compat close to the spirit of the original file,
  93     therefore I added this special decomposition rules here.
  94     '''
  95     special_decompose_dict = {
  96         (0x2215,): [0x002F], # ∕ → /
  97         (0x00B2,): [0x005E, 0x0032], # ² → ^2
  98         (0x03BC,): [0x00B5], # μ → µ (GREEK SMALL LETTER MU → MICRO SIGN)
  99         (0x2113,): [0x006C], # ℓ → l
 100         (0x00B3,): [0x005E, 0x0033], # ³ → ^3
 101         (0x00B5,): [0x0075], # µ → u
 102         (0x03BC, 0x2113): [0x03BC, 0x006C], # μℓ → μl
 103         (0x0072, 0x0061, 0x0064, 0x2215, 0x0073, 0x00B2): [
 104             0x0072, 0x0061, 0x0064, 0x002F, 0x0073, 0x00B2],
 105         (0x006D, 0x2215, 0x0073, 0x00B2): [0x006D, 0x002F, 0x0073, 0x00B2],
 106     }
 107     if tuple(code_point_list) in special_decompose_dict:
 108         return special_decompose_dict[tuple(code_point_list)]
 109     else:
 110         return code_point_list
 111
 112 def output_transliteration(translit_file):
 113     '''Write the new transliteration to the output file'''
 114     translit_file.write('\n')
 115     for code_point in sorted(unicode_utils.UNICODE_ATTRIBUTES):
 116         name = unicode_utils.UNICODE_ATTRIBUTES[code_point]['name']
 117         decomposition = unicode_utils.UNICODE_ATTRIBUTES[
 118             code_point]['decomposition']
 119         if decomposition.startswith('<square>'):
 120             decomposition = decomposition[9:]
 121             decomposed_code_points = [[int(x, 16)
 122                                        for x in decomposition.split(' ')]]
 123             if decomposed_code_points[0]:
 124                 while True:
 125                     special_decomposed_code_points = special_decompose(
 126                         decomposed_code_points[-1])
 127                     if (special_decomposed_code_points
 128                             != decomposed_code_points[-1]):
 129                         decomposed_code_points.append(
 130                             special_decomposed_code_points)
 131                         continue
 132                     special_decomposed_code_points = []
 133                     for decomposed_code_point in decomposed_code_points[-1]:
 134                         special_decomposed_code_points += special_decompose(
 135                             [decomposed_code_point])
 136                     if (special_decomposed_code_points
 137                             == decomposed_code_points[-1]):
 138                         break
 139                     decomposed_code_points.append(
 140                         special_decomposed_code_points)
 141                 translit_file.write('% {:s}\n'.format(name))
 142                 translit_file.write('{:s} '.format(
 143                     unicode_utils.ucs_symbol(code_point)))
 144                 for index in range(0, len(decomposed_code_points)):
 145                     if index > 0:
 146                         translit_file.write(';')
 147                     if len(decomposed_code_points[index]) > 1:
 148                         translit_file.write('"')
 149                     for decomposed_code_point in decomposed_code_points[index]:
 150                         translit_file.write('{:s}'.format(
 151                             unicode_utils.ucs_symbol(decomposed_code_point)))
 152                     if len(decomposed_code_points[index]) > 1:
 153                         translit_file.write('"')
 154                 translit_file.write('\n')
 155     for code_point in sorted(unicode_utils.UNICODE_ATTRIBUTES):
 156         name = unicode_utils.UNICODE_ATTRIBUTES[code_point]['name']
 157         decomposition = unicode_utils.UNICODE_ATTRIBUTES[
 158             code_point]['decomposition']
 159         if decomposition and name.startswith('CJK COMPATIBILITY IDEOGRAPH'):
 160             decomposed_code_points = [int(x, 16)
 161                                       for x in decomposition.split(' ')]
 162             if len(decomposed_code_points) != 1:
 163                 sys.stderr.write(
 164                     'Unexpected decomposition length {:x} {:s} {:s}\n'.format(
 165                         code_point, name, decomposition))
 166                 exit(1)
 167             translit_file.write('% {:s}\n'.format(name))
 168             translit_file.write('{:s} '.format(
 169                 unicode_utils.ucs_symbol(code_point)))
 170             for decomposed_code_point in decomposed_code_points:
 171                 translit_file.write('{:s}'.format(
 172                     unicode_utils.ucs_symbol(decomposed_code_point)))
 173             translit_file.write('\n')
 174     translit_file.write('\n')
 175
 176 if __name__ == "__main__":
 177     PARSER = argparse.ArgumentParser(
 178         description='''
 179         Generate a translit_cjk_compat file from UnicodeData.txt.
 180         ''')
 181     PARSER.add_argument(
 182         '-u', '--unicode_data_file',
 183         nargs='?',
 184         type=str,
 185         default='UnicodeData.txt',
 186         help=('The UnicodeData.txt file to read, '
 187               + 'default: %(default)s'))
 188     PARSER.add_argument(
 189         '-i', '--input_file',
 190         nargs='?',
 191         type=str,
 192         help=''' The original glibc/localedata/locales/translit_cjk_compat
 193         file.''')
 194     PARSER.add_argument(
 195         '-o', '--output_file',
 196         nargs='?',
 197         type=str,
 198         default='translit_cjk_compat.new',
 199         help='''The new translit_cjk_compat file, default: %(default)s.  If the
 200         original glibc/localedata/locales/translit_cjk_compat file has
 201         been given as an option, the header up to the
 202         “translit_start” line and the tail from the “translit_end”
 203         line to the end of the file will be copied unchanged into the
 204         output file.  ''')
 205     PARSER.add_argument(
 206         '--unicode_version',
 207         nargs='?',
 208         required=True,
 209         type=str,
 210         help='The Unicode version of the input files used.')
 211     ARGS = PARSER.parse_args()
 212
 213     unicode_utils.fill_attributes(ARGS.unicode_data_file)
 214     HEAD = TAIL = ''
 215     if ARGS.input_file:
 216         (HEAD, TAIL) = read_input_file(ARGS.input_file)
 217     with open(ARGS.output_file, mode='w') as TRANSLIT_FILE:
 218         output_head(TRANSLIT_FILE, ARGS.unicode_version, head=HEAD)
 219         output_transliteration(TRANSLIT_FILE)
 220         output_tail(TRANSLIT_FILE, tail=TAIL)