translation_pot.py

   1 #!/usr/bin/env python3
   2 #
   3 # Copyright (C) 2004-2021 ZNC, see the NOTICE file for details.
   4 #
   5 # Licensed under the Apache License, Version 2.0 (the "License");
   6 # you may not use this file except in compliance with the License.
   7 # You may obtain a copy of the License at
   8 #
   9 #     http://www.apache.org/licenses/LICENSE-2.0
  10 #
  11 # Unless required by applicable law or agreed to in writing, software
  12 # distributed under the License is distributed on an "AS IS" BASIS,
  13 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  14 # See the License for the specific language governing permissions and
  15 # limitations under the License.
  16 #
  17
  18 import argparse
  19 import glob
  20 import os
  21 import re
  22 import subprocess
  23
  24 parser = argparse.ArgumentParser()
  25 parser.add_argument('--include_dir', action='store')
  26 parser.add_argument('--explicit_sources', action='append')
  27 parser.add_argument('--tmpl_dirs', action='append')
  28 parser.add_argument('--strip_prefix', action='store')
  29 parser.add_argument('--tmp_prefix', action='store')
  30 parser.add_argument('--output', action='store')
  31 args = parser.parse_args()
  32
  33 pot_list = []
  34
  35 # .tmpl
  36 tmpl_pot = args.tmp_prefix + '_tmpl.pot'
  37 tmpl_uniq_pot = args.tmp_prefix + '_tmpl_uniq.pot'
  38 tmpl = []
  39 pattern = re.compile(r'<\?\s*(?:FORMAT|(PLURAL))\s+(?:CTX="([^"]+?)"\s+)?"([^"]+?)"(?(1)\s+"([^"]+?)"|).*?(?:"TRANSLATORS:\s*([^"]+?)")?\s*\?>')
  40 for tmpl_dir in args.tmpl_dirs:
  41     for fname in glob.iglob(tmpl_dir + '/*.tmpl'):
  42         fbase = fname[len(args.strip_prefix):]
  43         with open(fname, 'rt', encoding='utf8') as f:
  44             for linenum, line in enumerate(f):
  45                 for x in pattern.finditer(line):
  46                     text, plural, context, comment = x.group(3), x.group(4), x.group(2), x.group(5)
  47                     if comment:
  48                         tmpl.append('#  {}'.format(comment))
  49                     tmpl.append('#: {}:{}'.format(fbase, linenum + 1))
  50                     if context:
  51                         tmpl.append('msgctxt "{}"'.format(context))
  52                     tmpl.append('msgid "{}"'.format(text))
  53                     if plural:
  54                         tmpl.append('msgid_plural "{}"'.format(plural))
  55                         tmpl.append('msgstr[0] ""')
  56                         tmpl.append('msgstr[1] ""')
  57                     else:
  58                         tmpl.append('msgstr ""')
  59                     tmpl.append('')
  60
  61 # Bundle header to .tmpl, even if there were no .tmpl files.
  62 # Some .tmpl files contain non-ASCII characters, and the header is needed
  63 # anyway, because it's omitted from xgettext call below.
  64 with open(tmpl_pot, 'wt', encoding='utf8') as f:
  65     print('msgid ""', file=f)
  66     print('msgstr ""', file=f)
  67     print(r'"Content-Type: text/plain; charset=UTF-8\n"', file=f)
  68     print(r'"Content-Transfer-Encoding: 8bit\n"', file=f)
  69     print(file=f)
  70     for line in tmpl:
  71         print(line, file=f)
  72 subprocess.check_call(['msguniq', '--force-po', '-o', tmpl_uniq_pot, tmpl_pot])
  73 pot_list.append(tmpl_uniq_pot)
  74
  75 # .cpp
  76 main_pot = args.tmp_prefix + '_main.pot'
  77 subprocess.check_call(['xgettext',
  78     '--omit-header',
  79     '-D', args.include_dir,
  80     '-o', main_pot,
  81     '--keyword=t_s:1,1t',   '--keyword=t_s:1,2c,2t',
  82     '--keyword=t_f:1,1t',   '--keyword=t_f:1,2c,2t',
  83     '--keyword=t_p:1,2,3t', '--keyword=t_p:1,2,4c,4t',
  84     '--keyword=t_d:1,1t',   '--keyword=t_d:1,2c,2t',
  85 ] + args.explicit_sources)
  86 if os.path.isfile(main_pot):
  87     pot_list.append(main_pot)
  88
  89 # combine
  90 if pot_list:
  91     subprocess.check_call(['msgcat', '-o', args.output] + pot_list)