1 #!/usr/bin/env python3
2 """
3 Convert the X11 locale.alias file into a mapping dictionary suitable
4 for locale.py.
5
6 Written by Marc-Andre Lemburg <mal@genix.com>, 2004-12-10.
7
8 """
9 import locale
10 import sys
11 _locale = locale
12
13 # Location of the X11 alias file.
14 LOCALE_ALIAS = '/usr/share/X11/locale/locale.alias'
15 # Location of the glibc SUPPORTED locales file.
16 SUPPORTED = '/usr/share/i18n/SUPPORTED'
17
18 def parse(filename):
19
20 with open(filename, encoding='latin1') as f:
21 lines = list(f)
22 # Remove mojibake in /usr/share/X11/locale/locale.alias.
23 # b'\xef\xbf\xbd' == '\ufffd'.encode('utf-8')
24 lines = [line for line in lines if '\xef\xbf\xbd' not in line]
25 data = {}
26 for line in lines:
27 line = line.strip()
28 if not line:
29 continue
30 if line[:1] == '#':
31 continue
32 locale, alias = line.split()
33 # Fix non-standard locale names, e.g. ks_IN@devanagari.UTF-8
34 if '@' in alias:
35 alias_lang, _, alias_mod = alias.partition('@')
36 if '.' in alias_mod:
37 alias_mod, _, alias_enc = alias_mod.partition('.')
38 alias = alias_lang + '.' + alias_enc + '@' + alias_mod
39 # Strip ':'
40 if locale[-1] == ':':
41 locale = locale[:-1]
42 # Lower-case locale
43 locale = locale.lower()
44 # Ignore one letter locale mappings (except for 'c')
45 if len(locale) == 1 and locale != 'c':
46 continue
47 # Normalize encoding, if given
48 if '.' in locale:
49 lang, encoding = locale.split('.')[:2]
50 encoding = encoding.replace('-', '')
51 encoding = encoding.replace('_', '')
52 locale = lang + '.' + encoding
53 data[locale] = alias
54 return data
55
56 def parse_glibc_supported(filename):
57
58 with open(filename, encoding='latin1') as f:
59 lines = list(f)
60 data = {}
61 for line in lines:
62 line = line.strip()
63 if not line:
64 continue
65 if line[:1] == '#':
66 continue
67 line = line.replace('/', ' ').strip()
68 line = line.rstrip('\\').rstrip()
69 words = line.split()
70 if len(words) != 2:
71 continue
72 alias, alias_encoding = words
73 # Lower-case locale
74 locale = alias.lower()
75 # Normalize encoding, if given
76 if '.' in locale:
77 lang, encoding = locale.split('.')[:2]
78 encoding = encoding.replace('-', '')
79 encoding = encoding.replace('_', '')
80 locale = lang + '.' + encoding
81 # Add an encoding to alias
82 alias, _, modifier = alias.partition('@')
83 alias = _locale._replace_encoding(alias, alias_encoding)
84 if modifier and not (modifier == 'euro' and alias_encoding == 'ISO-8859-15'):
85 alias += '@' + modifier
86 data[locale] = alias
87 return data
88
89 def pprint(data):
90 items = sorted(data.items())
91 for k, v in items:
92 print(' %-40s%a,' % ('%a:' % k, v))
93
94 def print_differences(data, olddata):
95 items = sorted(olddata.items())
96 for k, v in items:
97 if k not in data:
98 print('# removed %a' % k)
99 elif olddata[k] != data[k]:
100 print('# updated %a -> %a to %a' % \
101 (k, olddata[k], data[k]))
102 # Additions are not mentioned
103
104 def optimize(data):
105 locale_alias = locale.locale_alias
106 locale.locale_alias = data.copy()
107 for k, v in data.items():
108 del locale.locale_alias[k]
109 if locale.normalize(k) != v:
110 locale.locale_alias[k] = v
111 newdata = locale.locale_alias
112 errors = check(data)
113 locale.locale_alias = locale_alias
114 if errors:
115 sys.exit(1)
116 return newdata
117
118 def check(data):
119 # Check that all alias definitions from the X11 file
120 # are actually mapped to the correct alias locales.
121 errors = 0
122 for k, v in data.items():
123 if locale.normalize(k) != v:
124 print('ERROR: %a -> %a != %a' % (k, locale.normalize(k), v),
125 file=sys.stderr)
126 errors += 1
127 return errors
128
129 if __name__ == '__main__':
130 import argparse
131 parser = argparse.ArgumentParser()
132 parser.add_argument('--locale-alias', default=LOCALE_ALIAS,
133 help='location of the X11 alias file '
134 '(default: %a)' % LOCALE_ALIAS)
135 parser.add_argument('--glibc-supported', default=SUPPORTED,
136 help='location of the glibc SUPPORTED locales file '
137 '(default: %a)' % SUPPORTED)
138 args = parser.parse_args()
139
140 data = locale.locale_alias.copy()
141 data.update(parse_glibc_supported(args.glibc_supported))
142 data.update(parse(args.locale_alias))
143 while True:
144 # Repeat optimization while the size is decreased.
145 n = len(data)
146 data = optimize(data)
147 if len(data) == n:
148 break
149 print_differences(data, locale.locale_alias)
150 print()
151 print('locale_alias = {')
152 pprint(data)
153 print('}')