Repository navigation
Expand file tree
/
Copy pathgettext.py
More file actions
657 lines (564 loc) · 21 KB
/
Copy pathgettext.py
File metadata and controls
657 lines (564 loc) · 21 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
"""Internationalization and localization support.
This module provides internationalization (I18N) and localization (L10N)
support for your Python programs by providing an interface to the GNU gettext
message catalog library.
I18N refers to the operation by which a program is made aware of multiple
languages. L10N refers to the adaptation of your program, once
internationalized, to the local language and cultural habits.
"""
# This module represents the integration of work, contributions, feedback, and
# suggestions from the following people:
#
# Martin von Loewis, who wrote the initial implementation of the underlying
# C-based libintlmodule (later renamed _gettext), along with a skeletal
# gettext.py implementation.
#
# Peter Funk, who wrote fintl.py, a fairly complete wrapper around intlmodule,
# which also included a pure-Python implementation to read .mo files if
# intlmodule wasn't available.
#
# James Henstridge, who also wrote a gettext.py module, which has some
# interesting, but currently unsupported experimental features: the notion of
# a Catalog class and instances, and the ability to add to a catalog file via
# a Python API.
#
# Barry Warsaw integrated these modules, wrote the .install() API and code,
# and conformed all C and Python code to Python's coding standards.
#
# Francois Pinard and Marc-Andre Lemburg also contributed valuably to this
# module.
#
# J. David Ibanez implemented plural forms. Bruno Haible fixed some bugs.
#
# TODO:
# - Lazy loading of .mo files. Currently the entire catalog is loaded into
# memory, but that's probably bad for large translated programs. Instead,
# the lexical sort of original strings in GNU .mo files should be exploited
# to do binary searches and lazy initializations. Or you might want to use
# the undocumented double-hash algorithm for .mo files with hash tables, but
# you'll need to study the GNU gettext code to do this.
#
# - Support Solaris .mo file formats. Unfortunately, we've been unable to
# find this format documented anywhere.
import operator
import os
import re
import sys
__all__ = ['NullTranslations', 'GNUTranslations', 'Catalog',
'bindtextdomain', 'find', 'translation', 'install',
'textdomain', 'dgettext', 'dngettext', 'gettext',
'ngettext', 'pgettext', 'dpgettext', 'npgettext',
'dnpgettext'
]
_default_localedir = os.path.join(sys.base_prefix, 'share', 'locale')
# Expression parsing for plural form selection.
#
# The gettext library supports a small subset of C syntax. The only
# incompatible difference is that integer literals starting with zero are
# decimal.
#
# https://www.gnu.org/software/gettext/manual/gettext.html#Plural-forms
# http://git.savannah.gnu.org/cgit/gettext.git/tree/gettext-runtime/intl/plural.y
_token_pattern = re.compile(r"""
(?P<WHITESPACES>[ \t]+) | # spaces and horizontal tabs
(?P<NUMBER>[0-9]+\b) | # decimal integer
(?P<NAME>n\b) | # only n is allowed
(?P<PARENTHESIS>[()]) |
(?P<OPERATOR>[-*/%+?:]|[><!]=?|==|&&|\|\|) | # !, *, /, %, +, -, <, >,
# <=, >=, ==, !=, &&, ||,
# ? :
# unary and bitwise ops
# not allowed
(?P<INVALID>\w+|.) # invalid token
""", re.VERBOSE|re.DOTALL)
def _tokenize(plural):
for mo in re.finditer(_token_pattern, plural):
kind = mo.lastgroup
if kind == 'WHITESPACES':
continue
value = mo.group(kind)
if kind == 'INVALID':
raise ValueError('invalid token in plural form: %s' % value)
yield value
yield ''
def _error(value):
if value:
return ValueError('unexpected token in plural form: %s' % value)
else:
return ValueError('unexpected end of plural form')
_binary_ops = (
('||',),
('&&',),
('==', '!='),
('<', '>', '<=', '>='),
('+', '-'),
('*', '/', '%'),
)
_binary_ops = {op: i for i, ops in enumerate(_binary_ops, 1) for op in ops}
_c2py_ops = {'||': 'or', '&&': 'and', '/': '//'}
def _parse(tokens, priority=-1):
result = ''
nexttok = next(tokens)
while nexttok == '!':
result += 'not '
nexttok = next(tokens)
if nexttok == '(':
sub, nexttok = _parse(tokens)
result = '%s(%s)' % (result, sub)
if nexttok != ')':
raise ValueError('unbalanced parenthesis in plural form')
elif nexttok == 'n':
result = '%s%s' % (result, nexttok)
else:
try:
value = int(nexttok, 10)
except ValueError:
raise _error(nexttok) from None
result = '%s%d' % (result, value)
nexttok = next(tokens)
j = 100
while nexttok in _binary_ops:
i = _binary_ops[nexttok]
if i < priority:
break
# Break chained comparisons
if i in (3, 4) and j in (3, 4): # '==', '!=', '<', '>', '<=', '>='
result = '(%s)' % result
# Replace some C operators by their Python equivalents
op = _c2py_ops.get(nexttok, nexttok)
right, nexttok = _parse(tokens, i + 1)
result = '%s %s %s' % (result, op, right)
j = i
if j == priority == 4: # '<', '>', '<=', '>='
result = '(%s)' % result
if nexttok == '?' and priority <= 0:
if_true, nexttok = _parse(tokens, 0)
if nexttok != ':':
raise _error(nexttok)
if_false, nexttok = _parse(tokens)
result = '%s if %s else %s' % (if_true, result, if_false)
if priority == 0:
result = '(%s)' % result
return result, nexttok
def _as_int(n):
try:
round(n)
except TypeError:
raise TypeError('Plural value must be an integer, got %s' %
(n.__class__.__name__,)) from None
return _as_int2(n)
def _as_int2(n):
try:
return operator.index(n)
except TypeError:
pass
import warnings
frame = sys._getframe(1)
stacklevel = 2
while frame.f_back is not None and frame.f_globals.get('__name__') == __name__:
stacklevel += 1
frame = frame.f_back
warnings.warn('Plural value must be an integer, got %s' %
(n.__class__.__name__,),
DeprecationWarning,
stacklevel)
return n
def c2py(plural):
"""Gets a C expression as used in PO files for plural forms and returns a
Python function that implements an equivalent expression.
"""
if len(plural) > 1000:
raise ValueError('plural form expression is too long')
try:
result, nexttok = _parse(_tokenize(plural))
if nexttok:
raise _error(nexttok)
depth = 0
for c in result:
if c == '(':
depth += 1
if depth > 20:
# Python compiler limit is about 90.
# The most complex example has 2.
raise ValueError('plural form expression is too complex')
elif c == ')':
depth -= 1
ns = {'_as_int': _as_int, '__name__': __name__}
exec('''if True:
def func(n):
if not isinstance(n, int):
n = _as_int(n)
return int(%s)
''' % result, ns)
return ns['func']
except RecursionError:
# Recursion error can be raised in _parse() or exec().
raise ValueError('plural form expression is too complex')
def _expand_lang(loc):
import locale
loc = locale.normalize(loc)
COMPONENT_CODESET = 1 << 0
COMPONENT_TERRITORY = 1 << 1
COMPONENT_MODIFIER = 1 << 2
# split up the locale into its base components
mask = 0
pos = loc.find('@')
if pos >= 0:
modifier = loc[pos:]
loc = loc[:pos]
mask |= COMPONENT_MODIFIER
else:
modifier = ''
pos = loc.find('.')
if pos >= 0:
codeset = loc[pos:]
loc = loc[:pos]
mask |= COMPONENT_CODESET
else:
codeset = ''
pos = loc.find('_')
if pos >= 0:
territory = loc[pos:]
loc = loc[:pos]
mask |= COMPONENT_TERRITORY
else:
territory = ''
language = loc
ret = []
for i in range(mask+1):
if not (i & ~mask): # if all components for this combo exist ...
val = language
if i & COMPONENT_TERRITORY: val += territory
if i & COMPONENT_CODESET: val += codeset
if i & COMPONENT_MODIFIER: val += modifier
ret.append(val)
ret.reverse()
return ret
class NullTranslations:
def __init__(self, fp=None):
self._info = {}
self._charset = None
self._fallback = None
if fp is not None:
self._parse(fp)
def _parse(self, fp):
pass
def add_fallback(self, fallback):
if self._fallback:
self._fallback.add_fallback(fallback)
else:
self._fallback = fallback
def gettext(self, message):
if self._fallback:
return self._fallback.gettext(message)
return message
def ngettext(self, msgid1, msgid2, n):
if self._fallback:
return self._fallback.ngettext(msgid1, msgid2, n)
n = _as_int2(n)
if n == 1:
return msgid1
else:
return msgid2
def pgettext(self, context, message):
if self._fallback:
return self._fallback.pgettext(context, message)
return message
def npgettext(self, context, msgid1, msgid2, n):
if self._fallback:
return self._fallback.npgettext(context, msgid1, msgid2, n)
n = _as_int2(n)
if n == 1:
return msgid1
else:
return msgid2
def info(self):
return self._info
def charset(self):
return self._charset
def install(self, names=None):
import builtins
builtins.__dict__['_'] = self.gettext
if names is not None:
allowed = {'gettext', 'ngettext', 'npgettext', 'pgettext'}
for name in allowed & set(names):
builtins.__dict__[name] = getattr(self, name)
class GNUTranslations(NullTranslations):
# Magic number of .mo files
LE_MAGIC = 0x950412de
BE_MAGIC = 0xde120495
# The encoding of a msgctxt and a msgid in a .mo file is
# msgctxt + "\x04" + msgid (gettext version >= 0.15)
CONTEXT = "%s\x04%s"
# Acceptable .mo versions
VERSIONS = (0, 1)
def _get_versions(self, version):
"""Returns a tuple of major version, minor version"""
return (version >> 16, version & 0xffff)
def _parse(self, fp):
"""Override this method to support alternative .mo formats."""
# Delay struct import for speeding up gettext import when .mo files
# are not used.
from struct import unpack
filename = getattr(fp, 'name', '')
# Parse the .mo file header, which consists of 5 little endian 32
# bit words.
self._catalog = catalog = {}
self.plural = lambda n: int(n != 1) # germanic plural by default
buf = fp.read()
buflen = len(buf)
# Are we big endian or little endian?
magic = unpack('<I', buf[:4])[0]
if magic == self.LE_MAGIC:
version, msgcount, masteridx, transidx = unpack('<4I', buf[4:20])
ii = '<II'
elif magic == self.BE_MAGIC:
version, msgcount, masteridx, transidx = unpack('>4I', buf[4:20])
ii = '>II'
else:
raise OSError(0, 'Bad magic number', filename)
major_version, minor_version = self._get_versions(version)
if major_version not in self.VERSIONS:
raise OSError(0, 'Bad version number ' + str(major_version), filename)
# Now put all messages from the .mo file buffer into the catalog
# dictionary.
for i in range(0, msgcount):
mlen, moff = unpack(ii, buf[masteridx:masteridx+8])
mend = moff + mlen
tlen, toff = unpack(ii, buf[transidx:transidx+8])
tend = toff + tlen
if mend < buflen and tend < buflen:
msg = buf[moff:mend]
tmsg = buf[toff:tend]
else:
raise OSError(0, 'File is corrupt', filename)
# See if we're looking at GNU .mo conventions for metadata
if mlen == 0:
# Catalog description
lastk = None
for b_item in tmsg.split(b'\n'):
item = b_item.decode().strip()
if not item:
continue
# Skip over comment lines:
if item.startswith('#-#-#-#-#') and item.endswith('#-#-#-#-#'):
continue
k = v = None
if ':' in item:
k, v = item.split(':', 1)
k = k.strip().lower()
v = v.strip()
self._info[k] = v
lastk = k
elif lastk:
self._info[lastk] += '\n' + item
if k == 'content-type':
self._charset = v.split('charset=')[1]
elif k == 'plural-forms':
v = v.split(';')
plural = v[1].split('plural=')[1]
self.plural = c2py(plural)
# Note: we unconditionally convert both msgids and msgstrs to
# Unicode using the character encoding specified in the charset
# parameter of the Content-Type header. The gettext documentation
# strongly encourages msgids to be us-ascii, but some applications
# require alternative encodings (e.g. Zope's ZCML and ZPT). For
# traditional gettext applications, the msgid conversion will
# cause no problems since us-ascii should always be a subset of
# the charset encoding. We may want to fall back to 8-bit msgids
# if the Unicode conversion fails.
charset = self._charset or 'ascii'
if b'\x00' in msg:
# Plural forms
msgid1, msgid2 = msg.split(b'\x00')
tmsg = tmsg.split(b'\x00')
msgid1 = str(msgid1, charset)
for i, x in enumerate(tmsg):
catalog[(msgid1, i)] = str(x, charset)
else:
catalog[str(msg, charset)] = str(tmsg, charset)
# advance to next entry in the seek tables
masteridx += 8
transidx += 8
def gettext(self, message):
missing = object()
tmsg = self._catalog.get(message, missing)
if tmsg is missing:
tmsg = self._catalog.get((message, self.plural(1)), missing)
if tmsg is not missing:
return tmsg
if self._fallback:
return self._fallback.gettext(message)
return message
def ngettext(self, msgid1, msgid2, n):
try:
tmsg = self._catalog[(msgid1, self.plural(n))]
except KeyError:
if self._fallback:
return self._fallback.ngettext(msgid1, msgid2, n)
if n == 1:
tmsg = msgid1
else:
tmsg = msgid2
return tmsg
def pgettext(self, context, message):
ctxt_msg_id = self.CONTEXT % (context, message)
missing = object()
tmsg = self._catalog.get(ctxt_msg_id, missing)
if tmsg is missing:
tmsg = self._catalog.get((ctxt_msg_id, self.plural(1)), missing)
if tmsg is not missing:
return tmsg
if self._fallback:
return self._fallback.pgettext(context, message)
return message
def npgettext(self, context, msgid1, msgid2, n):
ctxt_msg_id = self.CONTEXT % (context, msgid1)
try:
tmsg = self._catalog[ctxt_msg_id, self.plural(n)]
except KeyError:
if self._fallback:
return self._fallback.npgettext(context, msgid1, msgid2, n)
if n == 1:
tmsg = msgid1
else:
tmsg = msgid2
return tmsg
# Locate a .mo file using the gettext strategy
def find(domain, localedir=None, languages=None, all=False):
# Get some reasonable defaults for arguments that were not supplied
if localedir is None:
localedir = _default_localedir
if languages is None:
languages = []
for envar in ('LANGUAGE', 'LC_ALL', 'LC_MESSAGES', 'LANG'):
val = os.environ.get(envar)
if val:
languages = val.split(':')
break
if 'C' not in languages:
languages.append('C')
# now normalize and expand the languages
nelangs = []
for lang in languages:
for nelang in _expand_lang(lang):
if nelang not in nelangs:
nelangs.append(nelang)
# select a language
if all:
result = []
else:
result = None
for lang in nelangs:
if lang == 'C':
break
mofile = os.path.join(localedir, lang, 'LC_MESSAGES', '%s.mo' % domain)
if os.path.exists(mofile):
if all:
result.append(mofile)
else:
return mofile
return result
# a mapping between absolute .mo file path and Translation object
_translations = {}
def translation(domain, localedir=None, languages=None,
class_=None, fallback=False):
if class_ is None:
class_ = GNUTranslations
mofiles = find(domain, localedir, languages, all=True)
if not mofiles:
if fallback:
return NullTranslations()
from errno import ENOENT
raise FileNotFoundError(ENOENT,
'No translation file found for domain', domain)
# Avoid opening, reading, and parsing the .mo file after it's been done
# once.
result = None
for mofile in mofiles:
key = (class_, os.path.abspath(mofile))
t = _translations.get(key)
if t is None:
with open(mofile, 'rb') as fp:
t = _translations.setdefault(key, class_(fp))
# Copy the translation object to allow setting fallbacks and
# output charset. All other instance data is shared with the
# cached object.
# Delay copy import for speeding up gettext import when .mo files
# are not used.
import copy
t = copy.copy(t)
if result is None:
result = t
else:
result.add_fallback(t)
return result
def install(domain, localedir=None, *, names=None):
t = translation(domain, localedir, fallback=True)
t.install(names)
# a mapping b/w domains and locale directories
_localedirs = {}
# current global domain, `messages' used for compatibility w/ GNU gettext
_current_domain = 'messages'
def textdomain(domain=None):
global _current_domain
if domain is not None:
_current_domain = domain
return _current_domain
def bindtextdomain(domain, localedir=None):
global _localedirs
if localedir is not None:
_localedirs[domain] = localedir
return _localedirs.get(domain, _default_localedir)
def dgettext(domain, message):
try:
t = translation(domain, _localedirs.get(domain, None))
except OSError:
return message
return t.gettext(message)
def dngettext(domain, msgid1, msgid2, n):
try:
t = translation(domain, _localedirs.get(domain, None))
except OSError:
n = _as_int2(n)
if n == 1:
return msgid1
else:
return msgid2
return t.ngettext(msgid1, msgid2, n)
def dpgettext(domain, context, message):
try:
t = translation(domain, _localedirs.get(domain, None))
except OSError:
return message
return t.pgettext(context, message)
def dnpgettext(domain, context, msgid1, msgid2, n):
try:
t = translation(domain, _localedirs.get(domain, None))
except OSError:
n = _as_int2(n)
if n == 1:
return msgid1
else:
return msgid2
return t.npgettext(context, msgid1, msgid2, n)
def gettext(message):
return dgettext(_current_domain, message)
def ngettext(msgid1, msgid2, n):
return dngettext(_current_domain, msgid1, msgid2, n)
def pgettext(context, message):
return dpgettext(_current_domain, context, message)
def npgettext(context, msgid1, msgid2, n):
return dnpgettext(_current_domain, context, msgid1, msgid2, n)
# dcgettext() has been deemed unnecessary and is not implemented.
# James Henstridge's Catalog constructor from GNOME gettext. Documented usage
# was:
#
# import gettext
# cat = gettext.Catalog(PACKAGE, localedir=LOCALEDIR)
# _ = cat.gettext
# print _('Hello World')
# The resulting catalog object currently don't support access through a
# dictionary API, which was supported (but apparently unused) in GNOME
# gettext.
Catalog = translation