1
1
"translation helper functions"
2
2
3
-
import os, re, sys
3
+
import locale
4
+
import os
5
+
import re
6
+
import sys
4
7
import gettext as gettext_module
5
8
from cStringIO import StringIO
6
9
from django.utils.functional import lazy
@@ -25,15 +28,25 @@ def currentThread():
25
28
# The default translation is based on the settings file.
26
29
_default = None
27
30
28
-
# This is a cache for accept-header to translation object mappings to prevent
29
-
# the accept parser to run multiple times for one user.
31
+
# This is a cache for normalised accept-header languages to prevent multiple
32
+
# file lookups when checking the same locale on repeated requests.
30
33
_accepted = {}
31
34
32
-
def to_locale(language):
35
+
# Format of Accept-Language header values. From RFC 2616, section 14.4 and 3.9.
36
+
accept_language_re = re.compile(r'''
37
+
([A-Za-z]{1,8}(?:-[A-Za-z]{1,8})*|\*) # "en", "en-au", "x-y-z", "*"
38
+
(?:;q=(0(?:\.\d{,3})?|1(?:.0{,3})?))? # Optional "q=1.00", "q=0.8"
39
+
(?:\s*,\s*|$) # Multiple accepts per header.
40
+
''', re.VERBOSE)
41
+
42
+
def to_locale(language, to_lower=False):
33
43
"Turns a language name (en-us) into a locale name (en_US)."
34
44
p = language.find('-')
35
45
if p >= 0:
36
-
return language[:p].lower()+'_'+language[p+1:].upper()
46
+
if to_lower:
47
+
return language[:p].lower()+'_'+language[p+1:].lower()
48
+
else:
49
+
return language[:p].lower()+'_'+language[p+1:].upper()
37
50
else:
38
51
return language.lower()
39
52
@@ -297,46 +310,40 @@ def get_language_from_request(request):
297
310
if lang_code in supported and lang_code is not None and check_for_language(lang_code):
298
311
return lang_code
299
312
300
-
lang_code = request.COOKIES.get('django_language', None)
301
-
if lang_code in supported and lang_code is not None and check_for_language(lang_code):
313
+
lang_code = request.COOKIES.get('django_language')
314
+
if lang_code and lang_code in supported and check_for_language(lang_code):
302
315
return lang_code
303
316
304
-
accept = request.META.get('HTTP_ACCEPT_LANGUAGE', None)
305
-
if accept is not None:
306
-
307
-
t = _accepted.get(accept, None)
308
-
if t is not None:
309
-
return t
310
-
311
-
def _parsed(el):
312
-
p = el.find(';q=')
313
-
if p >= 0:
314
-
lang = el[:p].strip()
315
-
order = int(float(el[p+3:].strip())*100)
316
-
else:
317
-
lang = el
318
-
order = 100
319
-
p = lang.find('-')
320
-
if p >= 0:
321
-
mainlang = lang[:p]
322
-
else:
323
-
mainlang = lang
324
-
return (lang, mainlang, order)
325
-
326
-
langs = [_parsed(el) for el in accept.split(',')]
327
-
langs.sort(lambda a,b: -1*cmp(a[2], b[2]))
328
-
329
-
for lang, mainlang, order in langs:
330
-
if lang in supported or mainlang in supported:
331
-
langfile = gettext_module.find('django', globalpath, [to_locale(lang)])
332
-
if langfile:
333
-
# reconstruct the actual language from the language
334
-
# filename, because otherwise we might incorrectly
335
-
# report de_DE if we only have de available, but
336
-
# did find de_DE because of language normalization
337
-
lang = langfile[len(globalpath):].split(os.path.sep)[1]
338
-
_accepted[accept] = lang
339
-
return lang
317
+
accept = request.META.get('HTTP_ACCEPT_LANGUAGE', '')
318
+
for lang, unused in parse_accept_lang_header(accept):
319
+
if lang == '*':
320
+
break
321
+
322
+
# We have a very restricted form for our language files (no encoding
323
+
# specifier, since they all must be UTF-8 and only one possible
324
+
# language each time. So we avoid the overhead of gettext.find() and
325
+
# look up the MO file manually.
326
+
327
+
normalized = locale.locale_alias.get(to_locale(lang, True))
328
+
if not normalized:
329
+
continue
330
+
331
+
# Remove the default encoding from locale_alias
332
+
normalized = normalized.split('.')[0]
333
+
334
+
if normalized in _accepted:
335
+
# We've seen this locale before and have an MO file for it, so no
336
+
# need to check again.
337
+
return _accepted[normalized]
338
+
339
+
for lang in (normalized, normalized.split('_')[0]):
340
+
if lang not in supported:
341
+
continue
342
+
langfile = os.path.join(globalpath, lang, 'LC_MESSAGES',
343
+
'django.mo')
344
+
if os.path.exists(langfile):
345
+
_accepted[normalized] = lang
346
+
return lang
340
347
341
348
return settings.LANGUAGE_CODE
342
349
@@ -457,3 +464,23 @@ def templatize(src):
457
464
else:
458
465
out.write(blankout(t.contents, 'X'))
459
466
return out.getvalue()
467
+
468
+
def parse_accept_lang_header(lang_string):
469
+
"""
470
+
Parses the lang_string, which is the body of an HTTP Accept-Language
471
+
header, and returns a list of (lang, q-value), ordered by 'q' values.
472
+
473
+
Any format errors in lang_string results in an empty list being returned.
474
+
"""
475
+
result = []
476
+
pieces = accept_language_re.split(lang_string)
477
+
if pieces[-1]:
478
+
return []
479
+
for i in range(0, len(pieces) - 1, 3):
480
+
first, lang, priority = pieces[i : i + 3]
481
+
if first:
482
+
return []
483
+
priority = priority and float(priority) or 1.0
484
+
result.append((lang, priority))
485
+
result.sort(lambda x, y: -cmp(x[1], y[1]))
486
+
return result
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4