Mercurial > babel > mirror
annotate babel/messages/mofile.py @ 173:c100331c727c trunk
Forgot to fix the pluralizable messages, regarding #28.
| author | palgarvio |
|---|---|
| date | Tue, 26 Jun 2007 08:41:29 +0000 |
| parents | 208c7dfb9041 |
| children | bd256296086c |
| rev | line source |
|---|---|
| 160 | 1 # -*- coding: utf-8 -*- |
| 2 # | |
| 3 # Copyright (C) 2007 Edgewall Software | |
| 4 # All rights reserved. | |
| 5 # | |
| 6 # This software is licensed as described in the file COPYING, which | |
| 7 # you should have received as part of this distribution. The terms | |
| 8 # are also available at http://babel.edgewall.org/wiki/License. | |
| 9 # | |
| 10 # This software consists of voluntary contributions made by many | |
| 11 # individuals. For the exact contribution history, see the revision | |
| 12 # history and logs, available at http://babel.edgewall.org/log/. | |
| 13 | |
| 14 """Writing of files in the ``gettext`` MO (machine object) format. | |
| 15 | |
| 16 :see: `The Format of MO Files | |
| 17 <http://www.gnu.org/software/gettext/manual/gettext.html#MO-Files>`_ | |
| 18 """ | |
| 19 | |
| 20 import array | |
| 21 import struct | |
| 22 | |
| 161 | 23 __all__ = ['write_mo'] |
| 24 __docformat__ = 'restructuredtext en' | |
| 25 | |
| 160 | 26 def write_mo(fileobj, catalog, use_fuzzy=False): |
| 27 """Write a catalog to the specified file-like object using the GNU MO file | |
| 28 format. | |
| 29 | |
| 30 >>> from babel.messages import Catalog | |
| 31 >>> from gettext import GNUTranslations | |
| 32 >>> from StringIO import StringIO | |
| 33 | |
| 34 >>> catalog = Catalog(locale='en_US') | |
| 35 >>> catalog.add('foo', 'Voh') | |
| 36 >>> catalog.add((u'bar', u'baz'), (u'Bahr', u'Batz')) | |
| 37 >>> catalog.add('fuz', 'Futz', flags=['fuzzy']) | |
| 172 | 38 >>> catalog.add('Fizz', '') |
| 160 | 39 >>> buf = StringIO() |
| 40 | |
| 41 >>> write_mo(buf, catalog) | |
| 42 >>> buf.seek(0) | |
| 43 >>> translations = GNUTranslations(fp=buf) | |
| 44 >>> translations.ugettext('foo') | |
| 45 u'Voh' | |
| 46 >>> translations.ungettext('bar', 'baz', 1) | |
| 47 u'Bahr' | |
| 48 >>> translations.ungettext('bar', 'baz', 2) | |
| 49 u'Batz' | |
| 50 >>> translations.ugettext('fuz') | |
| 51 u'fuz' | |
| 172 | 52 >>> translations.ugettext('Fizz') |
| 53 u'Fizz' | |
| 160 | 54 |
| 55 :param fileobj: the file-like object to write to | |
| 56 :param catalog: the `Catalog` instance | |
| 57 :param use_fuzzy: whether translations marked as "fuzzy" should be included | |
| 58 in the output | |
| 59 """ | |
| 60 messages = list(catalog) | |
| 61 if not use_fuzzy: | |
| 62 messages[1:] = [m for m in messages[1:] if not m.fuzzy] | |
| 63 messages.sort(lambda x,y: cmp(x.id, y.id)) | |
| 64 | |
| 65 ids = strs = '' | |
| 66 offsets = [] | |
| 67 | |
| 68 for message in messages: | |
| 69 # For each string, we need size and file offset. Each string is NUL | |
| 70 # terminated; the NUL does not count into the size. | |
| 71 if message.pluralizable: | |
| 72 msgid = '\x00'.join([ | |
| 73 msgid.encode(catalog.charset) for msgid in message.id | |
| 74 ]) | |
|
173
c100331c727c
Forgot to fix the pluralizable messages, regarding #28.
palgarvio
parents:
172
diff
changeset
|
75 msgstrs = [] |
|
c100331c727c
Forgot to fix the pluralizable messages, regarding #28.
palgarvio
parents:
172
diff
changeset
|
76 for idx, string in enumerate(message.string): |
|
c100331c727c
Forgot to fix the pluralizable messages, regarding #28.
palgarvio
parents:
172
diff
changeset
|
77 if not string: |
|
c100331c727c
Forgot to fix the pluralizable messages, regarding #28.
palgarvio
parents:
172
diff
changeset
|
78 msgstrs.append(message.id[idx]) |
|
c100331c727c
Forgot to fix the pluralizable messages, regarding #28.
palgarvio
parents:
172
diff
changeset
|
79 else: |
|
c100331c727c
Forgot to fix the pluralizable messages, regarding #28.
palgarvio
parents:
172
diff
changeset
|
80 msgstrs.append(string) |
| 160 | 81 msgstr = '\x00'.join([ |
|
173
c100331c727c
Forgot to fix the pluralizable messages, regarding #28.
palgarvio
parents:
172
diff
changeset
|
82 msgstr.encode(catalog.charset) for msgstr in msgstrs |
| 160 | 83 ]) |
| 84 else: | |
| 85 msgid = message.id.encode(catalog.charset) | |
| 172 | 86 if not message.string: |
| 87 msgstr = message.id.encode(catalog.charset) | |
| 88 else: | |
| 89 msgstr = message.string.encode(catalog.charset) | |
| 160 | 90 offsets.append((len(ids), len(msgid), len(strs), len(msgstr))) |
| 91 ids += msgid + '\x00' | |
| 92 strs += msgstr + '\x00' | |
| 93 | |
| 94 # The header is 7 32-bit unsigned integers. We don't use hash tables, so | |
| 95 # the keys start right after the index tables. | |
| 96 keystart = 7 * 4 + 16 * len(messages) | |
| 97 valuestart = keystart + len(ids) | |
| 98 | |
| 99 # The string table first has the list of keys, then the list of values. | |
| 100 # Each entry has first the size of the string, then the file offset. | |
| 101 koffsets = [] | |
| 102 voffsets = [] | |
| 103 for o1, l1, o2, l2 in offsets: | |
| 104 koffsets += [l1, o1 + keystart] | |
| 105 voffsets += [l2, o2 + valuestart] | |
| 106 offsets = koffsets + voffsets | |
| 107 | |
| 108 fileobj.write(struct.pack('Iiiiiii', | |
| 109 0x950412deL, # magic | |
| 110 0, # version | |
| 111 len(messages), # number of entries | |
| 112 7 * 4, # start of key index | |
| 113 7 * 4 + len(messages) * 8, # start of value index | |
| 114 0, 0 # size and offset of hash table | |
| 115 ) + array.array("i", offsets).tostring() + ids + strs) |