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
|
# -*- coding: utf-8 -*-
"""
babel.messages.mofile
~~~~~~~~~~~~~~~~~~~~~
Writing of files in the ``gettext`` MO (machine object) format.
:copyright: (c) 2013-2018 by the Babel Team.
:license: BSD, see LICENSE for more details.
"""
import array
import struct
from babel.messages.catalog import Catalog, Message
from babel._compat import range_type, array_tobytes
LE_MAGIC = 0x950412de
BE_MAGIC = 0xde120495
def read_mo(fileobj):
"""Read a binary MO file from the given file-like object and return a
corresponding `Catalog` object.
:param fileobj: the file-like object to read the MO file from
:note: The implementation of this function is heavily based on the
``GNUTranslations._parse`` method of the ``gettext`` module in the
standard library.
"""
catalog = Catalog()
headers = {}
filename = getattr(fileobj, 'name', '')
buf = fileobj.read()
buflen = len(buf)
unpack = struct.unpack
# Parse the .mo file header, which consists of 5 little endian 32
# bit words.
magic = unpack('<I', buf[:4])[0] # Are we big endian or little endian?
if magic == LE_MAGIC:
version, msgcount, origidx, transidx = unpack('<4I', buf[4:20])
ii = '<II'
elif magic == BE_MAGIC:
version, msgcount, origidx, transidx = unpack('>4I', buf[4:20])
ii = '>II'
else:
raise IOError(0, 'Bad magic number', filename)
# Now put all messages from the .mo file buffer into the catalog
# dictionary
for i in range_type(0, msgcount):
mlen, moff = unpack(ii, buf[origidx:origidx + 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 IOError(0, 'File is corrupt', filename)
# See if we're looking at GNU .mo conventions for metadata
if mlen == 0:
# Catalog description
lastkey = key = None
for item in tmsg.splitlines():
item = item.strip()
if not item:
continue
if b':' in item:
key, value = item.split(b':', 1)
lastkey = key = key.strip().lower()
headers[key] = value.strip()
elif lastkey:
headers[lastkey] += b'\n' + item
if b'\x04' in msg: # context
ctxt, msg = msg.split(b'\x04')
else:
ctxt = None
if b'\x00' in msg: # plural forms
msg = msg.split(b'\x00')
tmsg = tmsg.split(b'\x00')
if catalog.charset:
msg = [x.decode(catalog.charset) for x in msg]
tmsg = [x.decode(catalog.charset) for x in tmsg]
else:
if catalog.charset:
msg = msg.decode(catalog.charset)
tmsg = tmsg.decode(catalog.charset)
catalog[msg] = Message(msg, tmsg, context=ctxt)
# advance to next entry in the seek tables
origidx += 8
transidx += 8
catalog.mime_headers = headers.items()
return catalog
def write_mo(fileobj, catalog, use_fuzzy=False):
"""Write a catalog to the specified file-like object using the GNU MO file
format.
>>> import sys
>>> from babel.messages import Catalog
>>> from gettext import GNUTranslations
>>> from babel._compat import BytesIO
>>> catalog = Catalog(locale='en_US')
>>> catalog.add('foo', 'Voh')
<Message ...>
>>> catalog.add((u'bar', u'baz'), (u'Bahr', u'Batz'))
<Message ...>
>>> catalog.add('fuz', 'Futz', flags=['fuzzy'])
<Message ...>
>>> catalog.add('Fizz', '')
<Message ...>
>>> catalog.add(('Fuzz', 'Fuzzes'), ('', ''))
<Message ...>
>>> buf = BytesIO()
>>> write_mo(buf, catalog)
>>> x = buf.seek(0)
>>> translations = GNUTranslations(fp=buf)
>>> if sys.version_info[0] >= 3:
... translations.ugettext = translations.gettext
... translations.ungettext = translations.ngettext
>>> translations.ugettext('foo')
u'Voh'
>>> translations.ungettext('bar', 'baz', 1)
u'Bahr'
>>> translations.ungettext('bar', 'baz', 2)
u'Batz'
>>> translations.ugettext('fuz')
u'fuz'
>>> translations.ugettext('Fizz')
u'Fizz'
>>> translations.ugettext('Fuzz')
u'Fuzz'
>>> translations.ugettext('Fuzzes')
u'Fuzzes'
:param fileobj: the file-like object to write to
:param catalog: the `Catalog` instance
:param use_fuzzy: whether translations marked as "fuzzy" should be included
in the output
"""
messages = list(catalog)
messages[1:] = [m for m in messages[1:]
if m.string and (use_fuzzy or not m.fuzzy)]
messages.sort()
ids = strs = b''
offsets = []
for message in messages:
# For each string, we need size and file offset. Each string is NUL
# terminated; the NUL does not count into the size.
if message.pluralizable:
msgid = b'\x00'.join([
msgid.encode(catalog.charset) for msgid in message.id
])
msgstrs = []
for idx, string in enumerate(message.string):
if not string:
msgstrs.append(message.id[min(int(idx), 1)])
else:
msgstrs.append(string)
msgstr = b'\x00'.join([
msgstr.encode(catalog.charset) for msgstr in msgstrs
])
else:
msgid = message.id.encode(catalog.charset)
msgstr = message.string.encode(catalog.charset)
if message.context:
msgid = b'\x04'.join([message.context.encode(catalog.charset),
msgid])
offsets.append((len(ids), len(msgid), len(strs), len(msgstr)))
ids += msgid + b'\x00'
strs += msgstr + b'\x00'
# The header is 7 32-bit unsigned integers. We don't use hash tables, so
# the keys start right after the index tables.
keystart = 7 * 4 + 16 * len(messages)
valuestart = keystart + len(ids)
# The string table first has the list of keys, then the list of values.
# Each entry has first the size of the string, then the file offset.
koffsets = []
voffsets = []
for o1, l1, o2, l2 in offsets:
koffsets += [l1, o1 + keystart]
voffsets += [l2, o2 + valuestart]
offsets = koffsets + voffsets
fileobj.write(struct.pack('Iiiiiii',
LE_MAGIC, # magic
0, # version
len(messages), # number of entries
7 * 4, # start of key index
7 * 4 + len(messages) * 8, # start of value index
0, 0 # size and offset of hash table
) + array_tobytes(array.array("i", offsets)) + ids + strs)
|
