Show More
This diff has been collapsed as it changes many lines, (1487 lines changed) Show them Hide them | |||
@@ -1,1680 +1,1639 b'' | |||
|
1 | 1 | # -*- coding: utf-8 -*- |
|
2 | 2 | # no-check-code |
|
3 | 3 | # |
|
4 | 4 | # License: MIT (see LICENSE file provided) |
|
5 | 5 | # vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4: |
|
6 | 6 | |
|
7 | 7 | """ |
|
8 | **polib** allows you to manipulate, create, modify gettext files (pot, po | |
|
9 |
|
|
|
10 |
|
|
|
11 | from scratch. | |
|
12 | ||
|
13 | **polib** provides a simple and pythonic API, exporting only three | |
|
14 | convenience functions (*pofile*, *mofile* and *detect_encoding*), and the | |
|
15 | four core classes, *POFile*, *MOFile*, *POEntry* and *MOEntry* for creating | |
|
16 | new files/entries. | |
|
17 | ||
|
18 | **Basic example**: | |
|
8 | **polib** allows you to manipulate, create, modify gettext files (pot, po and | |
|
9 | mo files). You can load existing files, iterate through it's entries, add, | |
|
10 | modify entries, comments or metadata, etc. or create new po files from scratch. | |
|
19 | 11 | |
|
20 | >>> import polib | |
|
21 | >>> # load an existing po file | |
|
22 | >>> po = polib.pofile('tests/test_utf8.po') | |
|
23 | >>> for entry in po: | |
|
24 | ... # do something with entry... | |
|
25 | ... pass | |
|
26 | >>> # add an entry | |
|
27 | >>> entry = polib.POEntry(msgid='Welcome', msgstr='Bienvenue') | |
|
28 | >>> entry.occurrences = [('welcome.py', '12'), ('anotherfile.py', '34')] | |
|
29 | >>> po.append(entry) | |
|
30 | >>> # to save our modified po file: | |
|
31 | >>> # po.save() | |
|
32 | >>> # or you may want to compile the po file | |
|
33 | >>> # po.save_as_mofile('tests/test_utf8.mo') | |
|
12 | **polib** provides a simple and pythonic API via the :func:`~polib.pofile` and | |
|
13 | :func:`~polib.mofile` convenience functions. | |
|
34 | 14 | """ |
|
35 | 15 | |
|
36 |
__author__ = 'David J |
|
|
37 |
__version__ = '0. |
|
|
16 | __author__ = 'David Jean Louis <izimobil@gmail.com>' | |
|
17 | __version__ = '0.6.4' | |
|
38 | 18 | __all__ = ['pofile', 'POFile', 'POEntry', 'mofile', 'MOFile', 'MOEntry', |
|
39 | 19 | 'detect_encoding', 'escape', 'unescape', 'detect_encoding',] |
|
40 | 20 | |
|
21 | import array | |
|
41 | 22 | import codecs |
|
23 | import os | |
|
24 | import re | |
|
42 | 25 | import struct |
|
26 | import sys | |
|
43 | 27 | import textwrap |
|
44 | 28 | import types |
|
45 | import re | |
|
29 | ||
|
46 | 30 | |
|
31 | # the default encoding to use when encoding cannot be detected | |
|
47 | 32 | default_encoding = 'utf-8' |
|
48 | 33 | |
|
49 | # function pofile() {{{ | |
|
50 | ||
|
51 | def pofile(fpath, **kwargs): | |
|
52 | """ | |
|
53 | Convenience function that parse the po/pot file *fpath* and return | |
|
54 | a POFile instance. | |
|
55 | ||
|
56 | **Keyword arguments**: | |
|
57 | - *fpath*: string, full or relative path to the po/pot file to parse | |
|
58 | - *wrapwidth*: integer, the wrap width, only useful when -w option was | |
|
59 | passed to xgettext (optional, default to 78) | |
|
60 | - *autodetect_encoding*: boolean, if set to False the function will | |
|
61 | not try to detect the po file encoding (optional, default to True) | |
|
62 | - *encoding*: string, an encoding, only relevant if autodetect_encoding | |
|
63 | is set to False | |
|
64 | - *check_for_duplicates*: whether to check for duplicate entries when | |
|
65 | adding entries to the file, default: False (optional) | |
|
66 | ||
|
67 | **Example**: | |
|
34 | # _pofile_or_mofile {{{ | |
|
68 | 35 | |
|
69 | >>> import polib | |
|
70 | >>> po = polib.pofile('tests/test_weird_occurrences.po', | |
|
71 | ... check_for_duplicates=True) | |
|
72 | >>> po #doctest: +ELLIPSIS | |
|
73 | <POFile instance at ...> | |
|
74 | >>> import os, tempfile | |
|
75 | >>> all_attrs = ('msgctxt', 'msgid', 'msgstr', 'msgid_plural', | |
|
76 | ... 'msgstr_plural', 'obsolete', 'comment', 'tcomment', | |
|
77 | ... 'occurrences', 'flags', 'previous_msgctxt', | |
|
78 | ... 'previous_msgid', 'previous_msgid_plural') | |
|
79 | >>> for fname in ['test_iso-8859-15.po', 'test_utf8.po']: | |
|
80 | ... orig_po = polib.pofile('tests/'+fname) | |
|
81 | ... tmpf = tempfile.NamedTemporaryFile().name | |
|
82 | ... orig_po.save(tmpf) | |
|
83 | ... try: | |
|
84 | ... new_po = polib.pofile(tmpf) | |
|
85 | ... for old, new in zip(orig_po, new_po): | |
|
86 | ... for attr in all_attrs: | |
|
87 | ... if getattr(old, attr) != getattr(new, attr): | |
|
88 | ... getattr(old, attr) | |
|
89 | ... getattr(new, attr) | |
|
90 | ... finally: | |
|
91 | ... os.unlink(tmpf) | |
|
92 | >>> po_file = polib.pofile('tests/test_save_as_mofile.po') | |
|
93 | >>> tmpf = tempfile.NamedTemporaryFile().name | |
|
94 | >>> po_file.save_as_mofile(tmpf) | |
|
95 | >>> try: | |
|
96 | ... mo_file = polib.mofile(tmpf) | |
|
97 | ... for old, new in zip(po_file, mo_file): | |
|
98 | ... if po_file._encode(old.msgid) != mo_file._encode(new.msgid): | |
|
99 | ... 'OLD: ', po_file._encode(old.msgid) | |
|
100 | ... 'NEW: ', mo_file._encode(new.msgid) | |
|
101 | ... if po_file._encode(old.msgstr) != mo_file._encode(new.msgstr): | |
|
102 | ... 'OLD: ', po_file._encode(old.msgstr) | |
|
103 | ... 'NEW: ', mo_file._encode(new.msgstr) | |
|
104 | ... print new.msgstr | |
|
105 | ... finally: | |
|
106 | ... os.unlink(tmpf) | |
|
36 | def _pofile_or_mofile(f, type, **kwargs): | |
|
37 | """ | |
|
38 | Internal function used by :func:`polib.pofile` and :func:`polib.mofile` to | |
|
39 | honor the DRY concept. | |
|
107 | 40 | """ |
|
108 | if kwargs.get('autodetect_encoding', True): | |
|
109 |
|
|
|
110 | else: | |
|
111 | enc = kwargs.get('encoding', default_encoding) | |
|
112 | check_for_duplicates = kwargs.get('check_for_duplicates', False) | |
|
113 | parser = _POFileParser( | |
|
114 | fpath, | |
|
41 | # get the file encoding | |
|
42 | enc = kwargs.get('encoding') | |
|
43 | if enc is None: | |
|
44 | enc = detect_encoding(f, type == 'mofile') | |
|
45 | ||
|
46 | # parse the file | |
|
47 | kls = type == 'pofile' and _POFileParser or _MOFileParser | |
|
48 | parser = kls( | |
|
49 | f, | |
|
115 | 50 | encoding=enc, |
|
116 | 51 | check_for_duplicates=kwargs.get('check_for_duplicates', False) |
|
117 | 52 | ) |
|
118 | 53 | instance = parser.parse() |
|
119 | 54 | instance.wrapwidth = kwargs.get('wrapwidth', 78) |
|
120 | 55 | return instance |
|
121 | 56 | |
|
122 | 57 | # }}} |
|
58 | # function pofile() {{{ | |
|
59 | ||
|
60 | def pofile(pofile, **kwargs): | |
|
61 | """ | |
|
62 | Convenience function that parses the po or pot file ``pofile`` and returns | |
|
63 | a :class:`~polib.POFile` instance. | |
|
64 | ||
|
65 | Arguments: | |
|
66 | ||
|
67 | ``pofile`` | |
|
68 | string, full or relative path to the po/pot file or its content (data). | |
|
69 | ||
|
70 | ``wrapwidth`` | |
|
71 | integer, the wrap width, only useful when the ``-w`` option was passed | |
|
72 | to xgettext (optional, default: ``78``). | |
|
73 | ||
|
74 | ``encoding`` | |
|
75 | string, the encoding to use (e.g. "utf-8") (default: ``None``, the | |
|
76 | encoding will be auto-detected). | |
|
77 | ||
|
78 | ``check_for_duplicates`` | |
|
79 | whether to check for duplicate entries when adding entries to the | |
|
80 | file (optional, default: ``False``). | |
|
81 | """ | |
|
82 | return _pofile_or_mofile(pofile, 'pofile', **kwargs) | |
|
83 | ||
|
84 | # }}} | |
|
123 | 85 | # function mofile() {{{ |
|
124 | 86 | |
|
125 |
def mofile( |
|
|
87 | def mofile(mofile, **kwargs): | |
|
126 | 88 | """ |
|
127 |
Convenience function that parse the mo file |
|
|
128 |
|
|
|
89 | Convenience function that parses the mo file ``mofile`` and returns a | |
|
90 | :class:`~polib.MOFile` instance. | |
|
129 | 91 | |
|
130 |
|
|
|
131 | - *fpath*: string, full or relative path to the mo file to parse | |
|
132 | - *wrapwidth*: integer, the wrap width, only useful when -w option was | |
|
133 | passed to xgettext to generate the po file that was used to format | |
|
134 | the mo file (optional, default to 78) | |
|
135 | - *autodetect_encoding*: boolean, if set to False the function will | |
|
136 | not try to detect the po file encoding (optional, default to True) | |
|
137 | - *encoding*: string, an encoding, only relevant if autodetect_encoding | |
|
138 | is set to False | |
|
139 | - *check_for_duplicates*: whether to check for duplicate entries when | |
|
140 | adding entries to the file, default: False (optional) | |
|
92 | Arguments: | |
|
141 | 93 | |
|
142 | **Example**: | |
|
94 | ``mofile`` | |
|
95 | string, full or relative path to the mo file or its content (data). | |
|
143 | 96 | |
|
144 | >>> import polib | |
|
145 | >>> mo = polib.mofile('tests/test_utf8.mo', check_for_duplicates=True) | |
|
146 | >>> mo #doctest: +ELLIPSIS | |
|
147 | <MOFile instance at ...> | |
|
148 | >>> import os, tempfile | |
|
149 | >>> for fname in ['test_iso-8859-15.mo', 'test_utf8.mo']: | |
|
150 | ... orig_mo = polib.mofile('tests/'+fname) | |
|
151 | ... tmpf = tempfile.NamedTemporaryFile().name | |
|
152 | ... orig_mo.save(tmpf) | |
|
153 | ... try: | |
|
154 | ... new_mo = polib.mofile(tmpf) | |
|
155 | ... for old, new in zip(orig_mo, new_mo): | |
|
156 | ... if old.msgid != new.msgid: | |
|
157 | ... old.msgstr | |
|
158 | ... new.msgstr | |
|
159 | ... finally: | |
|
160 | ... os.unlink(tmpf) | |
|
97 | ``wrapwidth`` | |
|
98 | integer, the wrap width, only useful when the ``-w`` option was passed | |
|
99 | to xgettext to generate the po file that was used to format the mo file | |
|
100 | (optional, default: ``78``). | |
|
101 | ||
|
102 | ``encoding`` | |
|
103 | string, the encoding to use (e.g. "utf-8") (default: ``None``, the | |
|
104 | encoding will be auto-detected). | |
|
105 | ||
|
106 | ``check_for_duplicates`` | |
|
107 | whether to check for duplicate entries when adding entries to the | |
|
108 | file (optional, default: ``False``). | |
|
161 | 109 | """ |
|
162 | if kwargs.get('autodetect_encoding', True): | |
|
163 | enc = detect_encoding(fpath, True) | |
|
164 | else: | |
|
165 | enc = kwargs.get('encoding', default_encoding) | |
|
166 | parser = _MOFileParser( | |
|
167 | fpath, | |
|
168 | encoding=enc, | |
|
169 | check_for_duplicates=kwargs.get('check_for_duplicates', False) | |
|
170 | ) | |
|
171 | instance = parser.parse() | |
|
172 | instance.wrapwidth = kwargs.get('wrapwidth', 78) | |
|
173 | return instance | |
|
110 | return _pofile_or_mofile(mofile, 'mofile', **kwargs) | |
|
174 | 111 | |
|
175 | 112 | # }}} |
|
176 | 113 | # function detect_encoding() {{{ |
|
177 | 114 | |
|
178 |
def detect_encoding(f |
|
|
115 | def detect_encoding(file, binary_mode=False): | |
|
179 | 116 | """ |
|
180 |
Try to detect the encoding used by the |
|
|
181 | return polib default *encoding* if it's unable to detect it. | |
|
117 | Try to detect the encoding used by the ``file``. The ``file`` argument can | |
|
118 | be a PO or MO file path or a string containing the contents of the file. | |
|
119 | If the encoding cannot be detected, the function will return the value of | |
|
120 | ``default_encoding``. | |
|
182 | 121 | |
|
183 |
|
|
|
184 | - *fpath*: string, full or relative path to the mo file to parse. | |
|
122 | Arguments: | |
|
123 | ||
|
124 | ``file`` | |
|
125 | string, full or relative path to the po/mo file or its content. | |
|
185 | 126 | |
|
186 | **Examples**: | |
|
127 | ``binary_mode`` | |
|
128 | boolean, set this to True if ``file`` is a mo file. | |
|
129 | """ | |
|
130 | rx = re.compile(r'"?Content-Type:.+? charset=([\w_\-:\.]+)') | |
|
187 | 131 | |
|
188 | >>> print(detect_encoding('tests/test_noencoding.po')) | |
|
189 | utf-8 | |
|
190 | >>> print(detect_encoding('tests/test_utf8.po')) | |
|
191 | UTF-8 | |
|
192 | >>> print(detect_encoding('tests/test_utf8.mo', True)) | |
|
193 | UTF-8 | |
|
194 | >>> print(detect_encoding('tests/test_iso-8859-15.po')) | |
|
195 | ISO_8859-15 | |
|
196 | >>> print(detect_encoding('tests/test_iso-8859-15.mo', True)) | |
|
197 | ISO_8859-15 | |
|
198 | """ | |
|
199 | import re | |
|
200 | rx = re.compile(r'"?Content-Type:.+? charset=([\w_\-:\.]+)') | |
|
201 | if binary_mode: | |
|
202 | mode = 'rb' | |
|
132 | def charset_exists(charset): | |
|
133 | """Check whether ``charset`` is valid or not.""" | |
|
134 | try: | |
|
135 | codecs.lookup(charset) | |
|
136 | except LookupError: | |
|
137 | return False | |
|
138 | return True | |
|
139 | ||
|
140 | if not os.path.exists(file): | |
|
141 | match = rx.search(file) | |
|
142 | if match: | |
|
143 | enc = match.group(1).strip() | |
|
144 | if charset_exists(enc): | |
|
145 | return enc | |
|
203 | 146 | else: |
|
204 | mode = 'r' | |
|
205 | f = open(fpath, mode) | |
|
206 | for l in f.readlines(): | |
|
207 | match = rx.search(l) | |
|
208 | if match: | |
|
209 | f.close() | |
|
210 |
|
|
|
211 | f.close() | |
|
147 | if binary_mode: | |
|
148 | mode = 'rb' | |
|
149 | else: | |
|
150 | mode = 'r' | |
|
151 | f = open(file, mode) | |
|
152 | for l in f.readlines(): | |
|
153 | match = rx.search(l) | |
|
154 | if match: | |
|
155 | f.close() | |
|
156 | enc = match.group(1).strip() | |
|
157 | if charset_exists(enc): | |
|
158 | return enc | |
|
159 | f.close() | |
|
212 | 160 | return default_encoding |
|
213 | 161 | |
|
214 | 162 | # }}} |
|
215 | 163 | # function escape() {{{ |
|
216 | 164 | |
|
217 | 165 | def escape(st): |
|
218 | 166 | """ |
|
219 | Escape special chars and return the given string *st*. | |
|
220 | ||
|
221 | **Examples**: | |
|
222 | ||
|
223 | >>> escape('\\t and \\n and \\r and " and \\\\') | |
|
224 | '\\\\t and \\\\n and \\\\r and \\\\" and \\\\\\\\' | |
|
167 | Escapes the characters ``\\\\``, ``\\t``, ``\\n``, ``\\r`` and ``"`` in | |
|
168 | the given string ``st`` and returns it. | |
|
225 | 169 | """ |
|
226 | 170 | return st.replace('\\', r'\\')\ |
|
227 | 171 | .replace('\t', r'\t')\ |
|
228 | 172 | .replace('\r', r'\r')\ |
|
229 | 173 | .replace('\n', r'\n')\ |
|
230 | 174 | .replace('\"', r'\"') |
|
231 | 175 | |
|
232 | 176 | # }}} |
|
233 | 177 | # function unescape() {{{ |
|
234 | 178 | |
|
235 | 179 | def unescape(st): |
|
236 | 180 | """ |
|
237 | Unescape special chars and return the given string *st*. | |
|
238 | ||
|
239 | **Examples**: | |
|
240 | ||
|
241 | >>> unescape('\\\\t and \\\\n and \\\\r and \\\\" and \\\\\\\\') | |
|
242 | '\\t and \\n and \\r and " and \\\\' | |
|
243 | >>> unescape(r'\\n') | |
|
244 | '\\n' | |
|
245 | >>> unescape(r'\\\\n') | |
|
246 | '\\\\n' | |
|
247 | >>> unescape(r'\\\\n\\n') | |
|
248 | '\\\\n\\n' | |
|
181 | Unescapes the characters ``\\\\``, ``\\t``, ``\\n``, ``\\r`` and ``"`` in | |
|
182 | the given string ``st`` and returns it. | |
|
249 | 183 | """ |
|
250 | 184 | def unescape_repl(m): |
|
251 | 185 | m = m.group(1) |
|
252 | 186 | if m == 'n': |
|
253 | 187 | return '\n' |
|
254 | 188 | if m == 't': |
|
255 | 189 | return '\t' |
|
256 | 190 | if m == 'r': |
|
257 | 191 | return '\r' |
|
258 | 192 | if m == '\\': |
|
259 | 193 | return '\\' |
|
260 | 194 | return m # handles escaped double quote |
|
261 | 195 | return re.sub(r'\\(\\|n|t|r|")', unescape_repl, st) |
|
262 | 196 | |
|
263 | 197 | # }}} |
|
264 | 198 | # class _BaseFile {{{ |
|
265 | 199 | |
|
266 | 200 | class _BaseFile(list): |
|
267 | 201 | """ |
|
268 |
Common |
|
|
269 |
This class |
|
|
202 | Common base class for the :class:`~polib.POFile` and :class:`~polib.MOFile` | |
|
203 | classes. This class should **not** be instanciated directly. | |
|
270 | 204 | """ |
|
271 | 205 | |
|
272 | 206 | def __init__(self, *args, **kwargs): |
|
273 | 207 | """ |
|
274 | Constructor. | |
|
208 | Constructor, accepts the following keyword arguments: | |
|
209 | ||
|
210 | ``pofile`` | |
|
211 | string, the path to the po or mo file, or its content as a string. | |
|
275 | 212 | |
|
276 | **Keyword arguments**: | |
|
277 | - *fpath*: string, path to po or mo file | |
|
278 | - *wrapwidth*: integer, the wrap width, only useful when -w option | |
|
279 | was passed to xgettext to generate the po file that was used to | |
|
280 | format the mo file, default to 78 (optional), | |
|
281 |
|
|
|
282 |
|
|
|
283 | - *check_for_duplicates*: whether to check for duplicate entries | |
|
284 | when adding entries to the file, default: False (optional). | |
|
213 | ``wrapwidth`` | |
|
214 | integer, the wrap width, only useful when the ``-w`` option was | |
|
215 | passed to xgettext (optional, default: ``78``). | |
|
216 | ||
|
217 | ``encoding`` | |
|
218 | string, the encoding to use, defaults to ``default_encoding`` | |
|
219 | global variable (optional). | |
|
220 | ||
|
221 | ``check_for_duplicates`` | |
|
222 | whether to check for duplicate entries when adding entries to the | |
|
223 | file, (optional, default: ``False``). | |
|
285 | 224 | """ |
|
286 | 225 | list.__init__(self) |
|
287 | 226 | # the opened file handle |
|
288 |
|
|
|
227 | pofile = kwargs.get('pofile', None) | |
|
228 | if pofile and os.path.exists(pofile): | |
|
229 | self.fpath = pofile | |
|
230 | else: | |
|
231 | self.fpath = kwargs.get('fpath') | |
|
289 | 232 | # the width at which lines should be wrapped |
|
290 | 233 | self.wrapwidth = kwargs.get('wrapwidth', 78) |
|
291 | 234 | # the file encoding |
|
292 | 235 | self.encoding = kwargs.get('encoding', default_encoding) |
|
293 | 236 | # whether to check for duplicate entries or not |
|
294 | 237 | self.check_for_duplicates = kwargs.get('check_for_duplicates', False) |
|
295 | 238 | # header |
|
296 | 239 | self.header = '' |
|
297 | 240 | # both po and mo files have metadata |
|
298 | 241 | self.metadata = {} |
|
299 | 242 | self.metadata_is_fuzzy = 0 |
|
300 | 243 | |
|
301 |
def __ |
|
|
244 | def __unicode__(self): | |
|
302 | 245 | """ |
|
303 |
|
|
|
246 | Returns the unicode representation of the file. | |
|
304 | 247 | """ |
|
305 | 248 | ret = [] |
|
306 | 249 | entries = [self.metadata_as_entry()] + \ |
|
307 | 250 | [e for e in self if not e.obsolete] |
|
308 | 251 | for entry in entries: |
|
309 |
ret.append(entry.__ |
|
|
252 | ret.append(entry.__unicode__(self.wrapwidth)) | |
|
310 | 253 | for entry in self.obsolete_entries(): |
|
311 |
ret.append(entry.__ |
|
|
312 |
ret |
|
|
254 | ret.append(entry.__unicode__(self.wrapwidth)) | |
|
255 | ret = '\n'.join(ret) | |
|
256 | ||
|
257 | if type(ret) != types.UnicodeType: | |
|
258 | return unicode(ret, self.encoding) | |
|
259 | return ret | |
|
260 | ||
|
261 | def __str__(self): | |
|
262 | """ | |
|
263 | Returns the string representation of the file. | |
|
264 | """ | |
|
265 | return unicode(self).encode(self.encoding) | |
|
313 | 266 | |
|
314 | 267 | def __contains__(self, entry): |
|
315 | 268 | """ |
|
316 |
Overriden method to implement the membership test (in and |
|
|
317 | The method considers that an entry is in the file if it finds an | |
|
318 | entry that has the same msgid (case sensitive). | |
|
319 | ||
|
320 | **Keyword argument**: | |
|
321 | - *entry*: an instance of polib._BaseEntry | |
|
269 | Overriden ``list`` method to implement the membership test (in and | |
|
270 | not in). | |
|
271 | The method considers that an entry is in the file if it finds an entry | |
|
272 | that has the same msgid (the test is **case sensitive**). | |
|
322 | 273 | |
|
323 |
|
|
|
324 | >>> po = POFile() | |
|
325 | >>> e1 = POEntry(msgid='foobar', msgstr='spam') | |
|
326 | >>> e2 = POEntry(msgid='barfoo', msgstr='spam') | |
|
327 | >>> e3 = POEntry(msgid='foobar', msgstr='eggs') | |
|
328 | >>> e4 = POEntry(msgid='spameggs', msgstr='eggs') | |
|
329 | >>> po.append(e1) | |
|
330 | >>> po.append(e2) | |
|
331 | >>> e1 in po | |
|
332 | True | |
|
333 | >>> e2 not in po | |
|
334 | False | |
|
335 | >>> e3 in po | |
|
336 | True | |
|
337 | >>> e4 in po | |
|
338 | False | |
|
274 | Argument: | |
|
275 | ||
|
276 | ``entry`` | |
|
277 | an instance of :class:`~polib._BaseEntry`. | |
|
339 | 278 | """ |
|
340 | 279 | return self.find(entry.msgid, by='msgid') is not None |
|
280 | ||
|
281 | def __eq__(self, other): | |
|
282 | return unicode(self) == unicode(other) | |
|
341 | 283 | |
|
342 | 284 | def append(self, entry): |
|
343 | 285 | """ |
|
344 | 286 | Overriden method to check for duplicates entries, if a user tries to |
|
345 |
add an entry that already |
|
|
346 | exception. | |
|
347 | ||
|
348 | **Keyword argument**: | |
|
349 | - *entry*: an instance of polib._BaseEntry | |
|
287 | add an entry that is already in the file, the method will raise a | |
|
288 | ``ValueError`` exception. | |
|
350 | 289 | |
|
351 |
|
|
|
352 | >>> e1 = POEntry(msgid='foobar', msgstr='spam') | |
|
353 | >>> e2 = POEntry(msgid='foobar', msgstr='eggs') | |
|
354 | >>> po = POFile(check_for_duplicates=True) | |
|
355 | >>> po.append(e1) | |
|
356 | >>> try: | |
|
357 | ... po.append(e2) | |
|
358 | ... except ValueError, e: | |
|
359 | ... unicode(e) | |
|
360 | u'Entry "foobar" already exists' | |
|
290 | Argument: | |
|
291 | ||
|
292 | ``entry`` | |
|
293 | an instance of :class:`~polib._BaseEntry`. | |
|
361 | 294 | """ |
|
362 | 295 | if self.check_for_duplicates and entry in self: |
|
363 | 296 | raise ValueError('Entry "%s" already exists' % entry.msgid) |
|
364 | 297 | super(_BaseFile, self).append(entry) |
|
365 | 298 | |
|
366 | 299 | def insert(self, index, entry): |
|
367 | 300 | """ |
|
368 | 301 | Overriden method to check for duplicates entries, if a user tries to |
|
369 |
|
|
|
370 | exception. | |
|
302 | add an entry that is already in the file, the method will raise a | |
|
303 | ``ValueError`` exception. | |
|
371 | 304 | |
|
372 |
|
|
|
373 | - *index*: index at which the entry should be inserted | |
|
374 | - *entry*: an instance of polib._BaseEntry | |
|
305 | Arguments: | |
|
375 | 306 | |
|
376 | **Tests**: | |
|
377 | >>> import polib | |
|
378 | >>> polib.check_for_duplicates = True | |
|
379 | >>> e1 = POEntry(msgid='foobar', msgstr='spam') | |
|
380 | >>> e2 = POEntry(msgid='barfoo', msgstr='eggs') | |
|
381 | >>> e3 = POEntry(msgid='foobar', msgstr='eggs') | |
|
382 | >>> po = POFile(check_for_duplicates=True) | |
|
383 | >>> po.insert(0, e1) | |
|
384 | >>> po.insert(1, e2) | |
|
385 | >>> try: | |
|
386 | ... po.insert(0, e3) | |
|
387 | ... except ValueError, e: | |
|
388 | ... unicode(e) | |
|
389 | u'Entry "foobar" already exists' | |
|
307 | ``index`` | |
|
308 | index at which the entry should be inserted. | |
|
309 | ||
|
310 | ``entry`` | |
|
311 | an instance of :class:`~polib._BaseEntry`. | |
|
390 | 312 | """ |
|
391 | 313 | if self.check_for_duplicates and entry in self: |
|
392 | 314 | raise ValueError('Entry "%s" already exists' % entry.msgid) |
|
393 | 315 | super(_BaseFile, self).insert(index, entry) |
|
394 | 316 | |
|
395 | def __repr__(self): | |
|
396 | """Return the official string representation of the object.""" | |
|
397 | return '<%s instance at %x>' % (self.__class__.__name__, id(self)) | |
|
398 | ||
|
399 | 317 | def metadata_as_entry(self): |
|
400 | 318 | """ |
|
401 |
Return the metadata as a |
|
|
402 | ||
|
403 | >>> import polib | |
|
404 | >>> po = polib.pofile('tests/test_fuzzy_header.po') | |
|
405 | >>> unicode(po) == unicode(open('tests/test_fuzzy_header.po').read()) | |
|
406 | True | |
|
319 | Returns the file metadata as a :class:`~polib.POFile` instance. | |
|
407 | 320 | """ |
|
408 | 321 | e = POEntry(msgid='') |
|
409 | 322 | mdata = self.ordered_metadata() |
|
410 | 323 | if mdata: |
|
411 | 324 | strs = [] |
|
412 | e._multiline_str['msgstr'] = '' | |
|
413 | 325 | for name, value in mdata: |
|
414 | 326 | # Strip whitespace off each line in a multi-line entry |
|
415 | 327 | strs.append('%s: %s' % (name, value)) |
|
416 | 328 | e.msgstr = '\n'.join(strs) + '\n' |
|
417 | e._multiline_str['msgstr'] = '__POLIB__NL__'.join( | |
|
418 | [s + '\n' for s in strs]) | |
|
419 | 329 | if self.metadata_is_fuzzy: |
|
420 | 330 | e.flags.append('fuzzy') |
|
421 | 331 | return e |
|
422 | 332 | |
|
423 | 333 | def save(self, fpath=None, repr_method='__str__'): |
|
424 | 334 | """ |
|
425 |
Save the po file to |
|
|
426 | the object. If there's already an open file and no fpath is | |
|
427 |
|
|
|
428 | data. | |
|
335 | Saves the po file to ``fpath``. | |
|
336 | If it is an existing file and no ``fpath`` is provided, then the | |
|
337 | existing file is rewritten with the modified data. | |
|
338 | ||
|
339 | Keyword arguments: | |
|
429 | 340 | |
|
430 | **Keyword arguments**: | |
|
431 |
|
|
|
432 | - *repr_method*: string, the method to use for output. | |
|
341 | ``fpath`` | |
|
342 | string, full or relative path to the file. | |
|
343 | ||
|
344 | ``repr_method`` | |
|
345 | string, the method to use for output. | |
|
433 | 346 | """ |
|
434 | 347 | if self.fpath is None and fpath is None: |
|
435 | 348 | raise IOError('You must provide a file path to save() method') |
|
436 | 349 | contents = getattr(self, repr_method)() |
|
437 | 350 | if fpath is None: |
|
438 | 351 | fpath = self.fpath |
|
439 | 352 | if repr_method == 'to_binary': |
|
440 | 353 | fhandle = open(fpath, 'wb') |
|
441 | 354 | else: |
|
442 | 355 | fhandle = codecs.open(fpath, 'w', self.encoding) |
|
443 | 356 | if type(contents) != types.UnicodeType: |
|
444 | 357 | contents = contents.decode(self.encoding) |
|
445 | 358 | fhandle.write(contents) |
|
446 | 359 | fhandle.close() |
|
447 | ||
|
448 | def find(self, st, by='msgid'): | |
|
449 | """ | |
|
450 | Find entry which msgid (or property identified by the *by* | |
|
451 | attribute) matches the string *st*. | |
|
360 | # set the file path if not set | |
|
361 | if self.fpath is None and fpath: | |
|
362 | self.fpath = fpath | |
|
452 | 363 | |
|
453 | **Keyword arguments**: | |
|
454 | - *st*: string, the string to search for | |
|
455 | - *by*: string, the comparison attribute | |
|
364 | def find(self, st, by='msgid', include_obsolete_entries=False, | |
|
365 | msgctxt=False): | |
|
366 | """ | |
|
367 | Find the entry which msgid (or property identified by the ``by`` | |
|
368 | argument) matches the string ``st``. | |
|
456 | 369 | |
|
457 | **Examples**: | |
|
370 | Keyword arguments: | |
|
371 | ||
|
372 | ``st`` | |
|
373 | string, the string to search for. | |
|
458 | 374 | |
|
459 | >>> po = pofile('tests/test_utf8.po') | |
|
460 | >>> entry = po.find('Thursday') | |
|
461 | >>> entry.msgstr | |
|
462 | u'Jueves' | |
|
463 | >>> entry = po.find('Some unexistant msgid') | |
|
464 | >>> entry is None | |
|
465 | True | |
|
466 | >>> entry = po.find('Jueves', 'msgstr') | |
|
467 | >>> entry.msgid | |
|
468 | u'Thursday' | |
|
375 | ``by`` | |
|
376 | string, the property to use for comparison (default: ``msgid``). | |
|
377 | ||
|
378 | ``include_obsolete_entries`` | |
|
379 | boolean, whether to also search in entries that are obsolete. | |
|
380 | ||
|
381 | ``msgctxt`` | |
|
382 | string, allows to specify a specific message context for the | |
|
383 | search. | |
|
469 | 384 | """ |
|
470 | for e in self: | |
|
385 | if include_obsolete_entries: | |
|
386 | entries = self[:] | |
|
387 | else: | |
|
388 | entries = [e for e in self if not e.obsolete] | |
|
389 | for e in entries: | |
|
471 | 390 | if getattr(e, by) == st: |
|
391 | if msgctxt and e.msgctxt != msgctxt: | |
|
392 | continue | |
|
472 | 393 | return e |
|
473 | 394 | return None |
|
474 | 395 | |
|
475 | 396 | def ordered_metadata(self): |
|
476 | 397 | """ |
|
477 |
Convenience method that return |
|
|
478 |
value is list of tuples (metadata name, |
|
|
398 | Convenience method that returns an ordered version of the metadata | |
|
399 | dictionnary. The return value is list of tuples (metadata name, | |
|
400 | metadata_value). | |
|
479 | 401 | """ |
|
480 | 402 | # copy the dict first |
|
481 | 403 | metadata = self.metadata.copy() |
|
482 | 404 | data_order = [ |
|
483 | 405 | 'Project-Id-Version', |
|
484 | 406 | 'Report-Msgid-Bugs-To', |
|
485 | 407 | 'POT-Creation-Date', |
|
486 | 408 | 'PO-Revision-Date', |
|
487 | 409 | 'Last-Translator', |
|
488 | 410 | 'Language-Team', |
|
489 | 411 | 'MIME-Version', |
|
490 | 412 | 'Content-Type', |
|
491 | 413 | 'Content-Transfer-Encoding' |
|
492 | 414 | ] |
|
493 | 415 | ordered_data = [] |
|
494 | 416 | for data in data_order: |
|
495 | 417 | try: |
|
496 | 418 | value = metadata.pop(data) |
|
497 | 419 | ordered_data.append((data, value)) |
|
498 | 420 | except KeyError: |
|
499 | 421 | pass |
|
500 |
# the rest of the metadata w |
|
|
422 | # the rest of the metadata will be alphabetically ordered since there | |
|
423 | # are no specs for this AFAIK | |
|
501 | 424 | keys = metadata.keys() |
|
502 |
|
|
|
425 | keys.sort() | |
|
503 | 426 | for data in keys: |
|
504 | 427 | value = metadata[data] |
|
505 | 428 | ordered_data.append((data, value)) |
|
506 | 429 | return ordered_data |
|
507 | 430 | |
|
508 | 431 | def to_binary(self): |
|
509 | 432 | """ |
|
510 |
Return the |
|
|
433 | Return the binary representation of the file. | |
|
511 | 434 | """ |
|
512 | import array | |
|
513 | import struct | |
|
514 | import types | |
|
515 | 435 | offsets = [] |
|
516 | 436 | entries = self.translated_entries() |
|
517 | 437 | # the keys are sorted in the .mo file |
|
518 | 438 | def cmp(_self, other): |
|
519 | if _self.msgid > other.msgid: | |
|
439 | # msgfmt compares entries with msgctxt if it exists | |
|
440 | self_msgid = _self.msgctxt and _self.msgctxt or _self.msgid | |
|
441 | other_msgid = other.msgctxt and other.msgctxt or other.msgid | |
|
442 | if self_msgid > other_msgid: | |
|
520 | 443 | return 1 |
|
521 |
elif |
|
|
444 | elif self_msgid < other_msgid: | |
|
522 | 445 | return -1 |
|
523 | 446 | else: |
|
524 | 447 | return 0 |
|
525 | 448 | # add metadata entry |
|
526 | 449 | entries.sort(cmp) |
|
527 | 450 | mentry = self.metadata_as_entry() |
|
528 | mentry.msgstr = mentry.msgstr.replace('\\n', '').lstrip() | |
|
451 | #mentry.msgstr = mentry.msgstr.replace('\\n', '').lstrip() | |
|
529 | 452 | entries = [mentry] + entries |
|
530 | 453 | entries_len = len(entries) |
|
531 | 454 | ids, strs = '', '' |
|
532 | 455 | for e in entries: |
|
533 | 456 | # For each string, we need size and file offset. Each string is |
|
534 | 457 | # NUL terminated; the NUL does not count into the size. |
|
458 | msgid = '' | |
|
459 | if e.msgctxt: | |
|
460 | # Contexts are stored by storing the concatenation of the | |
|
461 | # context, a <EOT> byte, and the original string | |
|
462 | msgid = self._encode(e.msgctxt + '\4') | |
|
535 | 463 | if e.msgid_plural: |
|
536 | 464 | indexes = e.msgstr_plural.keys() |
|
537 | 465 | indexes.sort() |
|
538 | 466 | msgstr = [] |
|
539 | 467 | for index in indexes: |
|
540 | 468 | msgstr.append(e.msgstr_plural[index]) |
|
541 | msgid = self._encode(e.msgid + '\0' + e.msgid_plural) | |
|
469 | msgid += self._encode(e.msgid + '\0' + e.msgid_plural) | |
|
542 | 470 | msgstr = self._encode('\0'.join(msgstr)) |
|
543 | 471 | else: |
|
544 | msgid = self._encode(e.msgid) | |
|
472 | msgid += self._encode(e.msgid) | |
|
545 | 473 | msgstr = self._encode(e.msgstr) |
|
546 | 474 | offsets.append((len(ids), len(msgid), len(strs), len(msgstr))) |
|
547 | 475 | ids += msgid + '\0' |
|
548 | 476 | strs += msgstr + '\0' |
|
477 | ||
|
549 | 478 | # The header is 7 32-bit unsigned integers. |
|
550 | 479 | keystart = 7*4+16*entries_len |
|
551 | 480 | # and the values start after the keys |
|
552 | 481 | valuestart = keystart + len(ids) |
|
553 | 482 | koffsets = [] |
|
554 | 483 | voffsets = [] |
|
555 | 484 | # The string table first has the list of keys, then the list of values. |
|
556 | 485 | # Each entry has first the size of the string, then the file offset. |
|
557 | 486 | for o1, l1, o2, l2 in offsets: |
|
558 | 487 | koffsets += [l1, o1+keystart] |
|
559 | 488 | voffsets += [l2, o2+valuestart] |
|
560 | 489 | offsets = koffsets + voffsets |
|
561 | output = struct.pack("IIIIIII", | |
|
562 | 0x950412de, # Magic number | |
|
563 | 0, # Version | |
|
564 | entries_len, # # of entries | |
|
565 | 7*4, # start of key index | |
|
566 | 7*4+entries_len*8, # start of value index | |
|
567 | 0, 0) # size and offset of hash table | |
|
568 | output += array.array("I", offsets).tostring() | |
|
490 | # check endianness for magic number | |
|
491 | if struct.pack('@h', 1) == struct.pack('<h', 1): | |
|
492 | magic_number = MOFile.LITTLE_ENDIAN | |
|
493 | else: | |
|
494 | magic_number = MOFile.BIG_ENDIAN | |
|
495 | ||
|
496 | output = struct.pack( | |
|
497 | "Iiiiiii", | |
|
498 | magic_number, # Magic number | |
|
499 | 0, # Version | |
|
500 | entries_len, # # of entries | |
|
501 | 7*4, # start of key index | |
|
502 | 7*4+entries_len*8, # start of value index | |
|
503 | 0, keystart # size and offset of hash table | |
|
504 | # Important: we don't use hash tables | |
|
505 | ) | |
|
506 | output += array.array("i", offsets).tostring() | |
|
569 | 507 | output += ids |
|
570 | 508 | output += strs |
|
571 | 509 | return output |
|
572 | 510 | |
|
573 | 511 | def _encode(self, mixed): |
|
574 | 512 | """ |
|
575 |
Encode the given argument with the file encoding if |
|
|
576 | and return the encoded string. | |
|
513 | Encodes the given ``mixed`` argument with the file encoding if and | |
|
514 | only if it's an unicode string and returns the encoded string. | |
|
577 | 515 | """ |
|
578 | 516 | if type(mixed) == types.UnicodeType: |
|
579 | 517 | return mixed.encode(self.encoding) |
|
580 | 518 | return mixed |
|
581 | 519 | |
|
582 | 520 | # }}} |
|
583 | 521 | # class POFile {{{ |
|
584 | 522 | |
|
585 | 523 | class POFile(_BaseFile): |
|
586 | ''' | |
|
524 | """ | |
|
587 | 525 | Po (or Pot) file reader/writer. |
|
588 | POFile objects inherit the list objects methods. | |
|
589 | ||
|
590 | **Example**: | |
|
526 | This class inherits the :class:`~polib._BaseFile` class and, by extension, | |
|
527 | the python ``list`` type. | |
|
528 | """ | |
|
591 | 529 | |
|
592 | >>> po = POFile() | |
|
593 | >>> entry1 = POEntry( | |
|
594 | ... msgid="Some english text", | |
|
595 | ... msgstr="Un texte en anglais" | |
|
596 | ... ) | |
|
597 | >>> entry1.occurrences = [('testfile', 12),('another_file', 1)] | |
|
598 | >>> entry1.comment = "Some useful comment" | |
|
599 | >>> entry2 = POEntry( | |
|
600 | ... msgid="Peace in some languages", | |
|
601 | ... msgstr="Pace سلام שלום Hasîtî 和平" | |
|
602 | ... ) | |
|
603 | >>> entry2.occurrences = [('testfile', 15),('another_file', 5)] | |
|
604 | >>> entry2.comment = "Another useful comment" | |
|
605 | >>> entry3 = POEntry( | |
|
606 | ... msgid='Some entry with quotes " \\"', | |
|
607 | ... msgstr='Un message unicode avec des quotes " \\"' | |
|
608 | ... ) | |
|
609 | >>> entry3.comment = "Test string quoting" | |
|
610 | >>> po.append(entry1) | |
|
611 | >>> po.append(entry2) | |
|
612 | >>> po.append(entry3) | |
|
613 | >>> po.header = "Some Header" | |
|
614 | >>> print(po) | |
|
615 | # Some Header | |
|
616 | msgid "" | |
|
617 | msgstr "" | |
|
618 | <BLANKLINE> | |
|
619 | #. Some useful comment | |
|
620 | #: testfile:12 another_file:1 | |
|
621 | msgid "Some english text" | |
|
622 | msgstr "Un texte en anglais" | |
|
623 | <BLANKLINE> | |
|
624 | #. Another useful comment | |
|
625 | #: testfile:15 another_file:5 | |
|
626 | msgid "Peace in some languages" | |
|
627 | msgstr "Pace سلام שלום Hasîtî 和平" | |
|
628 | <BLANKLINE> | |
|
629 | #. Test string quoting | |
|
630 | msgid "Some entry with quotes \\" \\"" | |
|
631 | msgstr "Un message unicode avec des quotes \\" \\"" | |
|
632 | <BLANKLINE> | |
|
633 | ''' | |
|
634 | ||
|
635 | def __str__(self): | |
|
636 | """Return the string representation of the po file""" | |
|
530 | def __unicode__(self): | |
|
531 | """ | |
|
532 | Returns the unicode representation of the po file. | |
|
533 | """ | |
|
637 | 534 | ret, headers = '', self.header.split('\n') |
|
638 | 535 | for header in headers: |
|
639 | 536 | if header[:1] in [',', ':']: |
|
640 | 537 | ret += '#%s\n' % header |
|
641 | 538 | else: |
|
642 | 539 | ret += '# %s\n' % header |
|
643 | return ret + _BaseFile.__str__(self) | |
|
540 | ||
|
541 | if type(ret) != types.UnicodeType: | |
|
542 | ret = unicode(ret, self.encoding) | |
|
543 | ||
|
544 | return ret + _BaseFile.__unicode__(self) | |
|
644 | 545 | |
|
645 | 546 | def save_as_mofile(self, fpath): |
|
646 | 547 | """ |
|
647 |
Save the binary representation of the file to |
|
|
548 | Saves the binary representation of the file to given ``fpath``. | |
|
648 | 549 | |
|
649 |
|
|
|
650 | - *fpath*: string, full or relative path to the file. | |
|
550 | Keyword argument: | |
|
551 | ||
|
552 | ``fpath`` | |
|
553 | string, full or relative path to the mo file. | |
|
651 | 554 | """ |
|
652 | 555 | _BaseFile.save(self, fpath, 'to_binary') |
|
653 | 556 | |
|
654 | 557 | def percent_translated(self): |
|
655 | 558 | """ |
|
656 | Convenience method that return the percentage of translated | |
|
559 | Convenience method that returns the percentage of translated | |
|
657 | 560 | messages. |
|
658 | ||
|
659 | **Example**: | |
|
660 | ||
|
661 | >>> import polib | |
|
662 | >>> po = polib.pofile('tests/test_pofile_helpers.po') | |
|
663 | >>> po.percent_translated() | |
|
664 | 50 | |
|
665 | >>> po = POFile() | |
|
666 | >>> po.percent_translated() | |
|
667 | 100 | |
|
668 | 561 | """ |
|
669 | 562 | total = len([e for e in self if not e.obsolete]) |
|
670 | 563 | if total == 0: |
|
671 | 564 | return 100 |
|
672 | 565 | translated = len(self.translated_entries()) |
|
673 | 566 | return int((100.00 / float(total)) * translated) |
|
674 | 567 | |
|
675 | 568 | def translated_entries(self): |
|
676 | 569 | """ |
|
677 |
Convenience method that return |
|
|
678 | ||
|
679 | **Example**: | |
|
680 | ||
|
681 | >>> import polib | |
|
682 | >>> po = polib.pofile('tests/test_pofile_helpers.po') | |
|
683 | >>> len(po.translated_entries()) | |
|
684 | 6 | |
|
570 | Convenience method that returns the list of translated entries. | |
|
685 | 571 | """ |
|
686 | 572 | return [e for e in self if e.translated()] |
|
687 | 573 | |
|
688 | 574 | def untranslated_entries(self): |
|
689 | 575 | """ |
|
690 |
Convenience method that return |
|
|
691 | ||
|
692 | **Example**: | |
|
693 | ||
|
694 | >>> import polib | |
|
695 | >>> po = polib.pofile('tests/test_pofile_helpers.po') | |
|
696 | >>> len(po.untranslated_entries()) | |
|
697 | 4 | |
|
576 | Convenience method that returns the list of untranslated entries. | |
|
698 | 577 | """ |
|
699 | 578 | return [e for e in self if not e.translated() and not e.obsolete \ |
|
700 | 579 | and not 'fuzzy' in e.flags] |
|
701 | 580 | |
|
702 | 581 | def fuzzy_entries(self): |
|
703 | 582 | """ |
|
704 |
Convenience method that return the list of |
|
|
705 | ||
|
706 | **Example**: | |
|
707 | ||
|
708 | >>> import polib | |
|
709 | >>> po = polib.pofile('tests/test_pofile_helpers.po') | |
|
710 | >>> len(po.fuzzy_entries()) | |
|
711 | 2 | |
|
583 | Convenience method that returns the list of fuzzy entries. | |
|
712 | 584 | """ |
|
713 | 585 | return [e for e in self if 'fuzzy' in e.flags] |
|
714 | 586 | |
|
715 | 587 | def obsolete_entries(self): |
|
716 | 588 | """ |
|
717 | Convenience method that return the list of obsolete entries. | |
|
718 | ||
|
719 | **Example**: | |
|
720 | ||
|
721 | >>> import polib | |
|
722 | >>> po = polib.pofile('tests/test_pofile_helpers.po') | |
|
723 | >>> len(po.obsolete_entries()) | |
|
724 | 4 | |
|
589 | Convenience method that returns the list of obsolete entries. | |
|
725 | 590 | """ |
|
726 | 591 | return [e for e in self if e.obsolete] |
|
727 | 592 | |
|
728 | 593 | def merge(self, refpot): |
|
729 | 594 | """ |
|
730 | XXX this could not work if encodings are different, needs thinking | |
|
731 | and general refactoring of how polib handles encoding... | |
|
732 | ||
|
733 | Convenience method that merge the current pofile with the pot file | |
|
595 | Convenience method that merges the current pofile with the pot file | |
|
734 | 596 | provided. It behaves exactly as the gettext msgmerge utility: |
|
735 | 597 | |
|
736 |
|
|
|
737 |
|
|
|
738 |
|
|
|
739 |
|
|
|
740 | ||
|
741 | **Keyword argument**: | |
|
742 | - *refpot*: object POFile, the reference catalog. | |
|
598 | * comments of this file will be preserved, but extracted comments and | |
|
599 | occurrences will be discarded; | |
|
600 | * any translations or comments in the file will be discarded, however, | |
|
601 | dot comments and file positions will be preserved; | |
|
602 | * the fuzzy flags are preserved. | |
|
743 | 603 | |
|
744 | **Example**: | |
|
604 | Keyword argument: | |
|
745 | 605 | |
|
746 | >>> import polib | |
|
747 | >>> refpot = polib.pofile('tests/test_merge.pot') | |
|
748 | >>> po = polib.pofile('tests/test_merge_before.po') | |
|
749 | >>> po.merge(refpot) | |
|
750 | >>> expected_po = polib.pofile('tests/test_merge_after.po') | |
|
751 | >>> unicode(po) == unicode(expected_po) | |
|
752 | True | |
|
606 | ``refpot`` | |
|
607 | object POFile, the reference catalog. | |
|
753 | 608 | """ |
|
754 | 609 | for entry in refpot: |
|
755 | e = self.find(entry.msgid) | |
|
610 | e = self.find(entry.msgid, include_obsolete_entries=True) | |
|
756 | 611 | if e is None: |
|
757 | 612 | e = POEntry() |
|
758 | 613 | self.append(e) |
|
759 | 614 | e.merge(entry) |
|
760 | # ok, now we must "obsolete" entries that are not in the refpot | |
|
761 | # anymore | |
|
615 | # ok, now we must "obsolete" entries that are not in the refpot anymore | |
|
762 | 616 | for entry in self: |
|
763 | 617 | if refpot.find(entry.msgid) is None: |
|
764 | 618 | entry.obsolete = True |
|
765 | 619 | |
|
766 | 620 | # }}} |
|
767 | 621 | # class MOFile {{{ |
|
768 | 622 | |
|
769 | 623 | class MOFile(_BaseFile): |
|
770 | ''' | |
|
624 | """ | |
|
771 | 625 | Mo file reader/writer. |
|
772 | MOFile objects inherit the list objects methods. | |
|
773 | ||
|
774 | **Example**: | |
|
775 | ||
|
776 | >>> mo = MOFile() | |
|
777 | >>> entry1 = POEntry( | |
|
778 | ... msgid="Some english text", | |
|
779 | ... msgstr="Un texte en anglais" | |
|
780 | ... ) | |
|
781 | >>> entry2 = POEntry( | |
|
782 | ... msgid="I need my dirty cheese", | |
|
783 | ... msgstr="Je veux mon sale fromage" | |
|
784 | ... ) | |
|
785 | >>> entry3 = MOEntry( | |
|
786 | ... msgid='Some entry with quotes " \\"', | |
|
787 | ... msgstr='Un message unicode avec des quotes " \\"' | |
|
788 | ... ) | |
|
789 | >>> mo.append(entry1) | |
|
790 | >>> mo.append(entry2) | |
|
791 | >>> mo.append(entry3) | |
|
792 | >>> print(mo) | |
|
793 | msgid "" | |
|
794 | msgstr "" | |
|
795 | <BLANKLINE> | |
|
796 | msgid "Some english text" | |
|
797 | msgstr "Un texte en anglais" | |
|
798 | <BLANKLINE> | |
|
799 | msgid "I need my dirty cheese" | |
|
800 | msgstr "Je veux mon sale fromage" | |
|
801 | <BLANKLINE> | |
|
802 | msgid "Some entry with quotes \\" \\"" | |
|
803 | msgstr "Un message unicode avec des quotes \\" \\"" | |
|
804 | <BLANKLINE> | |
|
805 | ''' | |
|
626 | This class inherits the :class:`~polib._BaseFile` class and, by | |
|
627 | extension, the python ``list`` type. | |
|
628 | """ | |
|
629 | BIG_ENDIAN = 0xde120495 | |
|
630 | LITTLE_ENDIAN = 0x950412de | |
|
806 | 631 | |
|
807 | 632 | def __init__(self, *args, **kwargs): |
|
808 | 633 | """ |
|
809 | MOFile constructor. Mo files have two other properties: | |
|
810 | - magic_number: the magic_number of the binary file, | |
|
811 | - version: the version of the mo spec. | |
|
634 | Constructor, accepts all keywords arguments accepted by | |
|
635 | :class:`~polib._BaseFile` class. | |
|
812 | 636 | """ |
|
813 | 637 | _BaseFile.__init__(self, *args, **kwargs) |
|
814 | 638 | self.magic_number = None |
|
815 | 639 | self.version = 0 |
|
816 | 640 | |
|
817 | 641 | def save_as_pofile(self, fpath): |
|
818 | 642 | """ |
|
819 |
Save |
|
|
643 | Saves the mofile as a pofile to ``fpath``. | |
|
820 | 644 | |
|
821 |
|
|
|
822 | - *fpath*: string, full or relative path to the file. | |
|
645 | Keyword argument: | |
|
646 | ||
|
647 | ``fpath`` | |
|
648 | string, full or relative path to the file. | |
|
823 | 649 | """ |
|
824 | 650 | _BaseFile.save(self, fpath) |
|
825 | 651 | |
|
826 | def save(self, fpath): | |
|
652 | def save(self, fpath=None): | |
|
827 | 653 | """ |
|
828 |
Save |
|
|
654 | Saves the mofile to ``fpath``. | |
|
829 | 655 | |
|
830 |
|
|
|
831 | - *fpath*: string, full or relative path to the file. | |
|
656 | Keyword argument: | |
|
657 | ||
|
658 | ``fpath`` | |
|
659 | string, full or relative path to the file. | |
|
832 | 660 | """ |
|
833 | 661 | _BaseFile.save(self, fpath, 'to_binary') |
|
834 | 662 | |
|
835 | 663 | def percent_translated(self): |
|
836 | 664 | """ |
|
837 | 665 | Convenience method to keep the same interface with POFile instances. |
|
838 | 666 | """ |
|
839 | 667 | return 100 |
|
840 | 668 | |
|
841 | 669 | def translated_entries(self): |
|
842 | 670 | """ |
|
843 | 671 | Convenience method to keep the same interface with POFile instances. |
|
844 | 672 | """ |
|
845 | 673 | return self |
|
846 | 674 | |
|
847 | 675 | def untranslated_entries(self): |
|
848 | 676 | """ |
|
849 | 677 | Convenience method to keep the same interface with POFile instances. |
|
850 | 678 | """ |
|
851 | 679 | return [] |
|
852 | 680 | |
|
853 | 681 | def fuzzy_entries(self): |
|
854 | 682 | """ |
|
855 | 683 | Convenience method to keep the same interface with POFile instances. |
|
856 | 684 | """ |
|
857 | 685 | return [] |
|
858 | 686 | |
|
859 | 687 | def obsolete_entries(self): |
|
860 | 688 | """ |
|
861 | 689 | Convenience method to keep the same interface with POFile instances. |
|
862 | 690 | """ |
|
863 | 691 | return [] |
|
864 | 692 | |
|
865 | 693 | # }}} |
|
866 | 694 | # class _BaseEntry {{{ |
|
867 | 695 | |
|
868 | 696 | class _BaseEntry(object): |
|
869 | 697 | """ |
|
870 |
Base class for POEntry |
|
|
871 |
This class |
|
|
698 | Base class for :class:`~polib.POEntry` and :class:`~polib.MOEntry` classes. | |
|
699 | This class should **not** be instanciated directly. | |
|
872 | 700 | """ |
|
873 | 701 | |
|
874 | 702 | def __init__(self, *args, **kwargs): |
|
875 | """Base Entry constructor.""" | |
|
703 | """ | |
|
704 | Constructor, accepts the following keyword arguments: | |
|
705 | ||
|
706 | ``msgid`` | |
|
707 | string, the entry msgid. | |
|
708 | ||
|
709 | ``msgstr`` | |
|
710 | string, the entry msgstr. | |
|
711 | ||
|
712 | ``msgid_plural`` | |
|
713 | string, the entry msgid_plural. | |
|
714 | ||
|
715 | ``msgstr_plural`` | |
|
716 | list, the entry msgstr_plural lines. | |
|
717 | ||
|
718 | ``msgctxt`` | |
|
719 | string, the entry context (msgctxt). | |
|
720 | ||
|
721 | ``obsolete`` | |
|
722 | bool, whether the entry is "obsolete" or not. | |
|
723 | ||
|
724 | ``encoding`` | |
|
725 | string, the encoding to use, defaults to ``default_encoding`` | |
|
726 | global variable (optional). | |
|
727 | """ | |
|
876 | 728 | self.msgid = kwargs.get('msgid', '') |
|
877 | 729 | self.msgstr = kwargs.get('msgstr', '') |
|
878 | 730 | self.msgid_plural = kwargs.get('msgid_plural', '') |
|
879 | 731 | self.msgstr_plural = kwargs.get('msgstr_plural', {}) |
|
732 | self.msgctxt = kwargs.get('msgctxt', None) | |
|
880 | 733 | self.obsolete = kwargs.get('obsolete', False) |
|
881 | 734 | self.encoding = kwargs.get('encoding', default_encoding) |
|
882 | self.msgctxt = kwargs.get('msgctxt', None) | |
|
883 | self._multiline_str = {} | |
|
884 | 735 | |
|
885 |
def __ |
|
|
886 | """Return the official string representation of the object.""" | |
|
887 | return '<%s instance at %x>' % (self.__class__.__name__, id(self)) | |
|
888 | ||
|
889 | def __str__(self, wrapwidth=78): | |
|
736 | def __unicode__(self, wrapwidth=78): | |
|
890 | 737 | """ |
|
891 |
|
|
|
892 | objects. | |
|
738 | Returns the unicode representation of the entry. | |
|
893 | 739 | """ |
|
894 | 740 | if self.obsolete: |
|
895 | 741 | delflag = '#~ ' |
|
896 | 742 | else: |
|
897 | 743 | delflag = '' |
|
898 | 744 | ret = [] |
|
899 | 745 | # write the msgctxt if any |
|
900 | 746 | if self.msgctxt is not None: |
|
901 | ret += self._str_field("msgctxt", delflag, "", self.msgctxt) | |
|
747 | ret += self._str_field("msgctxt", delflag, "", self.msgctxt, wrapwidth) | |
|
902 | 748 | # write the msgid |
|
903 | ret += self._str_field("msgid", delflag, "", self.msgid) | |
|
749 | ret += self._str_field("msgid", delflag, "", self.msgid, wrapwidth) | |
|
904 | 750 | # write the msgid_plural if any |
|
905 | 751 | if self.msgid_plural: |
|
906 | ret += self._str_field("msgid_plural", delflag, "", self.msgid_plural) | |
|
752 | ret += self._str_field("msgid_plural", delflag, "", self.msgid_plural, wrapwidth) | |
|
907 | 753 | if self.msgstr_plural: |
|
908 | 754 | # write the msgstr_plural if any |
|
909 | 755 | msgstrs = self.msgstr_plural |
|
910 | 756 | keys = list(msgstrs) |
|
911 | 757 | keys.sort() |
|
912 | 758 | for index in keys: |
|
913 | 759 | msgstr = msgstrs[index] |
|
914 | 760 | plural_index = '[%s]' % index |
|
915 | ret += self._str_field("msgstr", delflag, plural_index, msgstr) | |
|
761 | ret += self._str_field("msgstr", delflag, plural_index, msgstr, wrapwidth) | |
|
916 | 762 | else: |
|
917 | 763 | # otherwise write the msgstr |
|
918 | ret += self._str_field("msgstr", delflag, "", self.msgstr) | |
|
764 | ret += self._str_field("msgstr", delflag, "", self.msgstr, wrapwidth) | |
|
919 | 765 | ret.append('') |
|
920 |
ret |
|
|
766 | ret = '\n'.join(ret) | |
|
767 | ||
|
768 | if type(ret) != types.UnicodeType: | |
|
769 | return unicode(ret, self.encoding) | |
|
770 | return ret | |
|
771 | ||
|
772 | def __str__(self): | |
|
773 | """ | |
|
774 | Returns the string representation of the entry. | |
|
775 | """ | |
|
776 | return unicode(self).encode(self.encoding) | |
|
777 | ||
|
778 | def __eq__(self, other): | |
|
779 | return unicode(self) == unicode(other) | |
|
921 | 780 | |
|
922 | def _str_field(self, fieldname, delflag, plural_index, field): | |
|
923 | if (fieldname + plural_index) in self._multiline_str: | |
|
924 | field = self._multiline_str[fieldname + plural_index] | |
|
925 | lines = [''] + field.split('__POLIB__NL__') | |
|
781 | def _str_field(self, fieldname, delflag, plural_index, field, wrapwidth=78): | |
|
782 | lines = field.splitlines(True) | |
|
783 | if len(lines) > 1: | |
|
784 | lines = [''] + lines # start with initial empty line | |
|
926 | 785 | else: |
|
927 | lines = field.splitlines(True) | |
|
928 | if len(lines) > 1: | |
|
929 | lines = ['']+lines # start with initial empty line | |
|
786 | escaped_field = escape(field) | |
|
787 | specialchars_count = 0 | |
|
788 | for c in ['\\', '\n', '\r', '\t', '"']: | |
|
789 | specialchars_count += field.count(c) | |
|
790 | # comparison must take into account fieldname length + one space | |
|
791 | # + 2 quotes (eg. msgid "<string>") | |
|
792 | flength = len(fieldname) + 3 | |
|
793 | if plural_index: | |
|
794 | flength += len(plural_index) | |
|
795 | real_wrapwidth = wrapwidth - flength + specialchars_count | |
|
796 | if wrapwidth > 0 and len(field) > real_wrapwidth: | |
|
797 | # Wrap the line but take field name into account | |
|
798 | lines = [''] + [unescape(item) for item in wrap( | |
|
799 | escaped_field, | |
|
800 | wrapwidth - 2, # 2 for quotes "" | |
|
801 | drop_whitespace=False, | |
|
802 | break_long_words=False | |
|
803 | )] | |
|
930 | 804 | else: |
|
931 |
lines = [field] |
|
|
805 | lines = [field] | |
|
932 | 806 | if fieldname.startswith('previous_'): |
|
933 | 807 | # quick and dirty trick to get the real field name |
|
934 | 808 | fieldname = fieldname[9:] |
|
935 | 809 | |
|
936 | 810 | ret = ['%s%s%s "%s"' % (delflag, fieldname, plural_index, |
|
937 | 811 | escape(lines.pop(0)))] |
|
938 | 812 | for mstr in lines: |
|
939 | 813 | ret.append('%s"%s"' % (delflag, escape(mstr))) |
|
940 | 814 | return ret |
|
941 | 815 | |
|
942 | 816 | # }}} |
|
943 | 817 | # class POEntry {{{ |
|
944 | 818 | |
|
945 | 819 | class POEntry(_BaseEntry): |
|
946 | 820 | """ |
|
947 | 821 | Represents a po file entry. |
|
948 | ||
|
949 | **Examples**: | |
|
950 | ||
|
951 | >>> entry = POEntry(msgid='Welcome', msgstr='Bienvenue') | |
|
952 | >>> entry.occurrences = [('welcome.py', 12), ('anotherfile.py', 34)] | |
|
953 | >>> print(entry) | |
|
954 | #: welcome.py:12 anotherfile.py:34 | |
|
955 | msgid "Welcome" | |
|
956 | msgstr "Bienvenue" | |
|
957 | <BLANKLINE> | |
|
958 | >>> entry = POEntry() | |
|
959 | >>> entry.occurrences = [('src/some-very-long-filename-that-should-not-be-wrapped-even-if-it-is-larger-than-the-wrap-limit.c', 32), ('src/eggs.c', 45)] | |
|
960 | >>> entry.comment = 'A plural translation. This is a very very very long line please do not wrap, this is just for testing comment wrapping...' | |
|
961 | >>> entry.tcomment = 'A plural translation. This is a very very very long line please do not wrap, this is just for testing comment wrapping...' | |
|
962 | >>> entry.flags.append('c-format') | |
|
963 | >>> entry.previous_msgctxt = '@somecontext' | |
|
964 | >>> entry.previous_msgid = 'I had eggs but no spam !' | |
|
965 | >>> entry.previous_msgid_plural = 'I had eggs and %d spam !' | |
|
966 | >>> entry.msgctxt = '@somenewcontext' | |
|
967 | >>> entry.msgid = 'I have spam but no egg !' | |
|
968 | >>> entry.msgid_plural = 'I have spam and %d eggs !' | |
|
969 | >>> entry.msgstr_plural[0] = "J'ai du jambon mais aucun oeuf !" | |
|
970 | >>> entry.msgstr_plural[1] = "J'ai du jambon et %d oeufs !" | |
|
971 | >>> print(entry) | |
|
972 | #. A plural translation. This is a very very very long line please do not | |
|
973 | #. wrap, this is just for testing comment wrapping... | |
|
974 | # A plural translation. This is a very very very long line please do not wrap, | |
|
975 | # this is just for testing comment wrapping... | |
|
976 | #: src/some-very-long-filename-that-should-not-be-wrapped-even-if-it-is-larger-than-the-wrap-limit.c:32 | |
|
977 | #: src/eggs.c:45 | |
|
978 | #, c-format | |
|
979 | #| msgctxt "@somecontext" | |
|
980 | #| msgid "I had eggs but no spam !" | |
|
981 | #| msgid_plural "I had eggs and %d spam !" | |
|
982 | msgctxt "@somenewcontext" | |
|
983 | msgid "I have spam but no egg !" | |
|
984 | msgid_plural "I have spam and %d eggs !" | |
|
985 | msgstr[0] "J'ai du jambon mais aucun oeuf !" | |
|
986 | msgstr[1] "J'ai du jambon et %d oeufs !" | |
|
987 | <BLANKLINE> | |
|
988 | 822 | """ |
|
989 | 823 | |
|
990 | 824 | def __init__(self, *args, **kwargs): |
|
991 | """POEntry constructor.""" | |
|
825 | """ | |
|
826 | Constructor, accepts the following keyword arguments: | |
|
827 | ||
|
828 | ``comment`` | |
|
829 | string, the entry comment. | |
|
830 | ||
|
831 | ``tcomment`` | |
|
832 | string, the entry translator comment. | |
|
833 | ||
|
834 | ``occurrences`` | |
|
835 | list, the entry occurrences. | |
|
836 | ||
|
837 | ``flags`` | |
|
838 | list, the entry flags. | |
|
839 | ||
|
840 | ``previous_msgctxt`` | |
|
841 | string, the entry previous context. | |
|
842 | ||
|
843 | ``previous_msgid`` | |
|
844 | string, the entry previous msgid. | |
|
845 | ||
|
846 | ``previous_msgid_plural`` | |
|
847 | string, the entry previous msgid_plural. | |
|
848 | """ | |
|
992 | 849 | _BaseEntry.__init__(self, *args, **kwargs) |
|
993 | 850 | self.comment = kwargs.get('comment', '') |
|
994 | 851 | self.tcomment = kwargs.get('tcomment', '') |
|
995 | 852 | self.occurrences = kwargs.get('occurrences', []) |
|
996 | 853 | self.flags = kwargs.get('flags', []) |
|
997 | 854 | self.previous_msgctxt = kwargs.get('previous_msgctxt', None) |
|
998 | 855 | self.previous_msgid = kwargs.get('previous_msgid', None) |
|
999 | 856 | self.previous_msgid_plural = kwargs.get('previous_msgid_plural', None) |
|
1000 | 857 | |
|
1001 |
def __ |
|
|
858 | def __unicode__(self, wrapwidth=78): | |
|
1002 | 859 | """ |
|
1003 |
Return |
|
|
860 | Returns the unicode representation of the entry. | |
|
1004 | 861 | """ |
|
1005 | 862 | if self.obsolete: |
|
1006 |
return _BaseEntry.__ |
|
|
863 | return _BaseEntry.__unicode__(self, wrapwidth) | |
|
864 | ||
|
1007 | 865 | ret = [] |
|
1008 | # comment first, if any (with text wrapping as xgettext does) | |
|
1009 | if self.comment != '': | |
|
1010 |
|
|
|
1011 | if wrapwidth > 0 and len(comment) > wrapwidth-3: | |
|
1012 | ret += textwrap.wrap(comment, wrapwidth, | |
|
1013 | initial_indent='#. ', | |
|
1014 | subsequent_indent='#. ', | |
|
1015 | break_long_words=False) | |
|
1016 | else: | |
|
1017 | ret.append('#. %s' % comment) | |
|
1018 | # translator comment, if any (with text wrapping as xgettext does) | |
|
1019 | if self.tcomment != '': | |
|
1020 | for tcomment in self.tcomment.split('\n'): | |
|
1021 | if wrapwidth > 0 and len(tcomment) > wrapwidth-2: | |
|
1022 | ret += textwrap.wrap(tcomment, wrapwidth, | |
|
1023 | initial_indent='# ', | |
|
1024 | subsequent_indent='# ', | |
|
1025 | break_long_words=False) | |
|
1026 | else: | |
|
1027 | ret.append('# %s' % tcomment) | |
|
866 | # comments first, if any (with text wrapping as xgettext does) | |
|
867 | comments = [('comment', '#. '), ('tcomment', '# ')] | |
|
868 | for c in comments: | |
|
869 | val = getattr(self, c[0]) | |
|
870 | if val: | |
|
871 | for comment in val.split('\n'): | |
|
872 | if wrapwidth > 0 and len(comment) + len(c[1]) > wrapwidth: | |
|
873 | ret += wrap( | |
|
874 | comment, | |
|
875 | wrapwidth, | |
|
876 | initial_indent=c[1], | |
|
877 | subsequent_indent=c[1], | |
|
878 | break_long_words=False | |
|
879 | ) | |
|
880 | else: | |
|
881 | ret.append('%s%s' % (c[1], comment)) | |
|
882 | ||
|
1028 | 883 | # occurrences (with text wrapping as xgettext does) |
|
1029 | 884 | if self.occurrences: |
|
1030 | 885 | filelist = [] |
|
1031 | 886 | for fpath, lineno in self.occurrences: |
|
1032 | 887 | if lineno: |
|
1033 | 888 | filelist.append('%s:%s' % (fpath, lineno)) |
|
1034 | 889 | else: |
|
1035 | 890 | filelist.append(fpath) |
|
1036 | 891 | filestr = ' '.join(filelist) |
|
1037 | if wrapwidth > 0 and len(filestr)+3 > wrapwidth: | |
|
1038 |
# |
|
|
892 | if wrapwidth > 0 and len(filestr) + 3 > wrapwidth: | |
|
893 | # textwrap split words that contain hyphen, this is not | |
|
1039 | 894 | # what we want for filenames, so the dirty hack is to |
|
1040 | 895 | # temporally replace hyphens with a char that a file cannot |
|
1041 | 896 | # contain, like "*" |
|
1042 |
|
|
|
1043 | wrapwidth, | |
|
1044 | initial_indent='#: ', | |
|
1045 |
|
|
|
1046 | break_long_words=False) | |
|
1047 |
|
|
|
1048 |
|
|
|
1049 | ret.append(line.replace('*', '-')) | |
|
897 | ret += [l.replace('*', '-') for l in wrap( | |
|
898 | filestr.replace('-', '*'), | |
|
899 | wrapwidth, | |
|
900 | initial_indent='#: ', | |
|
901 | subsequent_indent='#: ', | |
|
902 | break_long_words=False | |
|
903 | )] | |
|
1050 | 904 | else: |
|
1051 | ret.append('#: '+filestr) | |
|
1052 | # flags | |
|
905 | ret.append('#: ' + filestr) | |
|
906 | ||
|
907 | # flags (TODO: wrapping ?) | |
|
1053 | 908 | if self.flags: |
|
1054 | flags = [] | |
|
1055 | for flag in self.flags: | |
|
1056 | flags.append(flag) | |
|
1057 | ret.append('#, %s' % ', '.join(flags)) | |
|
909 | ret.append('#, %s' % ', '.join(self.flags)) | |
|
1058 | 910 | |
|
1059 | 911 | # previous context and previous msgid/msgid_plural |
|
1060 | if self.previous_msgctxt: | |
|
1061 | ret += self._str_field("previous_msgctxt", "#| ", "", | |
|
1062 | self.previous_msgctxt) | |
|
1063 | if self.previous_msgid: | |
|
1064 |
ret += self._str_field( |
|
|
1065 | self.previous_msgid) | |
|
1066 | if self.previous_msgid_plural: | |
|
1067 | ret += self._str_field("previous_msgid_plural", "#| ", "", | |
|
1068 | self.previous_msgid_plural) | |
|
912 | fields = ['previous_msgctxt', 'previous_msgid', 'previous_msgid_plural'] | |
|
913 | for f in fields: | |
|
914 | val = getattr(self, f) | |
|
915 | if val: | |
|
916 | ret += self._str_field(f, "#| ", "", val, wrapwidth) | |
|
1069 | 917 | |
|
1070 |
ret.append(_BaseEntry.__ |
|
|
1071 |
ret |
|
|
918 | ret.append(_BaseEntry.__unicode__(self, wrapwidth)) | |
|
919 | ret = '\n'.join(ret) | |
|
920 | ||
|
921 | if type(ret) != types.UnicodeType: | |
|
922 | return unicode(ret, self.encoding) | |
|
923 | return ret | |
|
1072 | 924 | |
|
1073 | 925 | def __cmp__(self, other): |
|
1074 |
|
|
|
926 | """ | |
|
1075 | 927 | Called by comparison operations if rich comparison is not defined. |
|
1076 | ||
|
1077 | **Tests**: | |
|
1078 | >>> a = POEntry(msgid='a', occurrences=[('b.py', 1), ('b.py', 3)]) | |
|
1079 | >>> b = POEntry(msgid='b', occurrences=[('b.py', 1), ('b.py', 3)]) | |
|
1080 | >>> c1 = POEntry(msgid='c1', occurrences=[('a.py', 1), ('b.py', 1)]) | |
|
1081 | >>> c2 = POEntry(msgid='c2', occurrences=[('a.py', 1), ('a.py', 3)]) | |
|
1082 | >>> po = POFile() | |
|
1083 | >>> po.append(a) | |
|
1084 | >>> po.append(b) | |
|
1085 | >>> po.append(c1) | |
|
1086 | >>> po.append(c2) | |
|
1087 | >>> po.sort() | |
|
1088 | >>> print(po) | |
|
1089 | # | |
|
1090 | msgid "" | |
|
1091 | msgstr "" | |
|
1092 | <BLANKLINE> | |
|
1093 | #: a.py:1 a.py:3 | |
|
1094 | msgid "c2" | |
|
1095 | msgstr "" | |
|
1096 | <BLANKLINE> | |
|
1097 | #: a.py:1 b.py:1 | |
|
1098 | msgid "c1" | |
|
1099 | msgstr "" | |
|
1100 | <BLANKLINE> | |
|
1101 | #: b.py:1 b.py:3 | |
|
1102 | msgid "a" | |
|
1103 | msgstr "" | |
|
1104 | <BLANKLINE> | |
|
1105 | #: b.py:1 b.py:3 | |
|
1106 | msgid "b" | |
|
1107 | msgstr "" | |
|
1108 | <BLANKLINE> | |
|
1109 | ''' | |
|
928 | """ | |
|
1110 | 929 | def compare_occurrences(a, b): |
|
1111 | 930 | """ |
|
1112 | 931 | Compare an entry occurrence with another one. |
|
1113 | 932 | """ |
|
1114 | 933 | if a[0] != b[0]: |
|
1115 | 934 | return a[0] < b[0] |
|
1116 | 935 | if a[1] != b[1]: |
|
1117 | 936 | return a[1] < b[1] |
|
1118 | 937 | return 0 |
|
1119 | 938 | |
|
1120 | 939 | # First: Obsolete test |
|
1121 | 940 | if self.obsolete != other.obsolete: |
|
1122 | 941 | if self.obsolete: |
|
1123 | 942 | return -1 |
|
1124 | 943 | else: |
|
1125 | 944 | return 1 |
|
1126 | 945 | # Work on a copy to protect original |
|
1127 | 946 | occ1 = self.occurrences[:] |
|
1128 | 947 | occ2 = other.occurrences[:] |
|
1129 | 948 | # Sorting using compare method |
|
1130 | 949 | occ1.sort(compare_occurrences) |
|
1131 | 950 | occ2.sort(compare_occurrences) |
|
1132 | 951 | # Comparing sorted occurrences |
|
1133 | 952 | pos = 0 |
|
1134 | 953 | for entry1 in occ1: |
|
1135 | 954 | try: |
|
1136 | 955 | entry2 = occ2[pos] |
|
1137 | 956 | except IndexError: |
|
1138 | 957 | return 1 |
|
1139 | 958 | pos = pos + 1 |
|
1140 | 959 | if entry1[0] != entry2[0]: |
|
1141 | 960 | if entry1[0] > entry2[0]: |
|
1142 | 961 | return 1 |
|
1143 | 962 | else: |
|
1144 | 963 | return -1 |
|
1145 | 964 | if entry1[1] != entry2[1]: |
|
1146 | 965 | if entry1[1] > entry2[1]: |
|
1147 | 966 | return 1 |
|
1148 | 967 | else: |
|
1149 | 968 | return -1 |
|
1150 | 969 | # Finally: Compare message ID |
|
1151 | 970 | if self.msgid > other.msgid: return 1 |
|
1152 | 971 | else: return -1 |
|
1153 | 972 | |
|
1154 | 973 | def translated(self): |
|
1155 | 974 | """ |
|
1156 |
Return True if the entry has been translated or |
|
|
975 | Returns ``True`` if the entry has been translated or ``False`` | |
|
976 | otherwise. | |
|
1157 | 977 | """ |
|
1158 | 978 | if self.obsolete or 'fuzzy' in self.flags: |
|
1159 | 979 | return False |
|
1160 | 980 | if self.msgstr != '': |
|
1161 | 981 | return True |
|
1162 | 982 | if self.msgstr_plural: |
|
1163 | 983 | for pos in self.msgstr_plural: |
|
1164 | 984 | if self.msgstr_plural[pos] == '': |
|
1165 | 985 | return False |
|
1166 | 986 | return True |
|
1167 | 987 | return False |
|
1168 | 988 | |
|
1169 | 989 | def merge(self, other): |
|
1170 | 990 | """ |
|
1171 | 991 | Merge the current entry with the given pot entry. |
|
1172 | 992 | """ |
|
1173 |
self.msgid |
|
|
1174 | self.occurrences = other.occurrences | |
|
1175 |
self. |
|
|
1176 | self.flags = other.flags | |
|
993 | self.msgid = other.msgid | |
|
994 | self.msgctxt = other.msgctxt | |
|
995 | self.occurrences = other.occurrences | |
|
996 | self.comment = other.comment | |
|
997 | fuzzy = 'fuzzy' in self.flags | |
|
998 | self.flags = other.flags[:] # clone flags | |
|
999 | if fuzzy: | |
|
1000 | self.flags.append('fuzzy') | |
|
1177 | 1001 | self.msgid_plural = other.msgid_plural |
|
1002 | self.obsolete = other.obsolete | |
|
1003 | self.previous_msgctxt = other.previous_msgctxt | |
|
1004 | self.previous_msgid = other.previous_msgid | |
|
1005 | self.previous_msgid_plural = other.previous_msgid_plural | |
|
1178 | 1006 | if other.msgstr_plural: |
|
1179 | 1007 | for pos in other.msgstr_plural: |
|
1180 | 1008 | try: |
|
1181 | 1009 | # keep existing translation at pos if any |
|
1182 | 1010 | self.msgstr_plural[pos] |
|
1183 | 1011 | except KeyError: |
|
1184 | 1012 | self.msgstr_plural[pos] = '' |
|
1185 | 1013 | |
|
1186 | 1014 | # }}} |
|
1187 | 1015 | # class MOEntry {{{ |
|
1188 | 1016 | |
|
1189 | 1017 | class MOEntry(_BaseEntry): |
|
1190 | 1018 | """ |
|
1191 | 1019 | Represents a mo file entry. |
|
1192 | ||
|
1193 | **Examples**: | |
|
1194 | ||
|
1195 | >>> entry = MOEntry() | |
|
1196 | >>> entry.msgid = 'translate me !' | |
|
1197 | >>> entry.msgstr = 'traduisez moi !' | |
|
1198 | >>> print(entry) | |
|
1199 | msgid "translate me !" | |
|
1200 | msgstr "traduisez moi !" | |
|
1201 | <BLANKLINE> | |
|
1202 | 1020 | """ |
|
1203 | ||
|
1204 | def __str__(self, wrapwidth=78): | |
|
1205 | """ | |
|
1206 | Return the string representation of the entry. | |
|
1207 | """ | |
|
1208 | return _BaseEntry.__str__(self, wrapwidth) | |
|
1021 | pass | |
|
1209 | 1022 | |
|
1210 | 1023 | # }}} |
|
1211 | 1024 | # class _POFileParser {{{ |
|
1212 | 1025 | |
|
1213 | 1026 | class _POFileParser(object): |
|
1214 | 1027 | """ |
|
1215 | 1028 | A finite state machine to parse efficiently and correctly po |
|
1216 | 1029 | file format. |
|
1217 | 1030 | """ |
|
1218 | 1031 | |
|
1219 |
def __init__(self, |
|
|
1032 | def __init__(self, pofile, *args, **kwargs): | |
|
1220 | 1033 | """ |
|
1221 | 1034 | Constructor. |
|
1222 | 1035 | |
|
1223 |
|
|
|
1224 | - *fpath*: string, path to the po file | |
|
1225 | - *encoding*: string, the encoding to use, defaults to | |
|
1226 | "default_encoding" global variable (optional), | |
|
1227 | - *check_for_duplicates*: whether to check for duplicate entries | |
|
1228 | when adding entries to the file, default: False (optional). | |
|
1036 | Keyword arguments: | |
|
1037 | ||
|
1038 | ``pofile`` | |
|
1039 | string, path to the po file or its content | |
|
1040 | ||
|
1041 | ``encoding`` | |
|
1042 | string, the encoding to use, defaults to ``default_encoding`` | |
|
1043 | global variable (optional). | |
|
1044 | ||
|
1045 | ``check_for_duplicates`` | |
|
1046 | whether to check for duplicate entries when adding entries to the | |
|
1047 | file (optional, default: ``False``). | |
|
1229 | 1048 | """ |
|
1230 | 1049 | enc = kwargs.get('encoding', default_encoding) |
|
1231 | check_dup = kwargs.get('check_for_duplicates', False) | |
|
1232 | try: | |
|
1233 |
self.fhandle = codecs.open( |
|
|
1234 | except LookupError: | |
|
1235 | enc = default_encoding | |
|
1236 |
self.fhandle = codecs.open( |
|
|
1050 | if os.path.exists(pofile): | |
|
1051 | try: | |
|
1052 | self.fhandle = codecs.open(pofile, 'rU', enc) | |
|
1053 | except LookupError: | |
|
1054 | enc = default_encoding | |
|
1055 | self.fhandle = codecs.open(pofile, 'rU', enc) | |
|
1056 | else: | |
|
1057 | self.fhandle = pofile.splitlines() | |
|
1058 | ||
|
1237 | 1059 | self.instance = POFile( |
|
1238 |
|
|
|
1060 | pofile=pofile, | |
|
1239 | 1061 | encoding=enc, |
|
1240 | check_for_duplicates=check_dup | |
|
1062 | check_for_duplicates=kwargs.get('check_for_duplicates', False) | |
|
1241 | 1063 | ) |
|
1242 | 1064 | self.transitions = {} |
|
1243 | 1065 | self.current_entry = POEntry() |
|
1244 | 1066 | self.current_state = 'ST' |
|
1245 | 1067 | self.current_token = None |
|
1246 | 1068 | # two memo flags used in handlers |
|
1247 | 1069 | self.msgstr_index = 0 |
|
1248 | 1070 | self.entry_obsolete = 0 |
|
1249 | 1071 | # Configure the state machine, by adding transitions. |
|
1250 | 1072 | # Signification of symbols: |
|
1251 | 1073 | # * ST: Beginning of the file (start) |
|
1252 | 1074 | # * HE: Header |
|
1253 | 1075 | # * TC: a translation comment |
|
1254 | 1076 | # * GC: a generated comment |
|
1255 | 1077 | # * OC: a file/line occurence |
|
1256 | 1078 | # * FL: a flags line |
|
1257 | 1079 | # * CT: a message context |
|
1258 | 1080 | # * PC: a previous msgctxt |
|
1259 | 1081 | # * PM: a previous msgid |
|
1260 | 1082 | # * PP: a previous msgid_plural |
|
1261 | 1083 | # * MI: a msgid |
|
1262 | 1084 | # * MP: a msgid plural |
|
1263 | 1085 | # * MS: a msgstr |
|
1264 | 1086 | # * MX: a msgstr plural |
|
1265 | 1087 | # * MC: a msgid or msgstr continuation line |
|
1266 | 1088 | all = ['ST', 'HE', 'GC', 'OC', 'FL', 'CT', 'PC', 'PM', 'PP', 'TC', |
|
1267 | 1089 | 'MS', 'MP', 'MX', 'MI'] |
|
1268 | 1090 | |
|
1269 | 1091 | self.add('TC', ['ST', 'HE'], 'HE') |
|
1270 | 1092 | self.add('TC', ['GC', 'OC', 'FL', 'TC', 'PC', 'PM', 'PP', 'MS', |
|
1271 | 1093 | 'MP', 'MX', 'MI'], 'TC') |
|
1272 | 1094 | self.add('GC', all, 'GC') |
|
1273 | 1095 | self.add('OC', all, 'OC') |
|
1274 | 1096 | self.add('FL', all, 'FL') |
|
1275 | 1097 | self.add('PC', all, 'PC') |
|
1276 | 1098 | self.add('PM', all, 'PM') |
|
1277 | 1099 | self.add('PP', all, 'PP') |
|
1278 | 1100 | self.add('CT', ['ST', 'HE', 'GC', 'OC', 'FL', 'TC', 'PC', 'PM', |
|
1279 | 1101 | 'PP', 'MS', 'MX'], 'CT') |
|
1280 | 1102 | self.add('MI', ['ST', 'HE', 'GC', 'OC', 'FL', 'CT', 'TC', 'PC', |
|
1281 | 1103 | 'PM', 'PP', 'MS', 'MX'], 'MI') |
|
1282 | 1104 | self.add('MP', ['TC', 'GC', 'PC', 'PM', 'PP', 'MI'], 'MP') |
|
1283 | 1105 | self.add('MS', ['MI', 'MP', 'TC'], 'MS') |
|
1284 | 1106 | self.add('MX', ['MI', 'MX', 'MP', 'TC'], 'MX') |
|
1285 | 1107 | self.add('MC', ['CT', 'MI', 'MP', 'MS', 'MX', 'PM', 'PP', 'PC'], 'MC') |
|
1286 | 1108 | |
|
1287 | 1109 | def parse(self): |
|
1288 | 1110 | """ |
|
1289 | 1111 | Run the state machine, parse the file line by line and call process() |
|
1290 | 1112 | with the current matched symbol. |
|
1291 | 1113 | """ |
|
1292 |
i |
|
|
1114 | i = 0 | |
|
1115 | ||
|
1116 | keywords = { | |
|
1117 | 'msgctxt': 'CT', | |
|
1118 | 'msgid': 'MI', | |
|
1119 | 'msgstr': 'MS', | |
|
1120 | 'msgid_plural': 'MP', | |
|
1121 | } | |
|
1122 | prev_keywords = { | |
|
1123 | 'msgid_plural': 'PP', | |
|
1124 | 'msgid': 'PM', | |
|
1125 | 'msgctxt': 'PC', | |
|
1126 | } | |
|
1127 | ||
|
1293 | 1128 | for line in self.fhandle: |
|
1129 | i += 1 | |
|
1294 | 1130 | line = line.strip() |
|
1295 | 1131 | if line == '': |
|
1296 | i = i+1 | |
|
1297 | 1132 | continue |
|
1298 | if line[:3] == '#~ ': | |
|
1299 | line = line[3:] | |
|
1133 | ||
|
1134 | tokens = line.split(None, 2) | |
|
1135 | nb_tokens = len(tokens) | |
|
1136 | ||
|
1137 | if tokens[0] == '#~' and nb_tokens > 1: | |
|
1138 | line = line[3:].strip() | |
|
1139 | tokens = tokens[1:] | |
|
1140 | nb_tokens -= 1 | |
|
1300 | 1141 | self.entry_obsolete = 1 |
|
1301 | 1142 | else: |
|
1302 | 1143 | self.entry_obsolete = 0 |
|
1144 | ||
|
1145 | # Take care of keywords like | |
|
1146 | # msgid, msgid_plural, msgctxt & msgstr. | |
|
1147 | if tokens[0] in keywords and nb_tokens > 1: | |
|
1148 | line = line[len(tokens[0]):].lstrip() | |
|
1149 | self.current_token = line | |
|
1150 | self.process(keywords[tokens[0]], i) | |
|
1151 | continue | |
|
1152 | ||
|
1303 | 1153 | self.current_token = line |
|
1304 | if line[:2] == '#:': | |
|
1154 | ||
|
1155 | if tokens[0] == '#:' and nb_tokens > 1: | |
|
1305 | 1156 | # we are on a occurrences line |
|
1306 | 1157 | self.process('OC', i) |
|
1307 | elif line[:9] == 'msgctxt "': | |
|
1308 | # we are on a msgctxt | |
|
1309 | self.process('CT', i) | |
|
1310 | elif line[:7] == 'msgid "': | |
|
1311 | # we are on a msgid | |
|
1312 | self.process('MI', i) | |
|
1313 | elif line[:8] == 'msgstr "': | |
|
1314 | # we are on a msgstr | |
|
1315 | self.process('MS', i) | |
|
1316 | elif line[:1] == '"' or line[:4] == '#| "': | |
|
1317 | # we are on a continuation line or some metadata | |
|
1158 | ||
|
1159 | elif line[:1] == '"': | |
|
1160 | # we are on a continuation line | |
|
1318 | 1161 | self.process('MC', i) |
|
1319 | elif line[:14] == 'msgid_plural "': | |
|
1320 | # we are on a msgid plural | |
|
1321 | self.process('MP', i) | |
|
1162 | ||
|
1322 | 1163 | elif line[:7] == 'msgstr[': |
|
1323 | 1164 | # we are on a msgstr plural |
|
1324 | 1165 | self.process('MX', i) |
|
1325 | elif line[:3] == '#, ': | |
|
1166 | ||
|
1167 | elif tokens[0] == '#,' and nb_tokens > 1: | |
|
1326 | 1168 | # we are on a flags line |
|
1327 | 1169 | self.process('FL', i) |
|
1328 | elif line[:2] == '# ' or line == '#': | |
|
1329 | if line == '#': line = line + ' ' | |
|
1170 | ||
|
1171 | elif tokens[0] == '#': | |
|
1172 | if line == '#': line += ' ' | |
|
1330 | 1173 | # we are on a translator comment line |
|
1331 | 1174 | self.process('TC', i) |
|
1332 | elif line[:2] == '#.': | |
|
1175 | ||
|
1176 | elif tokens[0] == '#.' and nb_tokens > 1: | |
|
1333 | 1177 | # we are on a generated comment line |
|
1334 | 1178 | self.process('GC', i) |
|
1335 | elif line[:15] == '#| msgid_plural': | |
|
1336 | # we are on a previous msgid_plural | |
|
1337 | self.process('PP', i) | |
|
1338 | elif line[:8] == '#| msgid': | |
|
1339 | self.process('PM', i) | |
|
1340 | # we are on a previous msgid | |
|
1341 | elif line[:10] == '#| msgctxt': | |
|
1342 | # we are on a previous msgctxt | |
|
1343 |
self. |
|
|
1344 | i = i+1 | |
|
1179 | ||
|
1180 | elif tokens[0] == '#|': | |
|
1181 | if nb_tokens < 2: | |
|
1182 | self.process('??', i) | |
|
1183 | continue | |
|
1184 | ||
|
1185 | # Remove the marker and any whitespace right after that. | |
|
1186 | line = line[2:].lstrip() | |
|
1187 | self.current_token = line | |
|
1188 | ||
|
1189 | if tokens[1].startswith('"'): | |
|
1190 | # Continuation of previous metadata. | |
|
1191 | self.process('MC', i) | |
|
1192 | continue | |
|
1193 | ||
|
1194 | if nb_tokens == 2: | |
|
1195 | # Invalid continuation line. | |
|
1196 | self.process('??', i) | |
|
1197 | ||
|
1198 | # we are on a "previous translation" comment line, | |
|
1199 | if tokens[1] not in prev_keywords: | |
|
1200 | # Unknown keyword in previous translation comment. | |
|
1201 | self.process('??', i) | |
|
1202 | ||
|
1203 | # Remove the keyword and any whitespace | |
|
1204 | # between it and the starting quote. | |
|
1205 | line = line[len(tokens[1]):].lstrip() | |
|
1206 | self.current_token = line | |
|
1207 | self.process(prev_keywords[tokens[1]], i) | |
|
1208 | ||
|
1209 | else: | |
|
1210 | self.process('??', i) | |
|
1345 | 1211 | |
|
1346 | 1212 | if self.current_entry: |
|
1347 | 1213 | # since entries are added when another entry is found, we must add |
|
1348 | 1214 | # the last entry here (only if there are lines) |
|
1349 | 1215 | self.instance.append(self.current_entry) |
|
1350 | 1216 | # before returning the instance, check if there's metadata and if |
|
1351 | 1217 | # so extract it in a dict |
|
1352 | 1218 | firstentry = self.instance[0] |
|
1353 | 1219 | if firstentry.msgid == '': # metadata found |
|
1354 | 1220 | # remove the entry |
|
1355 | 1221 | firstentry = self.instance.pop(0) |
|
1356 | 1222 | self.instance.metadata_is_fuzzy = firstentry.flags |
|
1357 | 1223 | key = None |
|
1358 | 1224 | for msg in firstentry.msgstr.splitlines(): |
|
1359 | 1225 | try: |
|
1360 | 1226 | key, val = msg.split(':', 1) |
|
1361 | 1227 | self.instance.metadata[key] = val.strip() |
|
1362 | 1228 | except: |
|
1363 | 1229 | if key is not None: |
|
1364 | 1230 | self.instance.metadata[key] += '\n'+ msg.strip() |
|
1365 | 1231 | # close opened file |
|
1366 |
self.fhandle |
|
|
1232 | if isinstance(self.fhandle, file): | |
|
1233 | self.fhandle.close() | |
|
1367 | 1234 | return self.instance |
|
1368 | 1235 | |
|
1369 | 1236 | def add(self, symbol, states, next_state): |
|
1370 | 1237 | """ |
|
1371 | 1238 | Add a transition to the state machine. |
|
1239 | ||
|
1372 | 1240 | Keywords arguments: |
|
1373 | 1241 | |
|
1374 | symbol -- string, the matched token (two chars symbol) | |
|
1375 | states -- list, a list of states (two chars symbols) | |
|
1376 | next_state -- the next state the fsm will have after the action | |
|
1242 | ``symbol`` | |
|
1243 | string, the matched token (two chars symbol). | |
|
1244 | ||
|
1245 | ``states`` | |
|
1246 | list, a list of states (two chars symbols). | |
|
1247 | ||
|
1248 | ``next_state`` | |
|
1249 | the next state the fsm will have after the action. | |
|
1377 | 1250 | """ |
|
1378 | 1251 | for state in states: |
|
1379 | 1252 | action = getattr(self, 'handle_%s' % next_state.lower()) |
|
1380 | 1253 | self.transitions[(symbol, state)] = (action, next_state) |
|
1381 | 1254 | |
|
1382 | 1255 | def process(self, symbol, linenum): |
|
1383 | 1256 | """ |
|
1384 | 1257 | Process the transition corresponding to the current state and the |
|
1385 | 1258 | symbol provided. |
|
1386 | 1259 | |
|
1387 | 1260 | Keywords arguments: |
|
1388 | symbol -- string, the matched token (two chars symbol) | |
|
1389 | linenum -- integer, the current line number of the parsed file | |
|
1261 | ||
|
1262 | ``symbol`` | |
|
1263 | string, the matched token (two chars symbol). | |
|
1264 | ||
|
1265 | ``linenum`` | |
|
1266 | integer, the current line number of the parsed file. | |
|
1390 | 1267 | """ |
|
1391 | 1268 | try: |
|
1392 | 1269 | (action, state) = self.transitions[(symbol, self.current_state)] |
|
1393 | 1270 | if action(): |
|
1394 | 1271 | self.current_state = state |
|
1395 | 1272 | except Exception, exc: |
|
1396 | 1273 | raise IOError('Syntax error in po file (line %s)' % linenum) |
|
1397 | 1274 | |
|
1398 | 1275 | # state handlers |
|
1399 | 1276 | |
|
1400 | 1277 | def handle_he(self): |
|
1401 | 1278 | """Handle a header comment.""" |
|
1402 | 1279 | if self.instance.header != '': |
|
1403 | 1280 | self.instance.header += '\n' |
|
1404 | 1281 | self.instance.header += self.current_token[2:] |
|
1405 | 1282 | return 1 |
|
1406 | 1283 | |
|
1407 | 1284 | def handle_tc(self): |
|
1408 | 1285 | """Handle a translator comment.""" |
|
1409 | 1286 | if self.current_state in ['MC', 'MS', 'MX']: |
|
1410 | 1287 | self.instance.append(self.current_entry) |
|
1411 | 1288 | self.current_entry = POEntry() |
|
1412 | 1289 | if self.current_entry.tcomment != '': |
|
1413 | 1290 | self.current_entry.tcomment += '\n' |
|
1414 | 1291 | self.current_entry.tcomment += self.current_token[2:] |
|
1415 | 1292 | return True |
|
1416 | 1293 | |
|
1417 | 1294 | def handle_gc(self): |
|
1418 | 1295 | """Handle a generated comment.""" |
|
1419 | 1296 | if self.current_state in ['MC', 'MS', 'MX']: |
|
1420 | 1297 | self.instance.append(self.current_entry) |
|
1421 | 1298 | self.current_entry = POEntry() |
|
1422 | 1299 | if self.current_entry.comment != '': |
|
1423 | 1300 | self.current_entry.comment += '\n' |
|
1424 | 1301 | self.current_entry.comment += self.current_token[3:] |
|
1425 | 1302 | return True |
|
1426 | 1303 | |
|
1427 | 1304 | def handle_oc(self): |
|
1428 | 1305 | """Handle a file:num occurence.""" |
|
1429 | 1306 | if self.current_state in ['MC', 'MS', 'MX']: |
|
1430 | 1307 | self.instance.append(self.current_entry) |
|
1431 | 1308 | self.current_entry = POEntry() |
|
1432 | 1309 | occurrences = self.current_token[3:].split() |
|
1433 | 1310 | for occurrence in occurrences: |
|
1434 | 1311 | if occurrence != '': |
|
1435 | 1312 | try: |
|
1436 | 1313 | fil, line = occurrence.split(':') |
|
1437 | 1314 | if not line.isdigit(): |
|
1438 | 1315 | fil = fil + line |
|
1439 | 1316 | line = '' |
|
1440 | 1317 | self.current_entry.occurrences.append((fil, line)) |
|
1441 | 1318 | except: |
|
1442 | 1319 | self.current_entry.occurrences.append((occurrence, '')) |
|
1443 | 1320 | return True |
|
1444 | 1321 | |
|
1445 | 1322 | def handle_fl(self): |
|
1446 | 1323 | """Handle a flags line.""" |
|
1447 | 1324 | if self.current_state in ['MC', 'MS', 'MX']: |
|
1448 | 1325 | self.instance.append(self.current_entry) |
|
1449 | 1326 | self.current_entry = POEntry() |
|
1450 | 1327 | self.current_entry.flags += self.current_token[3:].split(', ') |
|
1451 | 1328 | return True |
|
1452 | 1329 | |
|
1453 | 1330 | def handle_pp(self): |
|
1454 | 1331 | """Handle a previous msgid_plural line.""" |
|
1455 | 1332 | if self.current_state in ['MC', 'MS', 'MX']: |
|
1456 | 1333 | self.instance.append(self.current_entry) |
|
1457 | 1334 | self.current_entry = POEntry() |
|
1458 | 1335 | self.current_entry.previous_msgid_plural = \ |
|
1459 |
unescape(self.current_token[1 |
|
|
1336 | unescape(self.current_token[1:-1]) | |
|
1460 | 1337 | return True |
|
1461 | 1338 | |
|
1462 | 1339 | def handle_pm(self): |
|
1463 | 1340 | """Handle a previous msgid line.""" |
|
1464 | 1341 | if self.current_state in ['MC', 'MS', 'MX']: |
|
1465 | 1342 | self.instance.append(self.current_entry) |
|
1466 | 1343 | self.current_entry = POEntry() |
|
1467 | 1344 | self.current_entry.previous_msgid = \ |
|
1468 |
unescape(self.current_token[1 |
|
|
1345 | unescape(self.current_token[1:-1]) | |
|
1469 | 1346 | return True |
|
1470 | 1347 | |
|
1471 | 1348 | def handle_pc(self): |
|
1472 | 1349 | """Handle a previous msgctxt line.""" |
|
1473 | 1350 | if self.current_state in ['MC', 'MS', 'MX']: |
|
1474 | 1351 | self.instance.append(self.current_entry) |
|
1475 | 1352 | self.current_entry = POEntry() |
|
1476 | 1353 | self.current_entry.previous_msgctxt = \ |
|
1477 |
unescape(self.current_token[1 |
|
|
1354 | unescape(self.current_token[1:-1]) | |
|
1478 | 1355 | return True |
|
1479 | 1356 | |
|
1480 | 1357 | def handle_ct(self): |
|
1481 | 1358 | """Handle a msgctxt.""" |
|
1482 | 1359 | if self.current_state in ['MC', 'MS', 'MX']: |
|
1483 | 1360 | self.instance.append(self.current_entry) |
|
1484 | 1361 | self.current_entry = POEntry() |
|
1485 |
self.current_entry.msgctxt = unescape(self.current_token[ |
|
|
1362 | self.current_entry.msgctxt = unescape(self.current_token[1:-1]) | |
|
1486 | 1363 | return True |
|
1487 | 1364 | |
|
1488 | 1365 | def handle_mi(self): |
|
1489 | 1366 | """Handle a msgid.""" |
|
1490 | 1367 | if self.current_state in ['MC', 'MS', 'MX']: |
|
1491 | 1368 | self.instance.append(self.current_entry) |
|
1492 | 1369 | self.current_entry = POEntry() |
|
1493 | 1370 | self.current_entry.obsolete = self.entry_obsolete |
|
1494 |
self.current_entry.msgid = unescape(self.current_token[ |
|
|
1371 | self.current_entry.msgid = unescape(self.current_token[1:-1]) | |
|
1495 | 1372 | return True |
|
1496 | 1373 | |
|
1497 | 1374 | def handle_mp(self): |
|
1498 | 1375 | """Handle a msgid plural.""" |
|
1499 |
self.current_entry.msgid_plural = unescape(self.current_token[1 |
|
|
1376 | self.current_entry.msgid_plural = unescape(self.current_token[1:-1]) | |
|
1500 | 1377 | return True |
|
1501 | 1378 | |
|
1502 | 1379 | def handle_ms(self): |
|
1503 | 1380 | """Handle a msgstr.""" |
|
1504 |
self.current_entry.msgstr = unescape(self.current_token[ |
|
|
1381 | self.current_entry.msgstr = unescape(self.current_token[1:-1]) | |
|
1505 | 1382 | return True |
|
1506 | 1383 | |
|
1507 | 1384 | def handle_mx(self): |
|
1508 | 1385 | """Handle a msgstr plural.""" |
|
1509 | 1386 | index, value = self.current_token[7], self.current_token[11:-1] |
|
1510 | 1387 | self.current_entry.msgstr_plural[index] = unescape(value) |
|
1511 | 1388 | self.msgstr_index = index |
|
1512 | 1389 | return True |
|
1513 | 1390 | |
|
1514 | 1391 | def handle_mc(self): |
|
1515 | 1392 | """Handle a msgid or msgstr continuation line.""" |
|
1516 | 1393 | token = unescape(self.current_token[1:-1]) |
|
1517 | 1394 | if self.current_state == 'CT': |
|
1518 | 1395 | typ = 'msgctxt' |
|
1519 | 1396 | self.current_entry.msgctxt += token |
|
1520 | 1397 | elif self.current_state == 'MI': |
|
1521 | 1398 | typ = 'msgid' |
|
1522 | 1399 | self.current_entry.msgid += token |
|
1523 | 1400 | elif self.current_state == 'MP': |
|
1524 | 1401 | typ = 'msgid_plural' |
|
1525 | 1402 | self.current_entry.msgid_plural += token |
|
1526 | 1403 | elif self.current_state == 'MS': |
|
1527 | 1404 | typ = 'msgstr' |
|
1528 | 1405 | self.current_entry.msgstr += token |
|
1529 | 1406 | elif self.current_state == 'MX': |
|
1530 | 1407 | typ = 'msgstr[%s]' % self.msgstr_index |
|
1531 | 1408 | self.current_entry.msgstr_plural[self.msgstr_index] += token |
|
1532 | 1409 | elif self.current_state == 'PP': |
|
1533 | 1410 | typ = 'previous_msgid_plural' |
|
1534 | 1411 | token = token[3:] |
|
1535 | 1412 | self.current_entry.previous_msgid_plural += token |
|
1536 | 1413 | elif self.current_state == 'PM': |
|
1537 | 1414 | typ = 'previous_msgid' |
|
1538 | 1415 | token = token[3:] |
|
1539 | 1416 | self.current_entry.previous_msgid += token |
|
1540 | 1417 | elif self.current_state == 'PC': |
|
1541 | 1418 | typ = 'previous_msgctxt' |
|
1542 | 1419 | token = token[3:] |
|
1543 | 1420 | self.current_entry.previous_msgctxt += token |
|
1544 | if typ not in self.current_entry._multiline_str: | |
|
1545 | self.current_entry._multiline_str[typ] = token | |
|
1546 | else: | |
|
1547 | self.current_entry._multiline_str[typ] += "__POLIB__NL__" + token | |
|
1548 | 1421 | # don't change the current state |
|
1549 | 1422 | return False |
|
1550 | 1423 | |
|
1551 | 1424 | # }}} |
|
1552 | 1425 | # class _MOFileParser {{{ |
|
1553 | 1426 | |
|
1554 | 1427 | class _MOFileParser(object): |
|
1555 | 1428 | """ |
|
1556 | 1429 | A class to parse binary mo files. |
|
1557 | 1430 | """ |
|
1558 | BIG_ENDIAN = 0xde120495 | |
|
1559 | LITTLE_ENDIAN = 0x950412de | |
|
1560 | 1431 | |
|
1561 |
def __init__(self, |
|
|
1432 | def __init__(self, mofile, *args, **kwargs): | |
|
1562 | 1433 | """ |
|
1563 | 1434 | Constructor. |
|
1564 | 1435 | |
|
1565 |
|
|
|
1566 | - *fpath*: string, path to the po file | |
|
1567 | - *encoding*: string, the encoding to use, defaults to | |
|
1568 | "default_encoding" global variable (optional), | |
|
1569 | - *check_for_duplicates*: whether to check for duplicate entries | |
|
1570 | when adding entries to the file, default: False (optional). | |
|
1436 | Keyword arguments: | |
|
1437 | ||
|
1438 | ``mofile`` | |
|
1439 | string, path to the mo file or its content | |
|
1440 | ||
|
1441 | ``encoding`` | |
|
1442 | string, the encoding to use, defaults to ``default_encoding`` | |
|
1443 | global variable (optional). | |
|
1444 | ||
|
1445 | ``check_for_duplicates`` | |
|
1446 | whether to check for duplicate entries when adding entries to the | |
|
1447 | file (optional, default: ``False``). | |
|
1571 | 1448 | """ |
|
1572 | enc = kwargs.get('encoding', default_encoding) | |
|
1573 | check_dup = kwargs.get('check_for_duplicates', False) | |
|
1574 | self.fhandle = open(fpath, 'rb') | |
|
1449 | self.fhandle = open(mofile, 'rb') | |
|
1575 | 1450 | self.instance = MOFile( |
|
1576 |
fpath= |
|
|
1577 | encoding=enc, | |
|
1578 | check_for_duplicates=check_dup | |
|
1451 | fpath=mofile, | |
|
1452 | encoding=kwargs.get('encoding', default_encoding), | |
|
1453 | check_for_duplicates=kwargs.get('check_for_duplicates', False) | |
|
1579 | 1454 | ) |
|
1580 | 1455 | |
|
1581 | def parse_magicnumber(self): | |
|
1582 | """ | |
|
1583 | Parse the magic number and raise an exception if not valid. | |
|
1584 | """ | |
|
1585 | ||
|
1586 | 1456 | def parse(self): |
|
1587 | 1457 | """ |
|
1588 | 1458 | Build the instance with the file handle provided in the |
|
1589 | 1459 | constructor. |
|
1590 | 1460 | """ |
|
1461 | # parse magic number | |
|
1591 | 1462 | magic_number = self._readbinary('<I', 4) |
|
1592 |
if magic_number == |
|
|
1463 | if magic_number == MOFile.LITTLE_ENDIAN: | |
|
1593 | 1464 | ii = '<II' |
|
1594 |
elif magic_number == |
|
|
1465 | elif magic_number == MOFile.BIG_ENDIAN: | |
|
1595 | 1466 | ii = '>II' |
|
1596 | 1467 | else: |
|
1597 | 1468 | raise IOError('Invalid mo file, magic number is incorrect !') |
|
1598 | 1469 | self.instance.magic_number = magic_number |
|
1599 | 1470 | # parse the version number and the number of strings |
|
1600 | 1471 | self.instance.version, numofstrings = self._readbinary(ii, 8) |
|
1601 | 1472 | # original strings and translation strings hash table offset |
|
1602 | 1473 | msgids_hash_offset, msgstrs_hash_offset = self._readbinary(ii, 8) |
|
1603 | 1474 | # move to msgid hash table and read length and offset of msgids |
|
1604 | 1475 | self.fhandle.seek(msgids_hash_offset) |
|
1605 | 1476 | msgids_index = [] |
|
1606 | 1477 | for i in range(numofstrings): |
|
1607 | 1478 | msgids_index.append(self._readbinary(ii, 8)) |
|
1608 | 1479 | # move to msgstr hash table and read length and offset of msgstrs |
|
1609 | 1480 | self.fhandle.seek(msgstrs_hash_offset) |
|
1610 | 1481 | msgstrs_index = [] |
|
1611 | 1482 | for i in range(numofstrings): |
|
1612 | 1483 | msgstrs_index.append(self._readbinary(ii, 8)) |
|
1613 | 1484 | # build entries |
|
1614 | 1485 | for i in range(numofstrings): |
|
1615 | 1486 | self.fhandle.seek(msgids_index[i][1]) |
|
1616 | 1487 | msgid = self.fhandle.read(msgids_index[i][0]) |
|
1617 | 1488 | self.fhandle.seek(msgstrs_index[i][1]) |
|
1618 | 1489 | msgstr = self.fhandle.read(msgstrs_index[i][0]) |
|
1619 | 1490 | if i == 0: # metadata |
|
1620 | 1491 | raw_metadata, metadata = msgstr.split('\n'), {} |
|
1621 | 1492 | for line in raw_metadata: |
|
1622 | 1493 | tokens = line.split(':', 1) |
|
1623 | 1494 | if tokens[0] != '': |
|
1624 | 1495 | try: |
|
1625 | 1496 | metadata[tokens[0]] = tokens[1].strip() |
|
1626 | 1497 | except IndexError: |
|
1627 | 1498 | metadata[tokens[0]] = '' |
|
1628 | 1499 | self.instance.metadata = metadata |
|
1629 | 1500 | continue |
|
1630 | 1501 | # test if we have a plural entry |
|
1631 | 1502 | msgid_tokens = msgid.split('\0') |
|
1632 | 1503 | if len(msgid_tokens) > 1: |
|
1633 |
entry = |
|
|
1504 | entry = self._build_entry( | |
|
1634 | 1505 | msgid=msgid_tokens[0], |
|
1635 | 1506 | msgid_plural=msgid_tokens[1], |
|
1636 | msgstr_plural=dict((k,v) for k,v in \ | |
|
1637 | enumerate(msgstr.split('\0'))) | |
|
1507 | msgstr_plural=dict((k,v) for k,v in enumerate(msgstr.split('\0'))) | |
|
1638 | 1508 | ) |
|
1639 | 1509 | else: |
|
1640 |
entry = |
|
|
1510 | entry = self._build_entry(msgid=msgid, msgstr=msgstr) | |
|
1641 | 1511 | self.instance.append(entry) |
|
1642 | 1512 | # close opened file |
|
1643 | 1513 | self.fhandle.close() |
|
1644 | 1514 | return self.instance |
|
1515 | ||
|
1516 | def _build_entry(self, msgid, msgstr=None, msgid_plural=None, | |
|
1517 | msgstr_plural=None): | |
|
1518 | msgctxt_msgid = msgid.split('\x04') | |
|
1519 | if len(msgctxt_msgid) > 1: | |
|
1520 | kwargs = { | |
|
1521 | 'msgctxt': msgctxt_msgid[0], | |
|
1522 | 'msgid' : msgctxt_msgid[1], | |
|
1523 | } | |
|
1524 | else: | |
|
1525 | kwargs = {'msgid': msgid} | |
|
1526 | if msgstr: | |
|
1527 | kwargs['msgstr'] = msgstr | |
|
1528 | if msgid_plural: | |
|
1529 | kwargs['msgid_plural'] = msgid_plural | |
|
1530 | if msgstr_plural: | |
|
1531 | kwargs['msgstr_plural'] = msgstr_plural | |
|
1532 | return MOEntry(**kwargs) | |
|
1645 | 1533 | |
|
1646 | 1534 | def _readbinary(self, fmt, numbytes): |
|
1647 | 1535 | """ |
|
1648 | 1536 | Private method that unpack n bytes of data using format <fmt>. |
|
1649 | 1537 | It returns a tuple or a mixed value if the tuple length is 1. |
|
1650 | 1538 | """ |
|
1651 | 1539 | bytes = self.fhandle.read(numbytes) |
|
1652 | 1540 | tup = struct.unpack(fmt, bytes) |
|
1653 | 1541 | if len(tup) == 1: |
|
1654 | 1542 | return tup[0] |
|
1655 | 1543 | return tup |
|
1656 | 1544 | |
|
1657 | 1545 | # }}} |
|
1658 | # __main__ {{{ | |
|
1546 | # class TextWrapper {{{ | |
|
1659 | 1547 | |
|
1660 | if __name__ == '__main__': | |
|
1548 | class TextWrapper(textwrap.TextWrapper): | |
|
1661 | 1549 | """ |
|
1662 | **Main function**:: | |
|
1663 | - to **test** the module just run: *python polib.py [-v]* | |
|
1664 | - to **profile** the module: *python polib.py -p <some_pofile.po>* | |
|
1550 | Subclass of textwrap.TextWrapper that backport the | |
|
1551 | drop_whitespace option. | |
|
1665 | 1552 | """ |
|
1666 | import sys | |
|
1667 | if len(sys.argv) > 2 and sys.argv[1] == '-p': | |
|
1668 | def test(f): | |
|
1669 | if f.endswith('po'): | |
|
1670 | p = pofile(f) | |
|
1553 | def __init__(self, *args, **kwargs): | |
|
1554 | drop_whitespace = kwargs.pop('drop_whitespace', True) | |
|
1555 | textwrap.TextWrapper.__init__(self, *args, **kwargs) | |
|
1556 | self.drop_whitespace = drop_whitespace | |
|
1557 | ||
|
1558 | def _wrap_chunks(self, chunks): | |
|
1559 | """_wrap_chunks(chunks : [string]) -> [string] | |
|
1560 | ||
|
1561 | Wrap a sequence of text chunks and return a list of lines of | |
|
1562 | length 'self.width' or less. (If 'break_long_words' is false, | |
|
1563 | some lines may be longer than this.) Chunks correspond roughly | |
|
1564 | to words and the whitespace between them: each chunk is | |
|
1565 | indivisible (modulo 'break_long_words'), but a line break can | |
|
1566 | come between any two chunks. Chunks should not have internal | |
|
1567 | whitespace; ie. a chunk is either all whitespace or a "word". | |
|
1568 | Whitespace chunks will be removed from the beginning and end of | |
|
1569 | lines, but apart from that whitespace is preserved. | |
|
1570 | """ | |
|
1571 | lines = [] | |
|
1572 | if self.width <= 0: | |
|
1573 | raise ValueError("invalid width %r (must be > 0)" % self.width) | |
|
1574 | ||
|
1575 | # Arrange in reverse order so items can be efficiently popped | |
|
1576 | # from a stack of chucks. | |
|
1577 | chunks.reverse() | |
|
1578 | ||
|
1579 | while chunks: | |
|
1580 | ||
|
1581 | # Start the list of chunks that will make up the current line. | |
|
1582 | # cur_len is just the length of all the chunks in cur_line. | |
|
1583 | cur_line = [] | |
|
1584 | cur_len = 0 | |
|
1585 | ||
|
1586 | # Figure out which static string will prefix this line. | |
|
1587 | if lines: | |
|
1588 | indent = self.subsequent_indent | |
|
1671 | 1589 | else: |
|
1672 |
|
|
|
1673 | s = unicode(p) | |
|
1674 | import profile | |
|
1675 | profile.run('test("'+sys.argv[2]+'")') | |
|
1676 | else: | |
|
1677 | import doctest | |
|
1678 | doctest.testmod() | |
|
1590 | indent = self.initial_indent | |
|
1591 | ||
|
1592 | # Maximum width for this line. | |
|
1593 | width = self.width - len(indent) | |
|
1594 | ||
|
1595 | # First chunk on line is whitespace -- drop it, unless this | |
|
1596 | # is the very beginning of the text (ie. no lines started yet). | |
|
1597 | if self.drop_whitespace and chunks[-1].strip() == '' and lines: | |
|
1598 | del chunks[-1] | |
|
1599 | ||
|
1600 | while chunks: | |
|
1601 | l = len(chunks[-1]) | |
|
1602 | ||
|
1603 | # Can at least squeeze this chunk onto the current line. | |
|
1604 | if cur_len + l <= width: | |
|
1605 | cur_line.append(chunks.pop()) | |
|
1606 | cur_len += l | |
|
1607 | ||
|
1608 | # Nope, this line is full. | |
|
1609 | else: | |
|
1610 | break | |
|
1611 | ||
|
1612 | # The current line is full, and the next chunk is too big to | |
|
1613 | # fit on *any* line (not just this one). | |
|
1614 | if chunks and len(chunks[-1]) > width: | |
|
1615 | self._handle_long_word(chunks, cur_line, cur_len, width) | |
|
1616 | ||
|
1617 | # If the last chunk on this line is all whitespace, drop it. | |
|
1618 | if self.drop_whitespace and cur_line and cur_line[-1].strip() == '': | |
|
1619 | del cur_line[-1] | |
|
1620 | ||
|
1621 | # Convert current line back to a string and store it in list | |
|
1622 | # of all lines (return value). | |
|
1623 | if cur_line: | |
|
1624 | lines.append(indent + ''.join(cur_line)) | |
|
1625 | ||
|
1626 | return lines | |
|
1679 | 1627 | |
|
1680 | 1628 | # }}} |
|
1629 | # function wrap() {{{ | |
|
1630 | ||
|
1631 | def wrap(text, width=70, **kwargs): | |
|
1632 | """ | |
|
1633 | Wrap a single paragraph of text, returning a list of wrapped lines. | |
|
1634 | """ | |
|
1635 | if sys.version_info < (2, 6): | |
|
1636 | return TextWrapper(width=width, **kwargs).wrap(text) | |
|
1637 | return textwrap.wrap(text, width=width, **kwargs) | |
|
1638 | ||
|
1639 | #}}} |
General Comments 0
You need to be logged in to leave comments.
Login now