| 1 | # -*- coding: utf-8 -*- |
|---|
| 2 | # |
|---|
| 3 | # Copyright (C) 2007-2010 Edgewall Software |
|---|
| 4 | # All rights reserved. |
|---|
| 5 | # |
|---|
| 6 | # This software is licensed as described in the file COPYING, which |
|---|
| 7 | # you should have received as part of this distribution. The terms |
|---|
| 8 | # are also available at http://genshi.edgewall.org/wiki/License. |
|---|
| 9 | # |
|---|
| 10 | # This software consists of voluntary contributions made by many |
|---|
| 11 | # individuals. For the exact contribution history, see the revision |
|---|
| 12 | # history and logs, available at http://genshi.edgewall.org/log/. |
|---|
| 13 | |
|---|
| 14 | """Directives and utilities for internationalization and localization of |
|---|
| 15 | templates. |
|---|
| 16 | |
|---|
| 17 | :since: version 0.4 |
|---|
| 18 | :note: Directives support added since version 0.6 |
|---|
| 19 | """ |
|---|
| 20 | |
|---|
| 21 | try: |
|---|
| 22 | any |
|---|
| 23 | except NameError: |
|---|
| 24 | from genshi.util import any |
|---|
| 25 | from gettext import NullTranslations |
|---|
| 26 | import os |
|---|
| 27 | import re |
|---|
| 28 | from types import FunctionType |
|---|
| 29 | |
|---|
| 30 | from genshi.core import Attrs, Namespace, QName, START, END, TEXT, \ |
|---|
| 31 | XML_NAMESPACE, _ensure, StreamEventKind |
|---|
| 32 | from genshi.template.eval import _ast |
|---|
| 33 | from genshi.template.base import DirectiveFactory, EXPR, SUB, _apply_directives |
|---|
| 34 | from genshi.template.directives import Directive, StripDirective |
|---|
| 35 | from genshi.template.markup import MarkupTemplate, EXEC |
|---|
| 36 | |
|---|
| 37 | __all__ = ['Translator', 'extract'] |
|---|
| 38 | __docformat__ = 'restructuredtext en' |
|---|
| 39 | |
|---|
| 40 | |
|---|
| 41 | I18N_NAMESPACE = Namespace('http://genshi.edgewall.org/i18n') |
|---|
| 42 | |
|---|
| 43 | MSGBUF = StreamEventKind('MSGBUF') |
|---|
| 44 | SUB_START = StreamEventKind('SUB_START') |
|---|
| 45 | SUB_END = StreamEventKind('SUB_END') |
|---|
| 46 | |
|---|
| 47 | GETTEXT_FUNCTIONS = ('_', 'gettext', 'ngettext', 'dgettext', 'dngettext', |
|---|
| 48 | 'ugettext', 'ungettext') |
|---|
| 49 | |
|---|
| 50 | |
|---|
| 51 | class I18NDirective(Directive): |
|---|
| 52 | """Simple interface for i18n directives to support messages extraction.""" |
|---|
| 53 | |
|---|
| 54 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| 55 | return _apply_directives(stream, directives, ctxt, vars) |
|---|
| 56 | |
|---|
| 57 | |
|---|
| 58 | class ExtractableI18NDirective(I18NDirective): |
|---|
| 59 | """Simple interface for directives to support messages extraction.""" |
|---|
| 60 | |
|---|
| 61 | def extract(self, translator, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| 62 | search_text=True, comment_stack=None): |
|---|
| 63 | raise NotImplementedError |
|---|
| 64 | |
|---|
| 65 | |
|---|
| 66 | class CommentDirective(I18NDirective): |
|---|
| 67 | """Implementation of the ``i18n:comment`` template directive which adds |
|---|
| 68 | translation comments. |
|---|
| 69 | |
|---|
| 70 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 71 | ... <p i18n:comment="As in Foo Bar">Foo</p> |
|---|
| 72 | ... </html>''') |
|---|
| 73 | >>> translator = Translator() |
|---|
| 74 | >>> translator.setup(tmpl) |
|---|
| 75 | >>> list(translator.extract(tmpl.stream)) |
|---|
| 76 | [(2, None, u'Foo', [u'As in Foo Bar'])] |
|---|
| 77 | """ |
|---|
| 78 | __slots__ = ['comment'] |
|---|
| 79 | |
|---|
| 80 | def __init__(self, value, template=None, namespaces=None, lineno=-1, |
|---|
| 81 | offset=-1): |
|---|
| 82 | Directive.__init__(self, None, template, namespaces, lineno, offset) |
|---|
| 83 | self.comment = value |
|---|
| 84 | |
|---|
| 85 | |
|---|
| 86 | class MsgDirective(ExtractableI18NDirective): |
|---|
| 87 | r"""Implementation of the ``i18n:msg`` directive which marks inner content |
|---|
| 88 | as translatable. Consider the following examples: |
|---|
| 89 | |
|---|
| 90 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 91 | ... <div i18n:msg=""> |
|---|
| 92 | ... <p>Foo</p> |
|---|
| 93 | ... <p>Bar</p> |
|---|
| 94 | ... </div> |
|---|
| 95 | ... <p i18n:msg="">Foo <em>bar</em>!</p> |
|---|
| 96 | ... </html>''') |
|---|
| 97 | |
|---|
| 98 | >>> translator = Translator() |
|---|
| 99 | >>> translator.setup(tmpl) |
|---|
| 100 | >>> list(translator.extract(tmpl.stream)) |
|---|
| 101 | [(2, None, u'[1:Foo]\n [2:Bar]', []), (6, None, u'Foo [1:bar]!', [])] |
|---|
| 102 | >>> print(tmpl.generate().render()) |
|---|
| 103 | <html> |
|---|
| 104 | <div><p>Foo</p> |
|---|
| 105 | <p>Bar</p></div> |
|---|
| 106 | <p>Foo <em>bar</em>!</p> |
|---|
| 107 | </html> |
|---|
| 108 | |
|---|
| 109 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 110 | ... <div i18n:msg="fname, lname"> |
|---|
| 111 | ... <p>First Name: ${fname}</p> |
|---|
| 112 | ... <p>Last Name: ${lname}</p> |
|---|
| 113 | ... </div> |
|---|
| 114 | ... <p i18n:msg="">Foo <em>bar</em>!</p> |
|---|
| 115 | ... </html>''') |
|---|
| 116 | >>> translator.setup(tmpl) |
|---|
| 117 | >>> list(translator.extract(tmpl.stream)) #doctest: +NORMALIZE_WHITESPACE |
|---|
| 118 | [(2, None, u'[1:First Name: %(fname)s]\n [2:Last Name: %(lname)s]', []), |
|---|
| 119 | (6, None, u'Foo [1:bar]!', [])] |
|---|
| 120 | |
|---|
| 121 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 122 | ... <div i18n:msg="fname, lname"> |
|---|
| 123 | ... <p>First Name: ${fname}</p> |
|---|
| 124 | ... <p>Last Name: ${lname}</p> |
|---|
| 125 | ... </div> |
|---|
| 126 | ... <p i18n:msg="">Foo <em>bar</em>!</p> |
|---|
| 127 | ... </html>''') |
|---|
| 128 | >>> translator.setup(tmpl) |
|---|
| 129 | >>> print(tmpl.generate(fname='John', lname='Doe').render()) |
|---|
| 130 | <html> |
|---|
| 131 | <div><p>First Name: John</p> |
|---|
| 132 | <p>Last Name: Doe</p></div> |
|---|
| 133 | <p>Foo <em>bar</em>!</p> |
|---|
| 134 | </html> |
|---|
| 135 | |
|---|
| 136 | Starting and ending white-space is stripped of to make it simpler for |
|---|
| 137 | translators. Stripping it is not that important since it's on the html |
|---|
| 138 | source, the rendered output will remain the same. |
|---|
| 139 | """ |
|---|
| 140 | __slots__ = ['params', 'lineno'] |
|---|
| 141 | |
|---|
| 142 | def __init__(self, value, template=None, namespaces=None, lineno=-1, |
|---|
| 143 | offset=-1): |
|---|
| 144 | Directive.__init__(self, None, template, namespaces, lineno, offset) |
|---|
| 145 | self.params = [param.strip() for param in value.split(',') if param] |
|---|
| 146 | self.lineno = lineno |
|---|
| 147 | |
|---|
| 148 | @classmethod |
|---|
| 149 | def attach(cls, template, stream, value, namespaces, pos): |
|---|
| 150 | if type(value) is dict: |
|---|
| 151 | value = value.get('params', '').strip() |
|---|
| 152 | return super(MsgDirective, cls).attach(template, stream, value.strip(), |
|---|
| 153 | namespaces, pos) |
|---|
| 154 | |
|---|
| 155 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| 156 | gettext = ctxt.get('_i18n.gettext') |
|---|
| 157 | if ctxt.get('_i18n.domain'): |
|---|
| 158 | dgettext = ctxt.get('_i18n.dgettext') |
|---|
| 159 | assert hasattr(dgettext, '__call__'), \ |
|---|
| 160 | 'No domain gettext function passed' |
|---|
| 161 | gettext = lambda msg: dgettext(ctxt.get('_i18n.domain'), msg) |
|---|
| 162 | |
|---|
| 163 | def _generate(): |
|---|
| 164 | msgbuf = MessageBuffer(self) |
|---|
| 165 | previous = stream.next() |
|---|
| 166 | if previous[0] is START: |
|---|
| 167 | yield previous |
|---|
| 168 | else: |
|---|
| 169 | msgbuf.append(*previous) |
|---|
| 170 | previous = stream.next() |
|---|
| 171 | for kind, data, pos in stream: |
|---|
| 172 | msgbuf.append(*previous) |
|---|
| 173 | previous = kind, data, pos |
|---|
| 174 | if previous[0] is not END: |
|---|
| 175 | msgbuf.append(*previous) |
|---|
| 176 | previous = None |
|---|
| 177 | for event in msgbuf.translate(gettext(msgbuf.format())): |
|---|
| 178 | yield event |
|---|
| 179 | if previous: |
|---|
| 180 | yield previous |
|---|
| 181 | |
|---|
| 182 | return _apply_directives(_generate(), directives, ctxt, vars) |
|---|
| 183 | |
|---|
| 184 | def extract(self, translator, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| 185 | search_text=True, comment_stack=None): |
|---|
| 186 | msgbuf = MessageBuffer(self) |
|---|
| 187 | strip = False |
|---|
| 188 | |
|---|
| 189 | stream = iter(stream) |
|---|
| 190 | previous = stream.next() |
|---|
| 191 | if previous[0] is START: |
|---|
| 192 | for message in translator._extract_attrs(previous, |
|---|
| 193 | gettext_functions, |
|---|
| 194 | search_text=search_text): |
|---|
| 195 | yield message |
|---|
| 196 | previous = stream.next() |
|---|
| 197 | strip = True |
|---|
| 198 | for event in stream: |
|---|
| 199 | if event[0] is START: |
|---|
| 200 | for message in translator._extract_attrs(event, |
|---|
| 201 | gettext_functions, |
|---|
| 202 | search_text=search_text): |
|---|
| 203 | yield message |
|---|
| 204 | msgbuf.append(*previous) |
|---|
| 205 | previous = event |
|---|
| 206 | if not strip: |
|---|
| 207 | msgbuf.append(*previous) |
|---|
| 208 | |
|---|
| 209 | yield self.lineno, None, msgbuf.format(), comment_stack[-1:] |
|---|
| 210 | |
|---|
| 211 | |
|---|
| 212 | class ChooseBranchDirective(I18NDirective): |
|---|
| 213 | __slots__ = ['params'] |
|---|
| 214 | |
|---|
| 215 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| 216 | self.params = ctxt.get('_i18n.choose.params', [])[:] |
|---|
| 217 | msgbuf = MessageBuffer(self) |
|---|
| 218 | stream = _apply_directives(stream, directives, ctxt, vars) |
|---|
| 219 | |
|---|
| 220 | previous = stream.next() |
|---|
| 221 | if previous[0] is START: |
|---|
| 222 | yield previous |
|---|
| 223 | else: |
|---|
| 224 | msgbuf.append(*previous) |
|---|
| 225 | |
|---|
| 226 | try: |
|---|
| 227 | previous = stream.next() |
|---|
| 228 | except StopIteration: |
|---|
| 229 | # For example <i18n:singular> or <i18n:plural> directives |
|---|
| 230 | yield MSGBUF, (), -1 # the place holder for msgbuf output |
|---|
| 231 | ctxt['_i18n.choose.%s' % self.tagname] = msgbuf |
|---|
| 232 | return |
|---|
| 233 | |
|---|
| 234 | for event in stream: |
|---|
| 235 | msgbuf.append(*previous) |
|---|
| 236 | previous = event |
|---|
| 237 | yield MSGBUF, (), -1 # the place holder for msgbuf output |
|---|
| 238 | |
|---|
| 239 | if previous[0] is END: |
|---|
| 240 | yield previous # the outer end tag |
|---|
| 241 | else: |
|---|
| 242 | msgbuf.append(*previous) |
|---|
| 243 | ctxt['_i18n.choose.%s' % self.tagname] = msgbuf |
|---|
| 244 | |
|---|
| 245 | def extract(self, translator, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| 246 | search_text=True, comment_stack=None, msgbuf=None): |
|---|
| 247 | stream = iter(stream) |
|---|
| 248 | previous = stream.next() |
|---|
| 249 | |
|---|
| 250 | if previous[0] is START: |
|---|
| 251 | # skip the enclosing element |
|---|
| 252 | for message in translator._extract_attrs(previous, |
|---|
| 253 | gettext_functions, |
|---|
| 254 | search_text=search_text): |
|---|
| 255 | yield message |
|---|
| 256 | previous = stream.next() |
|---|
| 257 | |
|---|
| 258 | for event in stream: |
|---|
| 259 | if previous[0] is START: |
|---|
| 260 | for message in translator._extract_attrs(previous, |
|---|
| 261 | gettext_functions, |
|---|
| 262 | search_text=search_text): |
|---|
| 263 | yield message |
|---|
| 264 | msgbuf.append(*previous) |
|---|
| 265 | previous = event |
|---|
| 266 | |
|---|
| 267 | if previous[0] is not END: |
|---|
| 268 | msgbuf.append(*previous) |
|---|
| 269 | |
|---|
| 270 | |
|---|
| 271 | class SingularDirective(ChooseBranchDirective): |
|---|
| 272 | """Implementation of the ``i18n:singular`` directive to be used with the |
|---|
| 273 | ``i18n:choose`` directive.""" |
|---|
| 274 | |
|---|
| 275 | |
|---|
| 276 | class PluralDirective(ChooseBranchDirective): |
|---|
| 277 | """Implementation of the ``i18n:plural`` directive to be used with the |
|---|
| 278 | ``i18n:choose`` directive.""" |
|---|
| 279 | |
|---|
| 280 | |
|---|
| 281 | class ChooseDirective(ExtractableI18NDirective): |
|---|
| 282 | """Implementation of the ``i18n:choose`` directive which provides plural |
|---|
| 283 | internationalisation of strings. |
|---|
| 284 | |
|---|
| 285 | This directive requires at least one parameter, the one which evaluates to |
|---|
| 286 | an integer which will allow to choose the plural/singular form. If you also |
|---|
| 287 | have expressions inside the singular and plural version of the string you |
|---|
| 288 | also need to pass a name for those parameters. Consider the following |
|---|
| 289 | examples: |
|---|
| 290 | |
|---|
| 291 | >>> tmpl = MarkupTemplate('''\ |
|---|
| 292 | <html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 293 | ... <div i18n:choose="num; num"> |
|---|
| 294 | ... <p i18n:singular="">There is $num coin</p> |
|---|
| 295 | ... <p i18n:plural="">There are $num coins</p> |
|---|
| 296 | ... </div> |
|---|
| 297 | ... </html>''') |
|---|
| 298 | >>> translator = Translator() |
|---|
| 299 | >>> translator.setup(tmpl) |
|---|
| 300 | >>> list(translator.extract(tmpl.stream)) #doctest: +NORMALIZE_WHITESPACE |
|---|
| 301 | [(2, 'ngettext', (u'There is %(num)s coin', |
|---|
| 302 | u'There are %(num)s coins'), [])] |
|---|
| 303 | |
|---|
| 304 | >>> tmpl = MarkupTemplate('''\ |
|---|
| 305 | <html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 306 | ... <div i18n:choose="num; num"> |
|---|
| 307 | ... <p i18n:singular="">There is $num coin</p> |
|---|
| 308 | ... <p i18n:plural="">There are $num coins</p> |
|---|
| 309 | ... </div> |
|---|
| 310 | ... </html>''') |
|---|
| 311 | >>> translator.setup(tmpl) |
|---|
| 312 | >>> print(tmpl.generate(num=1).render()) |
|---|
| 313 | <html> |
|---|
| 314 | <div> |
|---|
| 315 | <p>There is 1 coin</p> |
|---|
| 316 | </div> |
|---|
| 317 | </html> |
|---|
| 318 | >>> print(tmpl.generate(num=2).render()) |
|---|
| 319 | <html> |
|---|
| 320 | <div> |
|---|
| 321 | <p>There are 2 coins</p> |
|---|
| 322 | </div> |
|---|
| 323 | </html> |
|---|
| 324 | |
|---|
| 325 | When used as a element and not as an attribute: |
|---|
| 326 | |
|---|
| 327 | >>> tmpl = MarkupTemplate('''\ |
|---|
| 328 | <html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 329 | ... <i18n:choose numeral="num" params="num"> |
|---|
| 330 | ... <p i18n:singular="">There is $num coin</p> |
|---|
| 331 | ... <p i18n:plural="">There are $num coins</p> |
|---|
| 332 | ... </i18n:choose> |
|---|
| 333 | ... </html>''') |
|---|
| 334 | >>> translator.setup(tmpl) |
|---|
| 335 | >>> list(translator.extract(tmpl.stream)) #doctest: +NORMALIZE_WHITESPACE |
|---|
| 336 | [(2, 'ngettext', (u'There is %(num)s coin', |
|---|
| 337 | u'There are %(num)s coins'), [])] |
|---|
| 338 | """ |
|---|
| 339 | __slots__ = ['numeral', 'params', 'lineno'] |
|---|
| 340 | |
|---|
| 341 | def __init__(self, value, template=None, namespaces=None, lineno=-1, |
|---|
| 342 | offset=-1): |
|---|
| 343 | Directive.__init__(self, None, template, namespaces, lineno, offset) |
|---|
| 344 | params = [v.strip() for v in value.split(';')] |
|---|
| 345 | self.numeral = self._parse_expr(params.pop(0), template, lineno, offset) |
|---|
| 346 | self.params = params and [name.strip() for name in |
|---|
| 347 | params[0].split(',') if name] or [] |
|---|
| 348 | self.lineno = lineno |
|---|
| 349 | |
|---|
| 350 | @classmethod |
|---|
| 351 | def attach(cls, template, stream, value, namespaces, pos): |
|---|
| 352 | if type(value) is dict: |
|---|
| 353 | numeral = value.get('numeral', '').strip() |
|---|
| 354 | assert numeral is not '', "at least pass the numeral param" |
|---|
| 355 | params = [v.strip() for v in value.get('params', '').split(',')] |
|---|
| 356 | value = '%s; ' % numeral + ', '.join(params) |
|---|
| 357 | return super(ChooseDirective, cls).attach(template, stream, value, |
|---|
| 358 | namespaces, pos) |
|---|
| 359 | |
|---|
| 360 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| 361 | ctxt.push({'_i18n.choose.params': self.params, |
|---|
| 362 | '_i18n.choose.singular': None, |
|---|
| 363 | '_i18n.choose.plural': None}) |
|---|
| 364 | |
|---|
| 365 | ngettext = ctxt.get('_i18n.ngettext') |
|---|
| 366 | assert hasattr(ngettext, '__call__'), 'No ngettext function available' |
|---|
| 367 | dngettext = ctxt.get('_i18n.dngettext') |
|---|
| 368 | if not dngettext: |
|---|
| 369 | dngettext = lambda d, s, p, n: ngettext(s, p, n) |
|---|
| 370 | |
|---|
| 371 | new_stream = [] |
|---|
| 372 | singular_stream = None |
|---|
| 373 | singular_msgbuf = None |
|---|
| 374 | plural_stream = None |
|---|
| 375 | plural_msgbuf = None |
|---|
| 376 | |
|---|
| 377 | numeral = self.numeral.evaluate(ctxt) |
|---|
| 378 | is_plural = self._is_plural(numeral, ngettext) |
|---|
| 379 | |
|---|
| 380 | for event in stream: |
|---|
| 381 | if event[0] is SUB and any(isinstance(d, ChooseBranchDirective) |
|---|
| 382 | for d in event[1][0]): |
|---|
| 383 | subdirectives, substream = event[1] |
|---|
| 384 | |
|---|
| 385 | if isinstance(subdirectives[0], SingularDirective): |
|---|
| 386 | singular_stream = list(_apply_directives(substream, |
|---|
| 387 | subdirectives, |
|---|
| 388 | ctxt, vars)) |
|---|
| 389 | new_stream.append((MSGBUF, None, (None, -1, -1))) |
|---|
| 390 | |
|---|
| 391 | elif isinstance(subdirectives[0], PluralDirective): |
|---|
| 392 | if is_plural: |
|---|
| 393 | plural_stream = list(_apply_directives(substream, |
|---|
| 394 | subdirectives, |
|---|
| 395 | ctxt, vars)) |
|---|
| 396 | |
|---|
| 397 | else: |
|---|
| 398 | new_stream.append(event) |
|---|
| 399 | |
|---|
| 400 | if ctxt.get('_i18n.domain'): |
|---|
| 401 | ngettext = lambda s, p, n: dngettext(ctxt.get('_i18n.domain'), |
|---|
| 402 | s, p, n) |
|---|
| 403 | |
|---|
| 404 | singular_msgbuf = ctxt.get('_i18n.choose.singular') |
|---|
| 405 | if is_plural: |
|---|
| 406 | plural_msgbuf = ctxt.get('_i18n.choose.plural') |
|---|
| 407 | msgbuf, choice = plural_msgbuf, plural_stream |
|---|
| 408 | else: |
|---|
| 409 | msgbuf, choice = singular_msgbuf, singular_stream |
|---|
| 410 | plural_msgbuf = MessageBuffer(self) |
|---|
| 411 | |
|---|
| 412 | for kind, data, pos in new_stream: |
|---|
| 413 | if kind is MSGBUF: |
|---|
| 414 | for event in choice: |
|---|
| 415 | if event[0] is MSGBUF: |
|---|
| 416 | translation = ngettext(singular_msgbuf.format(), |
|---|
| 417 | plural_msgbuf.format(), |
|---|
| 418 | numeral) |
|---|
| 419 | for subevent in msgbuf.translate(translation): |
|---|
| 420 | yield subevent |
|---|
| 421 | else: |
|---|
| 422 | yield event |
|---|
| 423 | else: |
|---|
| 424 | yield kind, data, pos |
|---|
| 425 | |
|---|
| 426 | ctxt.pop() |
|---|
| 427 | |
|---|
| 428 | def extract(self, translator, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| 429 | search_text=True, comment_stack=None): |
|---|
| 430 | strip = False |
|---|
| 431 | stream = iter(stream) |
|---|
| 432 | previous = stream.next() |
|---|
| 433 | |
|---|
| 434 | if previous[0] is START: |
|---|
| 435 | # skip the enclosing element |
|---|
| 436 | for message in translator._extract_attrs(previous, |
|---|
| 437 | gettext_functions, |
|---|
| 438 | search_text=search_text): |
|---|
| 439 | yield message |
|---|
| 440 | previous = stream.next() |
|---|
| 441 | strip = True |
|---|
| 442 | |
|---|
| 443 | singular_msgbuf = MessageBuffer(self) |
|---|
| 444 | plural_msgbuf = MessageBuffer(self) |
|---|
| 445 | |
|---|
| 446 | for event in stream: |
|---|
| 447 | if previous[0] is SUB: |
|---|
| 448 | directives, substream = previous[1] |
|---|
| 449 | for directive in directives: |
|---|
| 450 | if isinstance(directive, SingularDirective): |
|---|
| 451 | for message in directive.extract(translator, |
|---|
| 452 | substream, gettext_functions, search_text, |
|---|
| 453 | comment_stack, msgbuf=singular_msgbuf): |
|---|
| 454 | yield message |
|---|
| 455 | elif isinstance(directive, PluralDirective): |
|---|
| 456 | for message in directive.extract(translator, |
|---|
| 457 | substream, gettext_functions, search_text, |
|---|
| 458 | comment_stack, msgbuf=plural_msgbuf): |
|---|
| 459 | yield message |
|---|
| 460 | elif not isinstance(directive, StripDirective): |
|---|
| 461 | singular_msgbuf.append(*previous) |
|---|
| 462 | plural_msgbuf.append(*previous) |
|---|
| 463 | else: |
|---|
| 464 | if previous[0] is START: |
|---|
| 465 | for message in translator._extract_attrs(previous, |
|---|
| 466 | gettext_functions, |
|---|
| 467 | search_text): |
|---|
| 468 | yield message |
|---|
| 469 | singular_msgbuf.append(*previous) |
|---|
| 470 | plural_msgbuf.append(*previous) |
|---|
| 471 | previous = event |
|---|
| 472 | |
|---|
| 473 | if not strip: |
|---|
| 474 | singular_msgbuf.append(*previous) |
|---|
| 475 | plural_msgbuf.append(*previous) |
|---|
| 476 | |
|---|
| 477 | yield self.lineno, 'ngettext', \ |
|---|
| 478 | (singular_msgbuf.format(), plural_msgbuf.format()), \ |
|---|
| 479 | comment_stack[-1:] |
|---|
| 480 | |
|---|
| 481 | def _is_plural(self, numeral, ngettext): |
|---|
| 482 | # XXX: should we test which form was chosen like this!?!?!? |
|---|
| 483 | # There should be no match in any catalogue for these singular and |
|---|
| 484 | # plural test strings |
|---|
| 485 | singular = u'O\x85\xbe\xa9\xa8az\xc3?\xe6\xa1\x02n\x84\x93' |
|---|
| 486 | plural = u'\xcc\xfb+\xd3Pn\x9d\tT\xec\x1d\xda\x1a\x88\x00' |
|---|
| 487 | return ngettext(singular, plural, numeral) == plural |
|---|
| 488 | |
|---|
| 489 | |
|---|
| 490 | class DomainDirective(I18NDirective): |
|---|
| 491 | """Implementation of the ``i18n:domain`` directive which allows choosing |
|---|
| 492 | another i18n domain(catalog) to translate from. |
|---|
| 493 | |
|---|
| 494 | >>> from genshi.filters.tests.i18n import DummyTranslations |
|---|
| 495 | >>> tmpl = MarkupTemplate('''\ |
|---|
| 496 | <html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 497 | ... <p i18n:msg="">Bar</p> |
|---|
| 498 | ... <div i18n:domain="foo"> |
|---|
| 499 | ... <p i18n:msg="">FooBar</p> |
|---|
| 500 | ... <p>Bar</p> |
|---|
| 501 | ... <p i18n:domain="bar" i18n:msg="">Bar</p> |
|---|
| 502 | ... <p i18n:domain="">Bar</p> |
|---|
| 503 | ... </div> |
|---|
| 504 | ... <p>Bar</p> |
|---|
| 505 | ... </html>''') |
|---|
| 506 | |
|---|
| 507 | >>> translations = DummyTranslations({'Bar': 'Voh'}) |
|---|
| 508 | >>> translations.add_domain('foo', {'FooBar': 'BarFoo', 'Bar': 'foo_Bar'}) |
|---|
| 509 | >>> translations.add_domain('bar', {'Bar': 'bar_Bar'}) |
|---|
| 510 | >>> translator = Translator(translations) |
|---|
| 511 | >>> translator.setup(tmpl) |
|---|
| 512 | |
|---|
| 513 | >>> print(tmpl.generate().render()) |
|---|
| 514 | <html> |
|---|
| 515 | <p>Voh</p> |
|---|
| 516 | <div> |
|---|
| 517 | <p>BarFoo</p> |
|---|
| 518 | <p>foo_Bar</p> |
|---|
| 519 | <p>bar_Bar</p> |
|---|
| 520 | <p>Voh</p> |
|---|
| 521 | </div> |
|---|
| 522 | <p>Voh</p> |
|---|
| 523 | </html> |
|---|
| 524 | """ |
|---|
| 525 | __slots__ = ['domain'] |
|---|
| 526 | |
|---|
| 527 | def __init__(self, value, template=None, namespaces=None, lineno=-1, |
|---|
| 528 | offset=-1): |
|---|
| 529 | Directive.__init__(self, None, template, namespaces, lineno, offset) |
|---|
| 530 | self.domain = value and value.strip() or '__DEFAULT__' |
|---|
| 531 | |
|---|
| 532 | @classmethod |
|---|
| 533 | def attach(cls, template, stream, value, namespaces, pos): |
|---|
| 534 | if type(value) is dict: |
|---|
| 535 | value = value.get('name') |
|---|
| 536 | return super(DomainDirective, cls).attach(template, stream, value, |
|---|
| 537 | namespaces, pos) |
|---|
| 538 | |
|---|
| 539 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| 540 | ctxt.push({'_i18n.domain': self.domain}) |
|---|
| 541 | for event in _apply_directives(stream, directives, ctxt, vars): |
|---|
| 542 | yield event |
|---|
| 543 | ctxt.pop() |
|---|
| 544 | |
|---|
| 545 | |
|---|
| 546 | class Translator(DirectiveFactory): |
|---|
| 547 | """Can extract and translate localizable strings from markup streams and |
|---|
| 548 | templates. |
|---|
| 549 | |
|---|
| 550 | For example, assume the following template: |
|---|
| 551 | |
|---|
| 552 | >>> tmpl = MarkupTemplate('''<html xmlns:py="http://genshi.edgewall.org/"> |
|---|
| 553 | ... <head> |
|---|
| 554 | ... <title>Example</title> |
|---|
| 555 | ... </head> |
|---|
| 556 | ... <body> |
|---|
| 557 | ... <h1>Example</h1> |
|---|
| 558 | ... <p>${_("Hello, %(name)s") % dict(name=username)}</p> |
|---|
| 559 | ... </body> |
|---|
| 560 | ... </html>''', filename='example.html') |
|---|
| 561 | |
|---|
| 562 | For demonstration, we define a dummy ``gettext``-style function with a |
|---|
| 563 | hard-coded translation table, and pass that to the `Translator` initializer: |
|---|
| 564 | |
|---|
| 565 | >>> def pseudo_gettext(string): |
|---|
| 566 | ... return { |
|---|
| 567 | ... 'Example': 'Beispiel', |
|---|
| 568 | ... 'Hello, %(name)s': 'Hallo, %(name)s' |
|---|
| 569 | ... }[string] |
|---|
| 570 | >>> translator = Translator(pseudo_gettext) |
|---|
| 571 | |
|---|
| 572 | Next, the translator needs to be prepended to any already defined filters |
|---|
| 573 | on the template: |
|---|
| 574 | |
|---|
| 575 | >>> tmpl.filters.insert(0, translator) |
|---|
| 576 | |
|---|
| 577 | When generating the template output, our hard-coded translations should be |
|---|
| 578 | applied as expected: |
|---|
| 579 | |
|---|
| 580 | >>> print(tmpl.generate(username='Hans', _=pseudo_gettext)) |
|---|
| 581 | <html> |
|---|
| 582 | <head> |
|---|
| 583 | <title>Beispiel</title> |
|---|
| 584 | </head> |
|---|
| 585 | <body> |
|---|
| 586 | <h1>Beispiel</h1> |
|---|
| 587 | <p>Hallo, Hans</p> |
|---|
| 588 | </body> |
|---|
| 589 | </html> |
|---|
| 590 | |
|---|
| 591 | Note that elements defining ``xml:lang`` attributes that do not contain |
|---|
| 592 | variable expressions are ignored by this filter. That can be used to |
|---|
| 593 | exclude specific parts of a template from being extracted and translated. |
|---|
| 594 | """ |
|---|
| 595 | |
|---|
| 596 | directives = [ |
|---|
| 597 | ('domain', DomainDirective), |
|---|
| 598 | ('comment', CommentDirective), |
|---|
| 599 | ('msg', MsgDirective), |
|---|
| 600 | ('choose', ChooseDirective), |
|---|
| 601 | ('singular', SingularDirective), |
|---|
| 602 | ('plural', PluralDirective) |
|---|
| 603 | ] |
|---|
| 604 | |
|---|
| 605 | IGNORE_TAGS = frozenset([ |
|---|
| 606 | QName('script'), QName('http://www.w3.org/1999/xhtml}script'), |
|---|
| 607 | QName('style'), QName('http://www.w3.org/1999/xhtml}style') |
|---|
| 608 | ]) |
|---|
| 609 | INCLUDE_ATTRS = frozenset([ |
|---|
| 610 | 'abbr', 'alt', 'label', 'prompt', 'standby', 'summary', 'title', |
|---|
| 611 | 'placeholder', |
|---|
| 612 | ]) |
|---|
| 613 | NAMESPACE = I18N_NAMESPACE |
|---|
| 614 | |
|---|
| 615 | def __init__(self, translate=NullTranslations(), ignore_tags=IGNORE_TAGS, |
|---|
| 616 | include_attrs=INCLUDE_ATTRS, extract_text=True): |
|---|
| 617 | """Initialize the translator. |
|---|
| 618 | |
|---|
| 619 | :param translate: the translation function, for example ``gettext`` or |
|---|
| 620 | ``ugettext``. |
|---|
| 621 | :param ignore_tags: a set of tag names that should not be localized |
|---|
| 622 | :param include_attrs: a set of attribute names should be localized |
|---|
| 623 | :param extract_text: whether the content of text nodes should be |
|---|
| 624 | extracted, or only text in explicit ``gettext`` |
|---|
| 625 | function calls |
|---|
| 626 | |
|---|
| 627 | :note: Changed in 0.6: the `translate` parameter can now be either |
|---|
| 628 | a ``gettext``-style function, or an object compatible with the |
|---|
| 629 | ``NullTransalations`` or ``GNUTranslations`` interface |
|---|
| 630 | """ |
|---|
| 631 | self.translate = translate |
|---|
| 632 | self.ignore_tags = ignore_tags |
|---|
| 633 | self.include_attrs = include_attrs |
|---|
| 634 | self.extract_text = extract_text |
|---|
| 635 | |
|---|
| 636 | def __call__(self, stream, ctxt=None, translate_text=True, |
|---|
| 637 | translate_attrs=True): |
|---|
| 638 | """Translate any localizable strings in the given stream. |
|---|
| 639 | |
|---|
| 640 | This function shouldn't be called directly. Instead, an instance of |
|---|
| 641 | the `Translator` class should be registered as a filter with the |
|---|
| 642 | `Template` or the `TemplateLoader`, or applied as a regular stream |
|---|
| 643 | filter. If used as a template filter, it should be inserted in front of |
|---|
| 644 | all the default filters. |
|---|
| 645 | |
|---|
| 646 | :param stream: the markup event stream |
|---|
| 647 | :param ctxt: the template context (not used) |
|---|
| 648 | :param translate_text: whether text nodes should be translated (used |
|---|
| 649 | internally) |
|---|
| 650 | :param translate_attrs: whether attribute values should be translated |
|---|
| 651 | (used internally) |
|---|
| 652 | :return: the localized stream |
|---|
| 653 | """ |
|---|
| 654 | ignore_tags = self.ignore_tags |
|---|
| 655 | include_attrs = self.include_attrs |
|---|
| 656 | skip = 0 |
|---|
| 657 | xml_lang = XML_NAMESPACE['lang'] |
|---|
| 658 | if not self.extract_text: |
|---|
| 659 | translate_text = False |
|---|
| 660 | translate_attrs = False |
|---|
| 661 | |
|---|
| 662 | if type(self.translate) is FunctionType: |
|---|
| 663 | gettext = self.translate |
|---|
| 664 | if ctxt: |
|---|
| 665 | ctxt['_i18n.gettext'] = gettext |
|---|
| 666 | else: |
|---|
| 667 | gettext = self.translate.ugettext |
|---|
| 668 | ngettext = self.translate.ungettext |
|---|
| 669 | try: |
|---|
| 670 | dgettext = self.translate.dugettext |
|---|
| 671 | dngettext = self.translate.dungettext |
|---|
| 672 | except AttributeError: |
|---|
| 673 | dgettext = lambda _, y: gettext(y) |
|---|
| 674 | dngettext = lambda _, s, p, n: ngettext(s, p, n) |
|---|
| 675 | if ctxt: |
|---|
| 676 | ctxt['_i18n.gettext'] = gettext |
|---|
| 677 | ctxt['_i18n.ngettext'] = ngettext |
|---|
| 678 | ctxt['_i18n.dgettext'] = dgettext |
|---|
| 679 | ctxt['_i18n.dngettext'] = dngettext |
|---|
| 680 | |
|---|
| 681 | if ctxt and ctxt.get('_i18n.domain'): |
|---|
| 682 | gettext = lambda msg: dgettext(ctxt.get('_i18n.domain'), msg) |
|---|
| 683 | |
|---|
| 684 | for kind, data, pos in stream: |
|---|
| 685 | |
|---|
| 686 | # skip chunks that should not be localized |
|---|
| 687 | if skip: |
|---|
| 688 | if kind is START: |
|---|
| 689 | skip += 1 |
|---|
| 690 | elif kind is END: |
|---|
| 691 | skip -= 1 |
|---|
| 692 | yield kind, data, pos |
|---|
| 693 | continue |
|---|
| 694 | |
|---|
| 695 | # handle different events that can be localized |
|---|
| 696 | if kind is START: |
|---|
| 697 | tag, attrs = data |
|---|
| 698 | if tag in self.ignore_tags or \ |
|---|
| 699 | isinstance(attrs.get(xml_lang), basestring): |
|---|
| 700 | skip += 1 |
|---|
| 701 | yield kind, data, pos |
|---|
| 702 | continue |
|---|
| 703 | |
|---|
| 704 | new_attrs = [] |
|---|
| 705 | changed = False |
|---|
| 706 | |
|---|
| 707 | for name, value in attrs: |
|---|
| 708 | newval = value |
|---|
| 709 | if isinstance(value, basestring): |
|---|
| 710 | if translate_attrs and name in include_attrs: |
|---|
| 711 | newval = gettext(value) |
|---|
| 712 | else: |
|---|
| 713 | newval = list( |
|---|
| 714 | self(_ensure(value), ctxt, translate_text=False) |
|---|
| 715 | ) |
|---|
| 716 | if newval != value: |
|---|
| 717 | value = newval |
|---|
| 718 | changed = True |
|---|
| 719 | new_attrs.append((name, value)) |
|---|
| 720 | if changed: |
|---|
| 721 | attrs = Attrs(new_attrs) |
|---|
| 722 | |
|---|
| 723 | yield kind, (tag, attrs), pos |
|---|
| 724 | |
|---|
| 725 | elif translate_text and kind is TEXT: |
|---|
| 726 | text = data.strip() |
|---|
| 727 | if text: |
|---|
| 728 | data = data.replace(text, unicode(gettext(text))) |
|---|
| 729 | yield kind, data, pos |
|---|
| 730 | |
|---|
| 731 | elif kind is SUB: |
|---|
| 732 | directives, substream = data |
|---|
| 733 | current_domain = None |
|---|
| 734 | for idx, directive in enumerate(directives): |
|---|
| 735 | # Organize directives to make everything work |
|---|
| 736 | # FIXME: There's got to be a better way to do this! |
|---|
| 737 | if isinstance(directive, DomainDirective): |
|---|
| 738 | # Grab current domain and update context |
|---|
| 739 | current_domain = directive.domain |
|---|
| 740 | ctxt.push({'_i18n.domain': current_domain}) |
|---|
| 741 | # Put domain directive as the first one in order to |
|---|
| 742 | # update context before any other directives evaluation |
|---|
| 743 | directives.insert(0, directives.pop(idx)) |
|---|
| 744 | |
|---|
| 745 | # If this is an i18n directive, no need to translate text |
|---|
| 746 | # nodes here |
|---|
| 747 | is_i18n_directive = any([ |
|---|
| 748 | isinstance(d, ExtractableI18NDirective) |
|---|
| 749 | for d in directives |
|---|
| 750 | ]) |
|---|
| 751 | substream = list(self(substream, ctxt, |
|---|
| 752 | translate_text=not is_i18n_directive, |
|---|
| 753 | translate_attrs=translate_attrs)) |
|---|
| 754 | yield kind, (directives, substream), pos |
|---|
| 755 | |
|---|
| 756 | if current_domain: |
|---|
| 757 | ctxt.pop() |
|---|
| 758 | else: |
|---|
| 759 | yield kind, data, pos |
|---|
| 760 | |
|---|
| 761 | def extract(self, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| 762 | search_text=True, comment_stack=None): |
|---|
| 763 | """Extract localizable strings from the given template stream. |
|---|
| 764 | |
|---|
| 765 | For every string found, this function yields a ``(lineno, function, |
|---|
| 766 | message, comments)`` tuple, where: |
|---|
| 767 | |
|---|
| 768 | * ``lineno`` is the number of the line on which the string was found, |
|---|
| 769 | * ``function`` is the name of the ``gettext`` function used (if the |
|---|
| 770 | string was extracted from embedded Python code), and |
|---|
| 771 | * ``message`` is the string itself (a ``unicode`` object, or a tuple |
|---|
| 772 | of ``unicode`` objects for functions with multiple string |
|---|
| 773 | arguments). |
|---|
| 774 | * ``comments`` is a list of comments related to the message, extracted |
|---|
| 775 | from ``i18n:comment`` attributes found in the markup |
|---|
| 776 | |
|---|
| 777 | >>> tmpl = MarkupTemplate('''<html xmlns:py="http://genshi.edgewall.org/"> |
|---|
| 778 | ... <head> |
|---|
| 779 | ... <title>Example</title> |
|---|
| 780 | ... </head> |
|---|
| 781 | ... <body> |
|---|
| 782 | ... <h1>Example</h1> |
|---|
| 783 | ... <p>${_("Hello, %(name)s") % dict(name=username)}</p> |
|---|
| 784 | ... <p>${ngettext("You have %d item", "You have %d items", num)}</p> |
|---|
| 785 | ... </body> |
|---|
| 786 | ... </html>''', filename='example.html') |
|---|
| 787 | >>> for line, func, msg, comments in Translator().extract(tmpl.stream): |
|---|
| 788 | ... print('%d, %r, %r' % (line, func, msg)) |
|---|
| 789 | 3, None, u'Example' |
|---|
| 790 | 6, None, u'Example' |
|---|
| 791 | 7, '_', u'Hello, %(name)s' |
|---|
| 792 | 8, 'ngettext', (u'You have %d item', u'You have %d items', None) |
|---|
| 793 | |
|---|
| 794 | :param stream: the event stream to extract strings from; can be a |
|---|
| 795 | regular stream or a template stream |
|---|
| 796 | :param gettext_functions: a sequence of function names that should be |
|---|
| 797 | treated as gettext-style localization |
|---|
| 798 | functions |
|---|
| 799 | :param search_text: whether the content of text nodes should be |
|---|
| 800 | extracted (used internally) |
|---|
| 801 | |
|---|
| 802 | :note: Changed in 0.4.1: For a function with multiple string arguments |
|---|
| 803 | (such as ``ngettext``), a single item with a tuple of strings is |
|---|
| 804 | yielded, instead an item for each string argument. |
|---|
| 805 | :note: Changed in 0.6: The returned tuples now include a fourth |
|---|
| 806 | element, which is a list of comments for the translator. |
|---|
| 807 | """ |
|---|
| 808 | if not self.extract_text: |
|---|
| 809 | search_text = False |
|---|
| 810 | if comment_stack is None: |
|---|
| 811 | comment_stack = [] |
|---|
| 812 | skip = 0 |
|---|
| 813 | |
|---|
| 814 | xml_lang = XML_NAMESPACE['lang'] |
|---|
| 815 | |
|---|
| 816 | for kind, data, pos in stream: |
|---|
| 817 | if skip: |
|---|
| 818 | if kind is START: |
|---|
| 819 | skip += 1 |
|---|
| 820 | if kind is END: |
|---|
| 821 | skip -= 1 |
|---|
| 822 | |
|---|
| 823 | if kind is START and not skip: |
|---|
| 824 | tag, attrs = data |
|---|
| 825 | if tag in self.ignore_tags or \ |
|---|
| 826 | isinstance(attrs.get(xml_lang), basestring): |
|---|
| 827 | skip += 1 |
|---|
| 828 | continue |
|---|
| 829 | |
|---|
| 830 | for message in self._extract_attrs((kind, data, pos), |
|---|
| 831 | gettext_functions, |
|---|
| 832 | search_text=search_text): |
|---|
| 833 | yield message |
|---|
| 834 | |
|---|
| 835 | elif not skip and search_text and kind is TEXT: |
|---|
| 836 | text = data.strip() |
|---|
| 837 | if text and [ch for ch in text if ch.isalpha()]: |
|---|
| 838 | yield pos[1], None, text, comment_stack[-1:] |
|---|
| 839 | |
|---|
| 840 | elif kind is EXPR or kind is EXEC: |
|---|
| 841 | for funcname, strings in extract_from_code(data, |
|---|
| 842 | gettext_functions): |
|---|
| 843 | # XXX: Do we need to grab i18n:comment from comment_stack ??? |
|---|
| 844 | yield pos[1], funcname, strings, [] |
|---|
| 845 | |
|---|
| 846 | elif kind is SUB: |
|---|
| 847 | directives, substream = data |
|---|
| 848 | in_comment = False |
|---|
| 849 | |
|---|
| 850 | for idx, directive in enumerate(directives): |
|---|
| 851 | # Do a first loop to see if there's a comment directive |
|---|
| 852 | # If there is update context and pop it from directives |
|---|
| 853 | if isinstance(directive, CommentDirective): |
|---|
| 854 | in_comment = True |
|---|
| 855 | comment_stack.append(directive.comment) |
|---|
| 856 | if len(directives) == 1: |
|---|
| 857 | # in case we're in the presence of something like: |
|---|
| 858 | # <p i18n:comment="foo">Foo</p> |
|---|
| 859 | for message in self.extract( |
|---|
| 860 | substream, gettext_functions, |
|---|
| 861 | search_text=search_text and not skip, |
|---|
| 862 | comment_stack=comment_stack): |
|---|
| 863 | yield message |
|---|
| 864 | directives.pop(idx) |
|---|
| 865 | elif not isinstance(directive, I18NDirective): |
|---|
| 866 | # Remove all other non i18n directives from the process |
|---|
| 867 | directives.pop(idx) |
|---|
| 868 | |
|---|
| 869 | if not directives and not in_comment: |
|---|
| 870 | # Extract content if there's no directives because |
|---|
| 871 | # strip was pop'ed and not because comment was pop'ed. |
|---|
| 872 | # Extraction in this case has been taken care of. |
|---|
| 873 | for message in self.extract( |
|---|
| 874 | substream, gettext_functions, |
|---|
| 875 | search_text=search_text and not skip): |
|---|
| 876 | yield message |
|---|
| 877 | |
|---|
| 878 | for directive in directives: |
|---|
| 879 | if isinstance(directive, ExtractableI18NDirective): |
|---|
| 880 | for message in directive.extract(self, |
|---|
| 881 | substream, gettext_functions, |
|---|
| 882 | search_text=search_text and not skip, |
|---|
| 883 | comment_stack=comment_stack): |
|---|
| 884 | yield message |
|---|
| 885 | else: |
|---|
| 886 | for message in self.extract( |
|---|
| 887 | substream, gettext_functions, |
|---|
| 888 | search_text=search_text and not skip, |
|---|
| 889 | comment_stack=comment_stack): |
|---|
| 890 | yield message |
|---|
| 891 | |
|---|
| 892 | if in_comment: |
|---|
| 893 | comment_stack.pop() |
|---|
| 894 | |
|---|
| 895 | def get_directive_index(self, dir_cls): |
|---|
| 896 | total = len(self._dir_order) |
|---|
| 897 | if dir_cls in self._dir_order: |
|---|
| 898 | return self._dir_order.index(dir_cls) - total |
|---|
| 899 | return total |
|---|
| 900 | |
|---|
| 901 | def setup(self, template): |
|---|
| 902 | """Convenience function to register the `Translator` filter and the |
|---|
| 903 | related directives with the given template. |
|---|
| 904 | |
|---|
| 905 | :param template: a `Template` instance |
|---|
| 906 | """ |
|---|
| 907 | template.filters.insert(0, self) |
|---|
| 908 | if hasattr(template, 'add_directives'): |
|---|
| 909 | template.add_directives(Translator.NAMESPACE, self) |
|---|
| 910 | |
|---|
| 911 | def _extract_attrs(self, event, gettext_functions, search_text): |
|---|
| 912 | for name, value in event[1][1]: |
|---|
| 913 | if search_text and isinstance(value, basestring): |
|---|
| 914 | if name in self.include_attrs: |
|---|
| 915 | text = value.strip() |
|---|
| 916 | if text: |
|---|
| 917 | yield event[2][1], None, text, [] |
|---|
| 918 | else: |
|---|
| 919 | for message in self.extract(_ensure(value), gettext_functions, |
|---|
| 920 | search_text=False): |
|---|
| 921 | yield message |
|---|
| 922 | |
|---|
| 923 | |
|---|
| 924 | class MessageBuffer(object): |
|---|
| 925 | """Helper class for managing internationalized mixed content. |
|---|
| 926 | |
|---|
| 927 | :since: version 0.5 |
|---|
| 928 | """ |
|---|
| 929 | |
|---|
| 930 | def __init__(self, directive=None): |
|---|
| 931 | """Initialize the message buffer. |
|---|
| 932 | |
|---|
| 933 | :param directive: the directive owning the buffer |
|---|
| 934 | :type directive: I18NDirective |
|---|
| 935 | """ |
|---|
| 936 | # params list needs to be copied so that directives can be evaluated |
|---|
| 937 | # more than once |
|---|
| 938 | self.orig_params = self.params = directive.params[:] |
|---|
| 939 | self.directive = directive |
|---|
| 940 | self.string = [] |
|---|
| 941 | self.events = {} |
|---|
| 942 | self.values = {} |
|---|
| 943 | self.depth = 1 |
|---|
| 944 | self.order = 1 |
|---|
| 945 | self._prev_order = None |
|---|
| 946 | self.stack = [0] |
|---|
| 947 | self.subdirectives = {} |
|---|
| 948 | |
|---|
| 949 | def _add_event(self, order, event): |
|---|
| 950 | if order == self._prev_order: |
|---|
| 951 | self.events[order][-1].append(event) |
|---|
| 952 | else: |
|---|
| 953 | self._prev_order = order |
|---|
| 954 | self.events.setdefault(order, []) |
|---|
| 955 | self.events[order].append([event]) |
|---|
| 956 | |
|---|
| 957 | def append(self, kind, data, pos): |
|---|
| 958 | """Append a stream event to the buffer. |
|---|
| 959 | |
|---|
| 960 | :param kind: the stream event kind |
|---|
| 961 | :param data: the event data |
|---|
| 962 | :param pos: the position of the event in the source |
|---|
| 963 | """ |
|---|
| 964 | if kind is SUB: |
|---|
| 965 | # The order needs to be +1 because a new START kind event will |
|---|
| 966 | # happen and we we need to wrap those events into our custom kind(s) |
|---|
| 967 | order = self.stack[-1] + 1 |
|---|
| 968 | subdirectives, substream = data |
|---|
| 969 | # Store the directives that should be applied after translation |
|---|
| 970 | self.subdirectives.setdefault(order, []).extend(subdirectives) |
|---|
| 971 | self._add_event(order, (SUB_START, None, pos)) |
|---|
| 972 | for skind, sdata, spos in substream: |
|---|
| 973 | self.append(skind, sdata, spos) |
|---|
| 974 | self._add_event(order, (SUB_END, None, pos)) |
|---|
| 975 | elif kind is TEXT: |
|---|
| 976 | if '[' in data or ']' in data: |
|---|
| 977 | # Quote [ and ] if it ain't us adding it, ie, if the user is |
|---|
| 978 | # using those chars in his templates, escape them |
|---|
| 979 | data = data.replace('[', '\[').replace(']', '\]') |
|---|
| 980 | self.string.append(data) |
|---|
| 981 | self._add_event(self.stack[-1], (kind, data, pos)) |
|---|
| 982 | elif kind is EXPR: |
|---|
| 983 | if self.params: |
|---|
| 984 | param = self.params.pop(0) |
|---|
| 985 | else: |
|---|
| 986 | params = ', '.join(['"%s"' % p for p in self.orig_params if p]) |
|---|
| 987 | if params: |
|---|
| 988 | params = "(%s)" % params |
|---|
| 989 | raise IndexError("%d parameters%s given to 'i18n:%s' but " |
|---|
| 990 | "%d or more expressions used in '%s', line %s" |
|---|
| 991 | % (len(self.orig_params), params, |
|---|
| 992 | self.directive.tagname, |
|---|
| 993 | len(self.orig_params) + 1, |
|---|
| 994 | os.path.basename(pos[0] or |
|---|
| 995 | 'In-memory Template'), |
|---|
| 996 | pos[1])) |
|---|
| 997 | self.string.append('%%(%s)s' % param) |
|---|
| 998 | self._add_event(self.stack[-1], (kind, data, pos)) |
|---|
| 999 | self.values[param] = (kind, data, pos) |
|---|
| 1000 | else: |
|---|
| 1001 | if kind is START: |
|---|
| 1002 | self.string.append('[%d:' % self.order) |
|---|
| 1003 | self.stack.append(self.order) |
|---|
| 1004 | self._add_event(self.stack[-1], (kind, data, pos)) |
|---|
| 1005 | self.depth += 1 |
|---|
| 1006 | self.order += 1 |
|---|
| 1007 | elif kind is END: |
|---|
| 1008 | self.depth -= 1 |
|---|
| 1009 | if self.depth: |
|---|
| 1010 | self._add_event(self.stack[-1], (kind, data, pos)) |
|---|
| 1011 | self.string.append(']') |
|---|
| 1012 | self.stack.pop() |
|---|
| 1013 | |
|---|
| 1014 | def format(self): |
|---|
| 1015 | """Return a message identifier representing the content in the |
|---|
| 1016 | buffer. |
|---|
| 1017 | """ |
|---|
| 1018 | return ''.join(self.string).strip() |
|---|
| 1019 | |
|---|
| 1020 | def translate(self, string, regex=re.compile(r'%\((\w+)\)s')): |
|---|
| 1021 | """Interpolate the given message translation with the events in the |
|---|
| 1022 | buffer and return the translated stream. |
|---|
| 1023 | |
|---|
| 1024 | :param string: the translated message string |
|---|
| 1025 | """ |
|---|
| 1026 | substream = None |
|---|
| 1027 | |
|---|
| 1028 | def yield_parts(string): |
|---|
| 1029 | for idx, part in enumerate(regex.split(string)): |
|---|
| 1030 | if idx % 2: |
|---|
| 1031 | yield self.values[part] |
|---|
| 1032 | elif part: |
|---|
| 1033 | yield (TEXT, |
|---|
| 1034 | part.replace('\[', '[').replace('\]', ']'), |
|---|
| 1035 | (None, -1, -1) |
|---|
| 1036 | ) |
|---|
| 1037 | |
|---|
| 1038 | parts = parse_msg(string) |
|---|
| 1039 | parts_counter = {} |
|---|
| 1040 | for order, string in parts: |
|---|
| 1041 | parts_counter.setdefault(order, []).append(None) |
|---|
| 1042 | |
|---|
| 1043 | while parts: |
|---|
| 1044 | order, string = parts.pop(0) |
|---|
| 1045 | events = self.events[order] |
|---|
| 1046 | if events: |
|---|
| 1047 | events = events.pop(0) |
|---|
| 1048 | else: |
|---|
| 1049 | # create a dummy empty text event so any remaining |
|---|
| 1050 | # part of the translation can be processed. |
|---|
| 1051 | events = [(TEXT, "", (None, -1, -1))] |
|---|
| 1052 | parts_counter[order].pop() |
|---|
| 1053 | |
|---|
| 1054 | for event in events: |
|---|
| 1055 | if event[0] is SUB_START: |
|---|
| 1056 | substream = [] |
|---|
| 1057 | elif event[0] is SUB_END: |
|---|
| 1058 | # Yield a substream which might have directives to be |
|---|
| 1059 | # applied to it (after translation events) |
|---|
| 1060 | yield SUB, (self.subdirectives[order], substream), event[2] |
|---|
| 1061 | substream = None |
|---|
| 1062 | elif event[0] is TEXT: |
|---|
| 1063 | if string: |
|---|
| 1064 | for part in yield_parts(string): |
|---|
| 1065 | if substream is not None: |
|---|
| 1066 | substream.append(part) |
|---|
| 1067 | else: |
|---|
| 1068 | yield part |
|---|
| 1069 | # String handled, reset it |
|---|
| 1070 | string = None |
|---|
| 1071 | elif event[0] is START: |
|---|
| 1072 | if substream is not None: |
|---|
| 1073 | substream.append(event) |
|---|
| 1074 | else: |
|---|
| 1075 | yield event |
|---|
| 1076 | if string: |
|---|
| 1077 | for part in yield_parts(string): |
|---|
| 1078 | if substream is not None: |
|---|
| 1079 | substream.append(part) |
|---|
| 1080 | else: |
|---|
| 1081 | yield part |
|---|
| 1082 | # String handled, reset it |
|---|
| 1083 | string = None |
|---|
| 1084 | elif event[0] is END: |
|---|
| 1085 | if string: |
|---|
| 1086 | for part in yield_parts(string): |
|---|
| 1087 | if substream is not None: |
|---|
| 1088 | substream.append(part) |
|---|
| 1089 | else: |
|---|
| 1090 | yield part |
|---|
| 1091 | # String handled, reset it |
|---|
| 1092 | string = None |
|---|
| 1093 | if substream is not None: |
|---|
| 1094 | substream.append(event) |
|---|
| 1095 | else: |
|---|
| 1096 | yield event |
|---|
| 1097 | elif event[0] is EXPR: |
|---|
| 1098 | # These are handled on the strings itself |
|---|
| 1099 | continue |
|---|
| 1100 | else: |
|---|
| 1101 | if string: |
|---|
| 1102 | for part in yield_parts(string): |
|---|
| 1103 | if substream is not None: |
|---|
| 1104 | substream.append(part) |
|---|
| 1105 | else: |
|---|
| 1106 | yield part |
|---|
| 1107 | # String handled, reset it |
|---|
| 1108 | string = None |
|---|
| 1109 | if substream is not None: |
|---|
| 1110 | substream.append(event) |
|---|
| 1111 | else: |
|---|
| 1112 | yield event |
|---|
| 1113 | |
|---|
| 1114 | |
|---|
| 1115 | def parse_msg(string, regex=re.compile(r'(?:\[(\d+)\:)|(?<!\\)\]')): |
|---|
| 1116 | """Parse a translated message using Genshi mixed content message |
|---|
| 1117 | formatting. |
|---|
| 1118 | |
|---|
| 1119 | >>> parse_msg("See [1:Help].") |
|---|
| 1120 | [(0, 'See '), (1, 'Help'), (0, '.')] |
|---|
| 1121 | |
|---|
| 1122 | >>> parse_msg("See [1:our [2:Help] page] for details.") |
|---|
| 1123 | [(0, 'See '), (1, 'our '), (2, 'Help'), (1, ' page'), (0, ' for details.')] |
|---|
| 1124 | |
|---|
| 1125 | >>> parse_msg("[2:Details] finden Sie in [1:Hilfe].") |
|---|
| 1126 | [(2, 'Details'), (0, ' finden Sie in '), (1, 'Hilfe'), (0, '.')] |
|---|
| 1127 | |
|---|
| 1128 | >>> parse_msg("[1:] Bilder pro Seite anzeigen.") |
|---|
| 1129 | [(1, ''), (0, ' Bilder pro Seite anzeigen.')] |
|---|
| 1130 | |
|---|
| 1131 | :param string: the translated message string |
|---|
| 1132 | :return: a list of ``(order, string)`` tuples |
|---|
| 1133 | :rtype: `list` |
|---|
| 1134 | """ |
|---|
| 1135 | parts = [] |
|---|
| 1136 | stack = [0] |
|---|
| 1137 | while True: |
|---|
| 1138 | mo = regex.search(string) |
|---|
| 1139 | if not mo: |
|---|
| 1140 | break |
|---|
| 1141 | |
|---|
| 1142 | if mo.start() or stack[-1]: |
|---|
| 1143 | parts.append((stack[-1], string[:mo.start()])) |
|---|
| 1144 | string = string[mo.end():] |
|---|
| 1145 | |
|---|
| 1146 | orderno = mo.group(1) |
|---|
| 1147 | if orderno is not None: |
|---|
| 1148 | stack.append(int(orderno)) |
|---|
| 1149 | else: |
|---|
| 1150 | stack.pop() |
|---|
| 1151 | if not stack: |
|---|
| 1152 | break |
|---|
| 1153 | |
|---|
| 1154 | if string: |
|---|
| 1155 | parts.append((stack[-1], string)) |
|---|
| 1156 | |
|---|
| 1157 | return parts |
|---|
| 1158 | |
|---|
| 1159 | |
|---|
| 1160 | def extract_from_code(code, gettext_functions): |
|---|
| 1161 | """Extract strings from Python bytecode. |
|---|
| 1162 | |
|---|
| 1163 | >>> from genshi.template.eval import Expression |
|---|
| 1164 | >>> expr = Expression('_("Hello")') |
|---|
| 1165 | >>> list(extract_from_code(expr, GETTEXT_FUNCTIONS)) |
|---|
| 1166 | [('_', u'Hello')] |
|---|
| 1167 | |
|---|
| 1168 | >>> expr = Expression('ngettext("You have %(num)s item", ' |
|---|
| 1169 | ... '"You have %(num)s items", num)') |
|---|
| 1170 | >>> list(extract_from_code(expr, GETTEXT_FUNCTIONS)) |
|---|
| 1171 | [('ngettext', (u'You have %(num)s item', u'You have %(num)s items', None))] |
|---|
| 1172 | |
|---|
| 1173 | :param code: the `Code` object |
|---|
| 1174 | :type code: `genshi.template.eval.Code` |
|---|
| 1175 | :param gettext_functions: a sequence of function names |
|---|
| 1176 | :since: version 0.5 |
|---|
| 1177 | """ |
|---|
| 1178 | def _walk(node): |
|---|
| 1179 | if isinstance(node, _ast.Call) and isinstance(node.func, _ast.Name) \ |
|---|
| 1180 | and node.func.id in gettext_functions: |
|---|
| 1181 | strings = [] |
|---|
| 1182 | def _add(arg): |
|---|
| 1183 | if isinstance(arg, _ast.Str) and isinstance(arg.s, basestring): |
|---|
| 1184 | strings.append(unicode(arg.s, 'utf-8')) |
|---|
| 1185 | elif arg: |
|---|
| 1186 | strings.append(None) |
|---|
| 1187 | [_add(arg) for arg in node.args] |
|---|
| 1188 | _add(node.starargs) |
|---|
| 1189 | _add(node.kwargs) |
|---|
| 1190 | if len(strings) == 1: |
|---|
| 1191 | strings = strings[0] |
|---|
| 1192 | else: |
|---|
| 1193 | strings = tuple(strings) |
|---|
| 1194 | yield node.func.id, strings |
|---|
| 1195 | elif node._fields: |
|---|
| 1196 | children = [] |
|---|
| 1197 | for field in node._fields: |
|---|
| 1198 | child = getattr(node, field, None) |
|---|
| 1199 | if isinstance(child, list): |
|---|
| 1200 | for elem in child: |
|---|
| 1201 | children.append(elem) |
|---|
| 1202 | elif isinstance(child, _ast.AST): |
|---|
| 1203 | children.append(child) |
|---|
| 1204 | for child in children: |
|---|
| 1205 | for funcname, strings in _walk(child): |
|---|
| 1206 | yield funcname, strings |
|---|
| 1207 | return _walk(code.ast) |
|---|
| 1208 | |
|---|
| 1209 | |
|---|
| 1210 | def extract(fileobj, keywords, comment_tags, options): |
|---|
| 1211 | """Babel extraction method for Genshi templates. |
|---|
| 1212 | |
|---|
| 1213 | :param fileobj: the file-like object the messages should be extracted from |
|---|
| 1214 | :param keywords: a list of keywords (i.e. function names) that should be |
|---|
| 1215 | recognized as translation functions |
|---|
| 1216 | :param comment_tags: a list of translator tags to search for and include |
|---|
| 1217 | in the results |
|---|
| 1218 | :param options: a dictionary of additional options (optional) |
|---|
| 1219 | :return: an iterator over ``(lineno, funcname, message, comments)`` tuples |
|---|
| 1220 | :rtype: ``iterator`` |
|---|
| 1221 | """ |
|---|
| 1222 | template_class = options.get('template_class', MarkupTemplate) |
|---|
| 1223 | if isinstance(template_class, basestring): |
|---|
| 1224 | module, clsname = template_class.split(':', 1) |
|---|
| 1225 | template_class = getattr(__import__(module, {}, {}, [clsname]), clsname) |
|---|
| 1226 | encoding = options.get('encoding', None) |
|---|
| 1227 | |
|---|
| 1228 | extract_text = options.get('extract_text', True) |
|---|
| 1229 | if isinstance(extract_text, basestring): |
|---|
| 1230 | extract_text = extract_text.lower() in ('1', 'on', 'yes', 'true') |
|---|
| 1231 | |
|---|
| 1232 | ignore_tags = options.get('ignore_tags', Translator.IGNORE_TAGS) |
|---|
| 1233 | if isinstance(ignore_tags, basestring): |
|---|
| 1234 | ignore_tags = ignore_tags.split() |
|---|
| 1235 | ignore_tags = [QName(tag) for tag in ignore_tags] |
|---|
| 1236 | |
|---|
| 1237 | include_attrs = options.get('include_attrs', Translator.INCLUDE_ATTRS) |
|---|
| 1238 | if isinstance(include_attrs, basestring): |
|---|
| 1239 | include_attrs = include_attrs.split() |
|---|
| 1240 | include_attrs = [QName(attr) for attr in include_attrs] |
|---|
| 1241 | |
|---|
| 1242 | tmpl = template_class(fileobj, filename=getattr(fileobj, 'name', None), |
|---|
| 1243 | encoding=encoding) |
|---|
| 1244 | tmpl.loader = None |
|---|
| 1245 | |
|---|
| 1246 | translator = Translator(None, ignore_tags, include_attrs, extract_text) |
|---|
| 1247 | if hasattr(tmpl, 'add_directives'): |
|---|
| 1248 | tmpl.add_directives(Translator.NAMESPACE, translator) |
|---|
| 1249 | for message in translator.extract(tmpl.stream, gettext_functions=keywords): |
|---|
| 1250 | yield message |
|---|