| [637] | 1 | # -*- coding: utf-8 -*- |
|---|
| 2 | # |
|---|
| [1120] | 3 | # Copyright (C) 2007-2010 Edgewall Software |
|---|
| [637] | 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 | |
|---|
| [1072] | 14 | """Directives and utilities for internationalization and localization of |
|---|
| 15 | templates. |
|---|
| [538] | 16 | |
|---|
| [690] | 17 | :since: version 0.4 |
|---|
| [1072] | 18 | :note: Directives support added since version 0.6 |
|---|
| [690] | 19 | """ |
|---|
| 20 | |
|---|
| [1079] | 21 | try: |
|---|
| 22 | any |
|---|
| 23 | except NameError: |
|---|
| 24 | from genshi.util import any |
|---|
| [932] | 25 | from gettext import NullTranslations |
|---|
| [1072] | 26 | import os |
|---|
| [538] | 27 | import re |
|---|
| [932] | 28 | from types import FunctionType |
|---|
| [538] | 29 | |
|---|
| [1118] | 30 | from genshi.core import Attrs, Namespace, QName, START, END, TEXT, \ |
|---|
| 31 | XML_NAMESPACE, _ensure, StreamEventKind |
|---|
| [988] | 32 | from genshi.template.eval import _ast |
|---|
| [954] | 33 | from genshi.template.base import DirectiveFactory, EXPR, SUB, _apply_directives |
|---|
| [1072] | 34 | from genshi.template.directives import Directive, StripDirective |
|---|
| [634] | 35 | from genshi.template.markup import MarkupTemplate, EXEC |
|---|
| [1158] | 36 | from genshi.compat import IS_PYTHON2 |
|---|
| [538] | 37 | |
|---|
| [634] | 38 | __all__ = ['Translator', 'extract'] |
|---|
| [605] | 39 | __docformat__ = 'restructuredtext en' |
|---|
| 40 | |
|---|
| [1072] | 41 | |
|---|
| [671] | 42 | I18N_NAMESPACE = Namespace('http://genshi.edgewall.org/i18n') |
|---|
| [538] | 43 | |
|---|
| [1072] | 44 | MSGBUF = StreamEventKind('MSGBUF') |
|---|
| 45 | SUB_START = StreamEventKind('SUB_START') |
|---|
| 46 | SUB_END = StreamEventKind('SUB_END') |
|---|
| [671] | 47 | |
|---|
| [1115] | 48 | GETTEXT_FUNCTIONS = ('_', 'gettext', 'ngettext', 'dgettext', 'dngettext', |
|---|
| 49 | 'ugettext', 'ungettext') |
|---|
| [954] | 50 | |
|---|
| [1115] | 51 | |
|---|
| [1072] | 52 | class I18NDirective(Directive): |
|---|
| 53 | """Simple interface for i18n directives to support messages extraction.""" |
|---|
| [954] | 54 | |
|---|
| [1072] | 55 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| 56 | return _apply_directives(stream, directives, ctxt, vars) |
|---|
| [954] | 57 | |
|---|
| 58 | |
|---|
| [1072] | 59 | class ExtractableI18NDirective(I18NDirective): |
|---|
| 60 | """Simple interface for directives to support messages extraction.""" |
|---|
| [954] | 61 | |
|---|
| [1115] | 62 | def extract(self, translator, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| 63 | search_text=True, comment_stack=None): |
|---|
| [1072] | 64 | raise NotImplementedError |
|---|
| 65 | |
|---|
| 66 | |
|---|
| 67 | class CommentDirective(I18NDirective): |
|---|
| 68 | """Implementation of the ``i18n:comment`` template directive which adds |
|---|
| 69 | translation comments. |
|---|
| 70 | |
|---|
| 71 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 72 | ... <p i18n:comment="As in Foo Bar">Foo</p> |
|---|
| 73 | ... </html>''') |
|---|
| 74 | >>> translator = Translator() |
|---|
| 75 | >>> translator.setup(tmpl) |
|---|
| 76 | >>> list(translator.extract(tmpl.stream)) |
|---|
| 77 | [(2, None, u'Foo', [u'As in Foo Bar'])] |
|---|
| 78 | """ |
|---|
| 79 | __slots__ = ['comment'] |
|---|
| 80 | |
|---|
| [1115] | 81 | def __init__(self, value, template=None, namespaces=None, lineno=-1, |
|---|
| 82 | offset=-1): |
|---|
| [1072] | 83 | Directive.__init__(self, None, template, namespaces, lineno, offset) |
|---|
| 84 | self.comment = value |
|---|
| 85 | |
|---|
| 86 | |
|---|
| 87 | class MsgDirective(ExtractableI18NDirective): |
|---|
| 88 | r"""Implementation of the ``i18n:msg`` directive which marks inner content |
|---|
| 89 | as translatable. Consider the following examples: |
|---|
| 90 | |
|---|
| 91 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 92 | ... <div i18n:msg=""> |
|---|
| 93 | ... <p>Foo</p> |
|---|
| 94 | ... <p>Bar</p> |
|---|
| 95 | ... </div> |
|---|
| 96 | ... <p i18n:msg="">Foo <em>bar</em>!</p> |
|---|
| 97 | ... </html>''') |
|---|
| 98 | |
|---|
| 99 | >>> translator = Translator() |
|---|
| 100 | >>> translator.setup(tmpl) |
|---|
| 101 | >>> list(translator.extract(tmpl.stream)) |
|---|
| 102 | [(2, None, u'[1:Foo]\n [2:Bar]', []), (6, None, u'Foo [1:bar]!', [])] |
|---|
| [1076] | 103 | >>> print(tmpl.generate().render()) |
|---|
| [1072] | 104 | <html> |
|---|
| 105 | <div><p>Foo</p> |
|---|
| 106 | <p>Bar</p></div> |
|---|
| 107 | <p>Foo <em>bar</em>!</p> |
|---|
| 108 | </html> |
|---|
| 109 | |
|---|
| 110 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 111 | ... <div i18n:msg="fname, lname"> |
|---|
| 112 | ... <p>First Name: ${fname}</p> |
|---|
| 113 | ... <p>Last Name: ${lname}</p> |
|---|
| 114 | ... </div> |
|---|
| 115 | ... <p i18n:msg="">Foo <em>bar</em>!</p> |
|---|
| 116 | ... </html>''') |
|---|
| 117 | >>> translator.setup(tmpl) |
|---|
| 118 | >>> list(translator.extract(tmpl.stream)) #doctest: +NORMALIZE_WHITESPACE |
|---|
| 119 | [(2, None, u'[1:First Name: %(fname)s]\n [2:Last Name: %(lname)s]', []), |
|---|
| 120 | (6, None, u'Foo [1:bar]!', [])] |
|---|
| 121 | |
|---|
| 122 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| 123 | ... <div i18n:msg="fname, lname"> |
|---|
| 124 | ... <p>First Name: ${fname}</p> |
|---|
| 125 | ... <p>Last Name: ${lname}</p> |
|---|
| 126 | ... </div> |
|---|
| 127 | ... <p i18n:msg="">Foo <em>bar</em>!</p> |
|---|
| 128 | ... </html>''') |
|---|
| 129 | >>> translator.setup(tmpl) |
|---|
| [1076] | 130 | >>> print(tmpl.generate(fname='John', lname='Doe').render()) |
|---|
| [1072] | 131 | <html> |
|---|
| 132 | <div><p>First Name: John</p> |
|---|
| 133 | <p>Last Name: Doe</p></div> |
|---|
| 134 | <p>Foo <em>bar</em>!</p> |
|---|
| 135 | </html> |
|---|
| 136 | |
|---|
| 137 | Starting and ending white-space is stripped of to make it simpler for |
|---|
| 138 | translators. Stripping it is not that important since it's on the html |
|---|
| 139 | source, the rendered output will remain the same. |
|---|
| 140 | """ |
|---|
| [1115] | 141 | __slots__ = ['params', 'lineno'] |
|---|
| [954] | 142 | |
|---|
| [1115] | 143 | def __init__(self, value, template=None, namespaces=None, lineno=-1, |
|---|
| 144 | offset=-1): |
|---|
| [954] | 145 | Directive.__init__(self, None, template, namespaces, lineno, offset) |
|---|
| [1072] | 146 | self.params = [param.strip() for param in value.split(',') if param] |
|---|
| [1115] | 147 | self.lineno = lineno |
|---|
| [954] | 148 | |
|---|
| [1072] | 149 | @classmethod |
|---|
| 150 | def attach(cls, template, stream, value, namespaces, pos): |
|---|
| 151 | if type(value) is dict: |
|---|
| 152 | value = value.get('params', '').strip() |
|---|
| 153 | return super(MsgDirective, cls).attach(template, stream, value.strip(), |
|---|
| 154 | namespaces, pos) |
|---|
| 155 | |
|---|
| [954] | 156 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| [1072] | 157 | gettext = ctxt.get('_i18n.gettext') |
|---|
| 158 | if ctxt.get('_i18n.domain'): |
|---|
| [1114] | 159 | dgettext = ctxt.get('_i18n.dgettext') |
|---|
| [1077] | 160 | assert hasattr(dgettext, '__call__'), \ |
|---|
| 161 | 'No domain gettext function passed' |
|---|
| [1072] | 162 | gettext = lambda msg: dgettext(ctxt.get('_i18n.domain'), msg) |
|---|
| [954] | 163 | |
|---|
| [1072] | 164 | def _generate(): |
|---|
| 165 | msgbuf = MessageBuffer(self) |
|---|
| 166 | previous = stream.next() |
|---|
| 167 | if previous[0] is START: |
|---|
| 168 | yield previous |
|---|
| 169 | else: |
|---|
| 170 | msgbuf.append(*previous) |
|---|
| 171 | previous = stream.next() |
|---|
| 172 | for kind, data, pos in stream: |
|---|
| 173 | msgbuf.append(*previous) |
|---|
| 174 | previous = kind, data, pos |
|---|
| 175 | if previous[0] is not END: |
|---|
| 176 | msgbuf.append(*previous) |
|---|
| 177 | previous = None |
|---|
| 178 | for event in msgbuf.translate(gettext(msgbuf.format())): |
|---|
| 179 | yield event |
|---|
| 180 | if previous: |
|---|
| 181 | yield previous |
|---|
| 182 | |
|---|
| 183 | return _apply_directives(_generate(), directives, ctxt, vars) |
|---|
| 184 | |
|---|
| [1115] | 185 | def extract(self, translator, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| 186 | search_text=True, comment_stack=None): |
|---|
| [1072] | 187 | msgbuf = MessageBuffer(self) |
|---|
| [1115] | 188 | strip = False |
|---|
| [1072] | 189 | |
|---|
| [954] | 190 | stream = iter(stream) |
|---|
| [1072] | 191 | previous = stream.next() |
|---|
| 192 | if previous[0] is START: |
|---|
| [1115] | 193 | for message in translator._extract_attrs(previous, |
|---|
| 194 | gettext_functions, |
|---|
| 195 | search_text=search_text): |
|---|
| 196 | yield message |
|---|
| [1072] | 197 | previous = stream.next() |
|---|
| [1115] | 198 | strip = True |
|---|
| [1072] | 199 | for event in stream: |
|---|
| [1115] | 200 | if event[0] is START: |
|---|
| 201 | for message in translator._extract_attrs(event, |
|---|
| 202 | gettext_functions, |
|---|
| 203 | search_text=search_text): |
|---|
| 204 | yield message |
|---|
| [1072] | 205 | msgbuf.append(*previous) |
|---|
| 206 | previous = event |
|---|
| [1115] | 207 | if not strip: |
|---|
| 208 | msgbuf.append(*previous) |
|---|
| [1072] | 209 | |
|---|
| [1115] | 210 | yield self.lineno, None, msgbuf.format(), comment_stack[-1:] |
|---|
| [1072] | 211 | |
|---|
| 212 | |
|---|
| 213 | class ChooseBranchDirective(I18NDirective): |
|---|
| 214 | __slots__ = ['params'] |
|---|
| [1115] | 215 | |
|---|
| [1072] | 216 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| 217 | self.params = ctxt.get('_i18n.choose.params', [])[:] |
|---|
| 218 | msgbuf = MessageBuffer(self) |
|---|
| [1118] | 219 | stream = _apply_directives(stream, directives, ctxt, vars) |
|---|
| [1072] | 220 | |
|---|
| [954] | 221 | previous = stream.next() |
|---|
| [1094] | 222 | if previous[0] is START: |
|---|
| 223 | yield previous |
|---|
| 224 | else: |
|---|
| 225 | msgbuf.append(*previous) |
|---|
| [1118] | 226 | |
|---|
| [1094] | 227 | try: |
|---|
| 228 | previous = stream.next() |
|---|
| 229 | except StopIteration: |
|---|
| 230 | # For example <i18n:singular> or <i18n:plural> directives |
|---|
| 231 | yield MSGBUF, (), -1 # the place holder for msgbuf output |
|---|
| [1118] | 232 | ctxt['_i18n.choose.%s' % self.tagname] = msgbuf |
|---|
| [1094] | 233 | return |
|---|
| [1118] | 234 | |
|---|
| 235 | for event in stream: |
|---|
| [1072] | 236 | msgbuf.append(*previous) |
|---|
| [1118] | 237 | previous = event |
|---|
| [1072] | 238 | yield MSGBUF, (), -1 # the place holder for msgbuf output |
|---|
| [1094] | 239 | |
|---|
| 240 | if previous[0] is END: |
|---|
| 241 | yield previous # the outer end tag |
|---|
| 242 | else: |
|---|
| 243 | msgbuf.append(*previous) |
|---|
| [1118] | 244 | ctxt['_i18n.choose.%s' % self.tagname] = msgbuf |
|---|
| [1072] | 245 | |
|---|
| [1115] | 246 | def extract(self, translator, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| 247 | search_text=True, comment_stack=None, msgbuf=None): |
|---|
| [1072] | 248 | stream = iter(stream) |
|---|
| 249 | previous = stream.next() |
|---|
| [1115] | 250 | |
|---|
| [1072] | 251 | if previous[0] is START: |
|---|
| [1115] | 252 | # skip the enclosing element |
|---|
| 253 | for message in translator._extract_attrs(previous, |
|---|
| 254 | gettext_functions, |
|---|
| 255 | search_text=search_text): |
|---|
| 256 | yield message |
|---|
| [1072] | 257 | previous = stream.next() |
|---|
| [1115] | 258 | |
|---|
| [954] | 259 | for event in stream: |
|---|
| [1115] | 260 | if previous[0] is START: |
|---|
| 261 | for message in translator._extract_attrs(previous, |
|---|
| 262 | gettext_functions, |
|---|
| 263 | search_text=search_text): |
|---|
| 264 | yield message |
|---|
| [954] | 265 | msgbuf.append(*previous) |
|---|
| 266 | previous = event |
|---|
| [1115] | 267 | |
|---|
| [1072] | 268 | if previous[0] is not END: |
|---|
| 269 | msgbuf.append(*previous) |
|---|
| [954] | 270 | |
|---|
| [1072] | 271 | |
|---|
| 272 | class SingularDirective(ChooseBranchDirective): |
|---|
| 273 | """Implementation of the ``i18n:singular`` directive to be used with the |
|---|
| 274 | ``i18n:choose`` directive.""" |
|---|
| 275 | |
|---|
| 276 | |
|---|
| 277 | class PluralDirective(ChooseBranchDirective): |
|---|
| 278 | """Implementation of the ``i18n:plural`` directive to be used with the |
|---|
| 279 | ``i18n:choose`` directive.""" |
|---|
| 280 | |
|---|
| 281 | |
|---|
| 282 | class ChooseDirective(ExtractableI18NDirective): |
|---|
| 283 | """Implementation of the ``i18n:choose`` directive which provides plural |
|---|
| 284 | internationalisation of strings. |
|---|
| 285 | |
|---|
| 286 | This directive requires at least one parameter, the one which evaluates to |
|---|
| 287 | an integer which will allow to choose the plural/singular form. If you also |
|---|
| 288 | have expressions inside the singular and plural version of the string you |
|---|
| 289 | also need to pass a name for those parameters. Consider the following |
|---|
| 290 | examples: |
|---|
| 291 | |
|---|
| [1158] | 292 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| [1072] | 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 | |
|---|
| [1158] | 304 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| [1072] | 305 | ... <div i18n:choose="num; num"> |
|---|
| 306 | ... <p i18n:singular="">There is $num coin</p> |
|---|
| 307 | ... <p i18n:plural="">There are $num coins</p> |
|---|
| 308 | ... </div> |
|---|
| 309 | ... </html>''') |
|---|
| 310 | >>> translator.setup(tmpl) |
|---|
| [1076] | 311 | >>> print(tmpl.generate(num=1).render()) |
|---|
| [1072] | 312 | <html> |
|---|
| 313 | <div> |
|---|
| 314 | <p>There is 1 coin</p> |
|---|
| 315 | </div> |
|---|
| 316 | </html> |
|---|
| [1076] | 317 | >>> print(tmpl.generate(num=2).render()) |
|---|
| [1072] | 318 | <html> |
|---|
| 319 | <div> |
|---|
| 320 | <p>There are 2 coins</p> |
|---|
| 321 | </div> |
|---|
| 322 | </html> |
|---|
| 323 | |
|---|
| [1111] | 324 | When used as a element and not as an attribute: |
|---|
| [1072] | 325 | |
|---|
| [1158] | 326 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| [1072] | 327 | ... <i18n:choose numeral="num" params="num"> |
|---|
| 328 | ... <p i18n:singular="">There is $num coin</p> |
|---|
| 329 | ... <p i18n:plural="">There are $num coins</p> |
|---|
| 330 | ... </i18n:choose> |
|---|
| 331 | ... </html>''') |
|---|
| 332 | >>> translator.setup(tmpl) |
|---|
| 333 | >>> list(translator.extract(tmpl.stream)) #doctest: +NORMALIZE_WHITESPACE |
|---|
| 334 | [(2, 'ngettext', (u'There is %(num)s coin', |
|---|
| 335 | u'There are %(num)s coins'), [])] |
|---|
| 336 | """ |
|---|
| [1115] | 337 | __slots__ = ['numeral', 'params', 'lineno'] |
|---|
| [1072] | 338 | |
|---|
| [1115] | 339 | def __init__(self, value, template=None, namespaces=None, lineno=-1, |
|---|
| 340 | offset=-1): |
|---|
| [1072] | 341 | Directive.__init__(self, None, template, namespaces, lineno, offset) |
|---|
| 342 | params = [v.strip() for v in value.split(';')] |
|---|
| 343 | self.numeral = self._parse_expr(params.pop(0), template, lineno, offset) |
|---|
| 344 | self.params = params and [name.strip() for name in |
|---|
| 345 | params[0].split(',') if name] or [] |
|---|
| [1115] | 346 | self.lineno = lineno |
|---|
| [1072] | 347 | |
|---|
| 348 | @classmethod |
|---|
| 349 | def attach(cls, template, stream, value, namespaces, pos): |
|---|
| 350 | if type(value) is dict: |
|---|
| 351 | numeral = value.get('numeral', '').strip() |
|---|
| 352 | assert numeral is not '', "at least pass the numeral param" |
|---|
| 353 | params = [v.strip() for v in value.get('params', '').split(',')] |
|---|
| 354 | value = '%s; ' % numeral + ', '.join(params) |
|---|
| 355 | return super(ChooseDirective, cls).attach(template, stream, value, |
|---|
| 356 | namespaces, pos) |
|---|
| 357 | |
|---|
| 358 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| 359 | ctxt.push({'_i18n.choose.params': self.params, |
|---|
| [1118] | 360 | '_i18n.choose.singular': None, |
|---|
| 361 | '_i18n.choose.plural': None}) |
|---|
| [1072] | 362 | |
|---|
| [1118] | 363 | ngettext = ctxt.get('_i18n.ngettext') |
|---|
| 364 | assert hasattr(ngettext, '__call__'), 'No ngettext function available' |
|---|
| 365 | dngettext = ctxt.get('_i18n.dngettext') |
|---|
| 366 | if not dngettext: |
|---|
| 367 | dngettext = lambda d, s, p, n: ngettext(s, p, n) |
|---|
| 368 | |
|---|
| [1072] | 369 | new_stream = [] |
|---|
| 370 | singular_stream = None |
|---|
| 371 | singular_msgbuf = None |
|---|
| 372 | plural_stream = None |
|---|
| 373 | plural_msgbuf = None |
|---|
| 374 | |
|---|
| [1118] | 375 | numeral = self.numeral.evaluate(ctxt) |
|---|
| 376 | is_plural = self._is_plural(numeral, ngettext) |
|---|
| [1115] | 377 | |
|---|
| [1118] | 378 | for event in stream: |
|---|
| 379 | if event[0] is SUB and any(isinstance(d, ChooseBranchDirective) |
|---|
| 380 | for d in event[1][0]): |
|---|
| 381 | subdirectives, substream = event[1] |
|---|
| 382 | |
|---|
| 383 | if isinstance(subdirectives[0], SingularDirective): |
|---|
| [1072] | 384 | singular_stream = list(_apply_directives(substream, |
|---|
| 385 | subdirectives, |
|---|
| 386 | ctxt, vars)) |
|---|
| [1118] | 387 | new_stream.append((MSGBUF, None, (None, -1, -1))) |
|---|
| 388 | |
|---|
| 389 | elif isinstance(subdirectives[0], PluralDirective): |
|---|
| 390 | if is_plural: |
|---|
| 391 | plural_stream = list(_apply_directives(substream, |
|---|
| 392 | subdirectives, |
|---|
| 393 | ctxt, vars)) |
|---|
| 394 | |
|---|
| [1072] | 395 | else: |
|---|
| [1118] | 396 | new_stream.append(event) |
|---|
| [1072] | 397 | |
|---|
| 398 | if ctxt.get('_i18n.domain'): |
|---|
| 399 | ngettext = lambda s, p, n: dngettext(ctxt.get('_i18n.domain'), |
|---|
| 400 | s, p, n) |
|---|
| 401 | |
|---|
| [1118] | 402 | singular_msgbuf = ctxt.get('_i18n.choose.singular') |
|---|
| 403 | if is_plural: |
|---|
| 404 | plural_msgbuf = ctxt.get('_i18n.choose.plural') |
|---|
| 405 | msgbuf, choice = plural_msgbuf, plural_stream |
|---|
| [1096] | 406 | else: |
|---|
| [1118] | 407 | msgbuf, choice = singular_msgbuf, singular_stream |
|---|
| 408 | plural_msgbuf = MessageBuffer(self) |
|---|
| [1096] | 409 | |
|---|
| [1072] | 410 | for kind, data, pos in new_stream: |
|---|
| 411 | if kind is MSGBUF: |
|---|
| [1118] | 412 | for event in choice: |
|---|
| 413 | if event[0] is MSGBUF: |
|---|
| [1072] | 414 | translation = ngettext(singular_msgbuf.format(), |
|---|
| 415 | plural_msgbuf.format(), |
|---|
| [1118] | 416 | numeral) |
|---|
| 417 | for subevent in msgbuf.translate(translation): |
|---|
| 418 | yield subevent |
|---|
| [1072] | 419 | else: |
|---|
| [1118] | 420 | yield event |
|---|
| [1072] | 421 | else: |
|---|
| 422 | yield kind, data, pos |
|---|
| 423 | |
|---|
| 424 | ctxt.pop() |
|---|
| 425 | |
|---|
| [1115] | 426 | def extract(self, translator, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| 427 | search_text=True, comment_stack=None): |
|---|
| 428 | strip = False |
|---|
| [1072] | 429 | stream = iter(stream) |
|---|
| 430 | previous = stream.next() |
|---|
| 431 | |
|---|
| [1115] | 432 | if previous[0] is START: |
|---|
| 433 | # skip the enclosing element |
|---|
| 434 | for message in translator._extract_attrs(previous, |
|---|
| 435 | gettext_functions, |
|---|
| 436 | search_text=search_text): |
|---|
| 437 | yield message |
|---|
| 438 | previous = stream.next() |
|---|
| 439 | strip = True |
|---|
| 440 | |
|---|
| [1072] | 441 | singular_msgbuf = MessageBuffer(self) |
|---|
| 442 | plural_msgbuf = MessageBuffer(self) |
|---|
| 443 | |
|---|
| [1115] | 444 | for event in stream: |
|---|
| 445 | if previous[0] is SUB: |
|---|
| 446 | directives, substream = previous[1] |
|---|
| 447 | for directive in directives: |
|---|
| 448 | if isinstance(directive, SingularDirective): |
|---|
| 449 | for message in directive.extract(translator, |
|---|
| 450 | substream, gettext_functions, search_text, |
|---|
| 451 | comment_stack, msgbuf=singular_msgbuf): |
|---|
| 452 | yield message |
|---|
| 453 | elif isinstance(directive, PluralDirective): |
|---|
| 454 | for message in directive.extract(translator, |
|---|
| 455 | substream, gettext_functions, search_text, |
|---|
| 456 | comment_stack, msgbuf=plural_msgbuf): |
|---|
| 457 | yield message |
|---|
| 458 | elif not isinstance(directive, StripDirective): |
|---|
| 459 | singular_msgbuf.append(*previous) |
|---|
| 460 | plural_msgbuf.append(*previous) |
|---|
| [1072] | 461 | else: |
|---|
| [1115] | 462 | if previous[0] is START: |
|---|
| 463 | for message in translator._extract_attrs(previous, |
|---|
| 464 | gettext_functions, |
|---|
| 465 | search_text): |
|---|
| 466 | yield message |
|---|
| 467 | singular_msgbuf.append(*previous) |
|---|
| 468 | plural_msgbuf.append(*previous) |
|---|
| 469 | previous = event |
|---|
| [1072] | 470 | |
|---|
| [1115] | 471 | if not strip: |
|---|
| 472 | singular_msgbuf.append(*previous) |
|---|
| 473 | plural_msgbuf.append(*previous) |
|---|
| 474 | |
|---|
| 475 | yield self.lineno, 'ngettext', \ |
|---|
| [1072] | 476 | (singular_msgbuf.format(), plural_msgbuf.format()), \ |
|---|
| 477 | comment_stack[-1:] |
|---|
| 478 | |
|---|
| [1118] | 479 | def _is_plural(self, numeral, ngettext): |
|---|
| 480 | # XXX: should we test which form was chosen like this!?!?!? |
|---|
| 481 | # There should be no match in any catalogue for these singular and |
|---|
| 482 | # plural test strings |
|---|
| 483 | singular = u'O\x85\xbe\xa9\xa8az\xc3?\xe6\xa1\x02n\x84\x93' |
|---|
| 484 | plural = u'\xcc\xfb+\xd3Pn\x9d\tT\xec\x1d\xda\x1a\x88\x00' |
|---|
| 485 | return ngettext(singular, plural, numeral) == plural |
|---|
| [1072] | 486 | |
|---|
| [1118] | 487 | |
|---|
| [1072] | 488 | class DomainDirective(I18NDirective): |
|---|
| 489 | """Implementation of the ``i18n:domain`` directive which allows choosing |
|---|
| 490 | another i18n domain(catalog) to translate from. |
|---|
| 491 | |
|---|
| 492 | >>> from genshi.filters.tests.i18n import DummyTranslations |
|---|
| [1158] | 493 | >>> tmpl = MarkupTemplate('''<html xmlns:i18n="http://genshi.edgewall.org/i18n"> |
|---|
| [1072] | 494 | ... <p i18n:msg="">Bar</p> |
|---|
| 495 | ... <div i18n:domain="foo"> |
|---|
| 496 | ... <p i18n:msg="">FooBar</p> |
|---|
| 497 | ... <p>Bar</p> |
|---|
| 498 | ... <p i18n:domain="bar" i18n:msg="">Bar</p> |
|---|
| 499 | ... <p i18n:domain="">Bar</p> |
|---|
| 500 | ... </div> |
|---|
| 501 | ... <p>Bar</p> |
|---|
| 502 | ... </html>''') |
|---|
| 503 | |
|---|
| 504 | >>> translations = DummyTranslations({'Bar': 'Voh'}) |
|---|
| 505 | >>> translations.add_domain('foo', {'FooBar': 'BarFoo', 'Bar': 'foo_Bar'}) |
|---|
| 506 | >>> translations.add_domain('bar', {'Bar': 'bar_Bar'}) |
|---|
| 507 | >>> translator = Translator(translations) |
|---|
| 508 | >>> translator.setup(tmpl) |
|---|
| 509 | |
|---|
| [1076] | 510 | >>> print(tmpl.generate().render()) |
|---|
| [1072] | 511 | <html> |
|---|
| 512 | <p>Voh</p> |
|---|
| 513 | <div> |
|---|
| 514 | <p>BarFoo</p> |
|---|
| 515 | <p>foo_Bar</p> |
|---|
| 516 | <p>bar_Bar</p> |
|---|
| 517 | <p>Voh</p> |
|---|
| 518 | </div> |
|---|
| 519 | <p>Voh</p> |
|---|
| 520 | </html> |
|---|
| 521 | """ |
|---|
| 522 | __slots__ = ['domain'] |
|---|
| 523 | |
|---|
| [1115] | 524 | def __init__(self, value, template=None, namespaces=None, lineno=-1, |
|---|
| 525 | offset=-1): |
|---|
| [1072] | 526 | Directive.__init__(self, None, template, namespaces, lineno, offset) |
|---|
| [1118] | 527 | self.domain = value and value.strip() or '__DEFAULT__' |
|---|
| [1072] | 528 | |
|---|
| 529 | @classmethod |
|---|
| 530 | def attach(cls, template, stream, value, namespaces, pos): |
|---|
| 531 | if type(value) is dict: |
|---|
| 532 | value = value.get('name') |
|---|
| 533 | return super(DomainDirective, cls).attach(template, stream, value, |
|---|
| 534 | namespaces, pos) |
|---|
| 535 | |
|---|
| 536 | def __call__(self, stream, directives, ctxt, **vars): |
|---|
| 537 | ctxt.push({'_i18n.domain': self.domain}) |
|---|
| 538 | for event in _apply_directives(stream, directives, ctxt, vars): |
|---|
| [954] | 539 | yield event |
|---|
| [1072] | 540 | ctxt.pop() |
|---|
| [954] | 541 | |
|---|
| 542 | |
|---|
| 543 | class Translator(DirectiveFactory): |
|---|
| [538] | 544 | """Can extract and translate localizable strings from markup streams and |
|---|
| [544] | 545 | templates. |
|---|
| [538] | 546 | |
|---|
| [1072] | 547 | For example, assume the following template: |
|---|
| [538] | 548 | |
|---|
| 549 | >>> tmpl = MarkupTemplate('''<html xmlns:py="http://genshi.edgewall.org/"> |
|---|
| 550 | ... <head> |
|---|
| 551 | ... <title>Example</title> |
|---|
| 552 | ... </head> |
|---|
| 553 | ... <body> |
|---|
| 554 | ... <h1>Example</h1> |
|---|
| 555 | ... <p>${_("Hello, %(name)s") % dict(name=username)}</p> |
|---|
| 556 | ... </body> |
|---|
| 557 | ... </html>''', filename='example.html') |
|---|
| 558 | |
|---|
| 559 | For demonstration, we define a dummy ``gettext``-style function with a |
|---|
| 560 | hard-coded translation table, and pass that to the `Translator` initializer: |
|---|
| 561 | |
|---|
| 562 | >>> def pseudo_gettext(string): |
|---|
| 563 | ... return { |
|---|
| 564 | ... 'Example': 'Beispiel', |
|---|
| 565 | ... 'Hello, %(name)s': 'Hallo, %(name)s' |
|---|
| 566 | ... }[string] |
|---|
| 567 | >>> translator = Translator(pseudo_gettext) |
|---|
| 568 | |
|---|
| 569 | Next, the translator needs to be prepended to any already defined filters |
|---|
| 570 | on the template: |
|---|
| 571 | |
|---|
| 572 | >>> tmpl.filters.insert(0, translator) |
|---|
| 573 | |
|---|
| 574 | When generating the template output, our hard-coded translations should be |
|---|
| 575 | applied as expected: |
|---|
| 576 | |
|---|
| [1076] | 577 | >>> print(tmpl.generate(username='Hans', _=pseudo_gettext)) |
|---|
| [538] | 578 | <html> |
|---|
| 579 | <head> |
|---|
| 580 | <title>Beispiel</title> |
|---|
| 581 | </head> |
|---|
| 582 | <body> |
|---|
| 583 | <h1>Beispiel</h1> |
|---|
| 584 | <p>Hallo, Hans</p> |
|---|
| 585 | </body> |
|---|
| 586 | </html> |
|---|
| [1072] | 587 | |
|---|
| [626] | 588 | Note that elements defining ``xml:lang`` attributes that do not contain |
|---|
| 589 | variable expressions are ignored by this filter. That can be used to |
|---|
| 590 | exclude specific parts of a template from being extracted and translated. |
|---|
| [538] | 591 | """ |
|---|
| 592 | |
|---|
| [954] | 593 | directives = [ |
|---|
| [1072] | 594 | ('domain', DomainDirective), |
|---|
| [954] | 595 | ('comment', CommentDirective), |
|---|
| [1072] | 596 | ('msg', MsgDirective), |
|---|
| 597 | ('choose', ChooseDirective), |
|---|
| 598 | ('singular', SingularDirective), |
|---|
| 599 | ('plural', PluralDirective) |
|---|
| [954] | 600 | ] |
|---|
| 601 | |
|---|
| [544] | 602 | IGNORE_TAGS = frozenset([ |
|---|
| 603 | QName('script'), QName('http://www.w3.org/1999/xhtml}script'), |
|---|
| 604 | QName('style'), QName('http://www.w3.org/1999/xhtml}style') |
|---|
| 605 | ]) |
|---|
| [1072] | 606 | INCLUDE_ATTRS = frozenset([ |
|---|
| 607 | 'abbr', 'alt', 'label', 'prompt', 'standby', 'summary', 'title' |
|---|
| 608 | ]) |
|---|
| [954] | 609 | NAMESPACE = I18N_NAMESPACE |
|---|
| [538] | 610 | |
|---|
| [932] | 611 | def __init__(self, translate=NullTranslations(), ignore_tags=IGNORE_TAGS, |
|---|
| [708] | 612 | include_attrs=INCLUDE_ATTRS, extract_text=True): |
|---|
| [538] | 613 | """Initialize the translator. |
|---|
| 614 | |
|---|
| 615 | :param translate: the translation function, for example ``gettext`` or |
|---|
| 616 | ``ugettext``. |
|---|
| 617 | :param ignore_tags: a set of tag names that should not be localized |
|---|
| 618 | :param include_attrs: a set of attribute names should be localized |
|---|
| [708] | 619 | :param extract_text: whether the content of text nodes should be |
|---|
| 620 | extracted, or only text in explicit ``gettext`` |
|---|
| 621 | function calls |
|---|
| [1072] | 622 | |
|---|
| [932] | 623 | :note: Changed in 0.6: the `translate` parameter can now be either |
|---|
| 624 | a ``gettext``-style function, or an object compatible with the |
|---|
| 625 | ``NullTransalations`` or ``GNUTranslations`` interface |
|---|
| [538] | 626 | """ |
|---|
| [544] | 627 | self.translate = translate |
|---|
| [538] | 628 | self.ignore_tags = ignore_tags |
|---|
| 629 | self.include_attrs = include_attrs |
|---|
| [708] | 630 | self.extract_text = extract_text |
|---|
| [538] | 631 | |
|---|
| [1118] | 632 | def __call__(self, stream, ctxt=None, translate_text=True, |
|---|
| 633 | translate_attrs=True): |
|---|
| [540] | 634 | """Translate any localizable strings in the given stream. |
|---|
| 635 | |
|---|
| 636 | This function shouldn't be called directly. Instead, an instance of |
|---|
| 637 | the `Translator` class should be registered as a filter with the |
|---|
| 638 | `Template` or the `TemplateLoader`, or applied as a regular stream |
|---|
| 639 | filter. If used as a template filter, it should be inserted in front of |
|---|
| 640 | all the default filters. |
|---|
| 641 | |
|---|
| 642 | :param stream: the markup event stream |
|---|
| 643 | :param ctxt: the template context (not used) |
|---|
| [1118] | 644 | :param translate_text: whether text nodes should be translated (used |
|---|
| 645 | internally) |
|---|
| 646 | :param translate_attrs: whether attribute values should be translated |
|---|
| 647 | (used internally) |
|---|
| [540] | 648 | :return: the localized stream |
|---|
| 649 | """ |
|---|
| [544] | 650 | ignore_tags = self.ignore_tags |
|---|
| 651 | include_attrs = self.include_attrs |
|---|
| [954] | 652 | skip = 0 |
|---|
| 653 | xml_lang = XML_NAMESPACE['lang'] |
|---|
| [1118] | 654 | if not self.extract_text: |
|---|
| 655 | translate_text = False |
|---|
| 656 | translate_attrs = False |
|---|
| [954] | 657 | |
|---|
| [932] | 658 | if type(self.translate) is FunctionType: |
|---|
| 659 | gettext = self.translate |
|---|
| [1072] | 660 | if ctxt: |
|---|
| 661 | ctxt['_i18n.gettext'] = gettext |
|---|
| [932] | 662 | else: |
|---|
| [1158] | 663 | if IS_PYTHON2: |
|---|
| 664 | gettext = self.translate.ugettext |
|---|
| 665 | ngettext = self.translate.ungettext |
|---|
| 666 | else: |
|---|
| 667 | gettext = self.translate.gettext |
|---|
| 668 | ngettext = self.translate.ngettext |
|---|
| [1072] | 669 | try: |
|---|
| [1158] | 670 | if IS_PYTHON2: |
|---|
| 671 | dgettext = self.translate.dugettext |
|---|
| 672 | dngettext = self.translate.dungettext |
|---|
| 673 | else: |
|---|
| 674 | dgettext = self.translate.dgettext |
|---|
| 675 | dngettext = self.translate.dngettext |
|---|
| [1072] | 676 | except AttributeError: |
|---|
| [1118] | 677 | dgettext = lambda _, y: gettext(y) |
|---|
| 678 | dngettext = lambda _, s, p, n: ngettext(s, p, n) |
|---|
| [1072] | 679 | if ctxt: |
|---|
| 680 | ctxt['_i18n.gettext'] = gettext |
|---|
| [1118] | 681 | ctxt['_i18n.ngettext'] = ngettext |
|---|
| [1072] | 682 | ctxt['_i18n.dgettext'] = dgettext |
|---|
| 683 | ctxt['_i18n.dngettext'] = dngettext |
|---|
| 684 | |
|---|
| 685 | if ctxt and ctxt.get('_i18n.domain'): |
|---|
| [1158] | 686 | # TODO: This can cause infinite recursion if dgettext is defined |
|---|
| 687 | # via the AttributeError case above! |
|---|
| [1072] | 688 | gettext = lambda msg: dgettext(ctxt.get('_i18n.domain'), msg) |
|---|
| 689 | |
|---|
| [538] | 690 | for kind, data, pos in stream: |
|---|
| 691 | |
|---|
| 692 | # skip chunks that should not be localized |
|---|
| 693 | if skip: |
|---|
| 694 | if kind is START: |
|---|
| [626] | 695 | skip += 1 |
|---|
| [538] | 696 | elif kind is END: |
|---|
| [626] | 697 | skip -= 1 |
|---|
| [538] | 698 | yield kind, data, pos |
|---|
| 699 | continue |
|---|
| 700 | |
|---|
| 701 | # handle different events that can be localized |
|---|
| 702 | if kind is START: |
|---|
| 703 | tag, attrs = data |
|---|
| [626] | 704 | if tag in self.ignore_tags or \ |
|---|
| 705 | isinstance(attrs.get(xml_lang), basestring): |
|---|
| [538] | 706 | skip += 1 |
|---|
| 707 | yield kind, data, pos |
|---|
| 708 | continue |
|---|
| 709 | |
|---|
| [595] | 710 | new_attrs = [] |
|---|
| [538] | 711 | changed = False |
|---|
| [1072] | 712 | |
|---|
| [538] | 713 | for name, value in attrs: |
|---|
| [583] | 714 | newval = value |
|---|
| [1118] | 715 | if isinstance(value, basestring): |
|---|
| 716 | if translate_attrs and name in include_attrs: |
|---|
| [932] | 717 | newval = gettext(value) |
|---|
| [583] | 718 | else: |
|---|
| [1072] | 719 | newval = list( |
|---|
| [1118] | 720 | self(_ensure(value), ctxt, translate_text=False) |
|---|
| [583] | 721 | ) |
|---|
| 722 | if newval != value: |
|---|
| 723 | value = newval |
|---|
| 724 | changed = True |
|---|
| [538] | 725 | new_attrs.append((name, value)) |
|---|
| 726 | if changed: |
|---|
| [783] | 727 | attrs = Attrs(new_attrs) |
|---|
| [538] | 728 | |
|---|
| 729 | yield kind, (tag, attrs), pos |
|---|
| 730 | |
|---|
| [1118] | 731 | elif translate_text and kind is TEXT: |
|---|
| [954] | 732 | text = data.strip() |
|---|
| 733 | if text: |
|---|
| 734 | data = data.replace(text, unicode(gettext(text))) |
|---|
| 735 | yield kind, data, pos |
|---|
| [538] | 736 | |
|---|
| 737 | elif kind is SUB: |
|---|
| [954] | 738 | directives, substream = data |
|---|
| [1072] | 739 | current_domain = None |
|---|
| 740 | for idx, directive in enumerate(directives): |
|---|
| 741 | # Organize directives to make everything work |
|---|
| [1118] | 742 | # FIXME: There's got to be a better way to do this! |
|---|
| [1072] | 743 | if isinstance(directive, DomainDirective): |
|---|
| 744 | # Grab current domain and update context |
|---|
| 745 | current_domain = directive.domain |
|---|
| 746 | ctxt.push({'_i18n.domain': current_domain}) |
|---|
| 747 | # Put domain directive as the first one in order to |
|---|
| 748 | # update context before any other directives evaluation |
|---|
| 749 | directives.insert(0, directives.pop(idx)) |
|---|
| 750 | |
|---|
| 751 | # If this is an i18n directive, no need to translate text |
|---|
| [954] | 752 | # nodes here |
|---|
| [1079] | 753 | is_i18n_directive = any([ |
|---|
| 754 | isinstance(d, ExtractableI18NDirective) |
|---|
| 755 | for d in directives |
|---|
| 756 | ]) |
|---|
| [954] | 757 | substream = list(self(substream, ctxt, |
|---|
| [1118] | 758 | translate_text=not is_i18n_directive, |
|---|
| 759 | translate_attrs=translate_attrs)) |
|---|
| [954] | 760 | yield kind, (directives, substream), pos |
|---|
| [538] | 761 | |
|---|
| [1072] | 762 | if current_domain: |
|---|
| 763 | ctxt.pop() |
|---|
| [538] | 764 | else: |
|---|
| 765 | yield kind, data, pos |
|---|
| 766 | |
|---|
| [585] | 767 | def extract(self, stream, gettext_functions=GETTEXT_FUNCTIONS, |
|---|
| [1114] | 768 | search_text=True, comment_stack=None): |
|---|
| [538] | 769 | """Extract localizable strings from the given template stream. |
|---|
| 770 | |
|---|
| [544] | 771 | For every string found, this function yields a ``(lineno, function, |
|---|
| [929] | 772 | message, comments)`` tuple, where: |
|---|
| [538] | 773 | |
|---|
| [544] | 774 | * ``lineno`` is the number of the line on which the string was found, |
|---|
| 775 | * ``function`` is the name of the ``gettext`` function used (if the |
|---|
| 776 | string was extracted from embedded Python code), and |
|---|
| [569] | 777 | * ``message`` is the string itself (a ``unicode`` object, or a tuple |
|---|
| [929] | 778 | of ``unicode`` objects for functions with multiple string |
|---|
| 779 | arguments). |
|---|
| 780 | * ``comments`` is a list of comments related to the message, extracted |
|---|
| 781 | from ``i18n:comment`` attributes found in the markup |
|---|
| [538] | 782 | |
|---|
| 783 | >>> tmpl = MarkupTemplate('''<html xmlns:py="http://genshi.edgewall.org/"> |
|---|
| 784 | ... <head> |
|---|
| 785 | ... <title>Example</title> |
|---|
| 786 | ... </head> |
|---|
| 787 | ... <body> |
|---|
| 788 | ... <h1>Example</h1> |
|---|
| 789 | ... <p>${_("Hello, %(name)s") % dict(name=username)}</p> |
|---|
| [569] | 790 | ... <p>${ngettext("You have %d item", "You have %d items", num)}</p> |
|---|
| [538] | 791 | ... </body> |
|---|
| 792 | ... </html>''', filename='example.html') |
|---|
| [929] | 793 | >>> for line, func, msg, comments in Translator().extract(tmpl.stream): |
|---|
| [1076] | 794 | ... print('%d, %r, %r' % (line, func, msg)) |
|---|
| [544] | 795 | 3, None, u'Example' |
|---|
| 796 | 6, None, u'Example' |
|---|
| 797 | 7, '_', u'Hello, %(name)s' |
|---|
| [676] | 798 | 8, 'ngettext', (u'You have %d item', u'You have %d items', None) |
|---|
| [569] | 799 | |
|---|
| [544] | 800 | :param stream: the event stream to extract strings from; can be a |
|---|
| 801 | regular stream or a template stream |
|---|
| 802 | :param gettext_functions: a sequence of function names that should be |
|---|
| 803 | treated as gettext-style localization |
|---|
| 804 | functions |
|---|
| [585] | 805 | :param search_text: whether the content of text nodes should be |
|---|
| 806 | extracted (used internally) |
|---|
| [569] | 807 | |
|---|
| 808 | :note: Changed in 0.4.1: For a function with multiple string arguments |
|---|
| 809 | (such as ``ngettext``), a single item with a tuple of strings is |
|---|
| 810 | yielded, instead an item for each string argument. |
|---|
| [1072] | 811 | :note: Changed in 0.6: The returned tuples now include a fourth |
|---|
| 812 | element, which is a list of comments for the translator. |
|---|
| [538] | 813 | """ |
|---|
| [708] | 814 | if not self.extract_text: |
|---|
| 815 | search_text = False |
|---|
| [1072] | 816 | if comment_stack is None: |
|---|
| 817 | comment_stack = [] |
|---|
| [538] | 818 | skip = 0 |
|---|
| [1072] | 819 | |
|---|
| [626] | 820 | xml_lang = XML_NAMESPACE['lang'] |
|---|
| [538] | 821 | |
|---|
| 822 | for kind, data, pos in stream: |
|---|
| 823 | if skip: |
|---|
| 824 | if kind is START: |
|---|
| [626] | 825 | skip += 1 |
|---|
| [538] | 826 | if kind is END: |
|---|
| [626] | 827 | skip -= 1 |
|---|
| [538] | 828 | |
|---|
| [658] | 829 | if kind is START and not skip: |
|---|
| [538] | 830 | tag, attrs = data |
|---|
| [626] | 831 | if tag in self.ignore_tags or \ |
|---|
| 832 | isinstance(attrs.get(xml_lang), basestring): |
|---|
| [538] | 833 | skip += 1 |
|---|
| 834 | continue |
|---|
| 835 | |
|---|
| [1115] | 836 | for message in self._extract_attrs((kind, data, pos), |
|---|
| 837 | gettext_functions, |
|---|
| 838 | search_text=search_text): |
|---|
| 839 | yield message |
|---|
| [538] | 840 | |
|---|
| [658] | 841 | elif not skip and search_text and kind is TEXT: |
|---|
| [1114] | 842 | text = data.strip() |
|---|
| 843 | if text and [ch for ch in text if ch.isalpha()]: |
|---|
| 844 | yield pos[1], None, text, comment_stack[-1:] |
|---|
| [538] | 845 | |
|---|
| 846 | elif kind is EXPR or kind is EXEC: |
|---|
| [676] | 847 | for funcname, strings in extract_from_code(data, |
|---|
| [672] | 848 | gettext_functions): |
|---|
| [1072] | 849 | # XXX: Do we need to grab i18n:comment from comment_stack ??? |
|---|
| [929] | 850 | yield pos[1], funcname, strings, [] |
|---|
| [538] | 851 | |
|---|
| 852 | elif kind is SUB: |
|---|
| [1072] | 853 | directives, substream = data |
|---|
| 854 | in_comment = False |
|---|
| [634] | 855 | |
|---|
| [1072] | 856 | for idx, directive in enumerate(directives): |
|---|
| 857 | # Do a first loop to see if there's a comment directive |
|---|
| 858 | # If there is update context and pop it from directives |
|---|
| 859 | if isinstance(directive, CommentDirective): |
|---|
| 860 | in_comment = True |
|---|
| 861 | comment_stack.append(directive.comment) |
|---|
| 862 | if len(directives) == 1: |
|---|
| 863 | # in case we're in the presence of something like: |
|---|
| 864 | # <p i18n:comment="foo">Foo</p> |
|---|
| [1115] | 865 | for message in self.extract( |
|---|
| 866 | substream, gettext_functions, |
|---|
| 867 | search_text=search_text and not skip, |
|---|
| 868 | comment_stack=comment_stack): |
|---|
| 869 | yield message |
|---|
| [1072] | 870 | directives.pop(idx) |
|---|
| 871 | elif not isinstance(directive, I18NDirective): |
|---|
| 872 | # Remove all other non i18n directives from the process |
|---|
| 873 | directives.pop(idx) |
|---|
| [634] | 874 | |
|---|
| [1072] | 875 | if not directives and not in_comment: |
|---|
| 876 | # Extract content if there's no directives because |
|---|
| 877 | # strip was pop'ed and not because comment was pop'ed. |
|---|
| 878 | # Extraction in this case has been taken care of. |
|---|
| [1115] | 879 | for message in self.extract( |
|---|
| 880 | substream, gettext_functions, |
|---|
| 881 | search_text=search_text and not skip): |
|---|
| 882 | yield message |
|---|
| [1072] | 883 | |
|---|
| 884 | for directive in directives: |
|---|
| 885 | if isinstance(directive, ExtractableI18NDirective): |
|---|
| [1115] | 886 | for message in directive.extract(self, |
|---|
| 887 | substream, gettext_functions, |
|---|
| 888 | search_text=search_text and not skip, |
|---|
| 889 | comment_stack=comment_stack): |
|---|
| 890 | yield message |
|---|
| [1072] | 891 | else: |
|---|
| [1115] | 892 | for message in self.extract( |
|---|
| 893 | substream, gettext_functions, |
|---|
| 894 | search_text=search_text and not skip, |
|---|
| 895 | comment_stack=comment_stack): |
|---|
| 896 | yield message |
|---|
| [1072] | 897 | |
|---|
| 898 | if in_comment: |
|---|
| 899 | comment_stack.pop() |
|---|
| 900 | |
|---|
| 901 | def get_directive_index(self, dir_cls): |
|---|
| 902 | total = len(self._dir_order) |
|---|
| 903 | if dir_cls in self._dir_order: |
|---|
| 904 | return self._dir_order.index(dir_cls) - total |
|---|
| 905 | return total |
|---|
| 906 | |
|---|
| 907 | def setup(self, template): |
|---|
| 908 | """Convenience function to register the `Translator` filter and the |
|---|
| 909 | related directives with the given template. |
|---|
| 910 | |
|---|
| 911 | :param template: a `Template` instance |
|---|
| 912 | """ |
|---|
| 913 | template.filters.insert(0, self) |
|---|
| 914 | if hasattr(template, 'add_directives'): |
|---|
| 915 | template.add_directives(Translator.NAMESPACE, self) |
|---|
| 916 | |
|---|
| [1115] | 917 | def _extract_attrs(self, event, gettext_functions, search_text): |
|---|
| 918 | for name, value in event[1][1]: |
|---|
| 919 | if search_text and isinstance(value, basestring): |
|---|
| 920 | if name in self.include_attrs: |
|---|
| 921 | text = value.strip() |
|---|
| 922 | if text: |
|---|
| 923 | yield event[2][1], None, text, [] |
|---|
| 924 | else: |
|---|
| 925 | for message in self.extract(_ensure(value), gettext_functions, |
|---|
| 926 | search_text=False): |
|---|
| 927 | yield message |
|---|
| [1072] | 928 | |
|---|
| [1115] | 929 | |
|---|
| [671] | 930 | class MessageBuffer(object): |
|---|
| [862] | 931 | """Helper class for managing internationalized mixed content. |
|---|
| [690] | 932 | |
|---|
| 933 | :since: version 0.5 |
|---|
| 934 | """ |
|---|
| [671] | 935 | |
|---|
| [1072] | 936 | def __init__(self, directive=None): |
|---|
| [862] | 937 | """Initialize the message buffer. |
|---|
| 938 | |
|---|
| [1105] | 939 | :param directive: the directive owning the buffer |
|---|
| 940 | :type directive: I18NDirective |
|---|
| [862] | 941 | """ |
|---|
| [1072] | 942 | # params list needs to be copied so that directives can be evaluated |
|---|
| 943 | # more than once |
|---|
| 944 | self.orig_params = self.params = directive.params[:] |
|---|
| 945 | self.directive = directive |
|---|
| [862] | 946 | self.string = [] |
|---|
| [671] | 947 | self.events = {} |
|---|
| [901] | 948 | self.values = {} |
|---|
| [671] | 949 | self.depth = 1 |
|---|
| 950 | self.order = 1 |
|---|
| 951 | self.stack = [0] |
|---|
| [1072] | 952 | self.subdirectives = {} |
|---|
| [671] | 953 | |
|---|
| 954 | def append(self, kind, data, pos): |
|---|
| [862] | 955 | """Append a stream event to the buffer. |
|---|
| 956 | |
|---|
| 957 | :param kind: the stream event kind |
|---|
| 958 | :param data: the event data |
|---|
| 959 | :param pos: the position of the event in the source |
|---|
| 960 | """ |
|---|
| [1072] | 961 | if kind is SUB: |
|---|
| 962 | # The order needs to be +1 because a new START kind event will |
|---|
| 963 | # happen and we we need to wrap those events into our custom kind(s) |
|---|
| 964 | order = self.stack[-1] + 1 |
|---|
| 965 | subdirectives, substream = data |
|---|
| 966 | # Store the directives that should be applied after translation |
|---|
| 967 | self.subdirectives.setdefault(order, []).extend(subdirectives) |
|---|
| 968 | self.events.setdefault(order, []).append((SUB_START, None, pos)) |
|---|
| 969 | for skind, sdata, spos in substream: |
|---|
| 970 | self.append(skind, sdata, spos) |
|---|
| 971 | self.events.setdefault(order, []).append((SUB_END, None, pos)) |
|---|
| 972 | elif kind is TEXT: |
|---|
| 973 | if '[' in data or ']' in data: |
|---|
| 974 | # Quote [ and ] if it ain't us adding it, ie, if the user is |
|---|
| 975 | # using those chars in his templates, escape them |
|---|
| 976 | data = data.replace('[', '\[').replace(']', '\]') |
|---|
| [862] | 977 | self.string.append(data) |
|---|
| [1072] | 978 | self.events.setdefault(self.stack[-1], []).append((kind, data, pos)) |
|---|
| [901] | 979 | elif kind is EXPR: |
|---|
| [1072] | 980 | if self.params: |
|---|
| 981 | param = self.params.pop(0) |
|---|
| 982 | else: |
|---|
| 983 | params = ', '.join(['"%s"' % p for p in self.orig_params if p]) |
|---|
| 984 | if params: |
|---|
| 985 | params = "(%s)" % params |
|---|
| 986 | raise IndexError("%d parameters%s given to 'i18n:%s' but " |
|---|
| 987 | "%d or more expressions used in '%s', line %s" |
|---|
| 988 | % (len(self.orig_params), params, |
|---|
| 989 | self.directive.tagname, |
|---|
| [1115] | 990 | len(self.orig_params) + 1, |
|---|
| [1072] | 991 | os.path.basename(pos[0] or |
|---|
| [1095] | 992 | 'In-memory Template'), |
|---|
| [1072] | 993 | pos[1])) |
|---|
| [901] | 994 | self.string.append('%%(%s)s' % param) |
|---|
| [1072] | 995 | self.events.setdefault(self.stack[-1], []).append((kind, data, pos)) |
|---|
| [901] | 996 | self.values[param] = (kind, data, pos) |
|---|
| [671] | 997 | else: |
|---|
| [1072] | 998 | if kind is START: |
|---|
| [1077] | 999 | self.string.append('[%d:' % self.order) |
|---|
| [671] | 1000 | self.stack.append(self.order) |
|---|
| [1072] | 1001 | self.events.setdefault(self.stack[-1], |
|---|
| 1002 | []).append((kind, data, pos)) |
|---|
| [671] | 1003 | self.depth += 1 |
|---|
| 1004 | self.order += 1 |
|---|
| 1005 | elif kind is END: |
|---|
| 1006 | self.depth -= 1 |
|---|
| 1007 | if self.depth: |
|---|
| 1008 | self.events[self.stack[-1]].append((kind, data, pos)) |
|---|
| [1077] | 1009 | self.string.append(']') |
|---|
| [671] | 1010 | self.stack.pop() |
|---|
| 1011 | |
|---|
| 1012 | def format(self): |
|---|
| [862] | 1013 | """Return a message identifier representing the content in the |
|---|
| 1014 | buffer. |
|---|
| 1015 | """ |
|---|
| [1077] | 1016 | return ''.join(self.string).strip() |
|---|
| [671] | 1017 | |
|---|
| [901] | 1018 | def translate(self, string, regex=re.compile(r'%\((\w+)\)s')): |
|---|
| [862] | 1019 | """Interpolate the given message translation with the events in the |
|---|
| 1020 | buffer and return the translated stream. |
|---|
| 1021 | |
|---|
| 1022 | :param string: the translated message string |
|---|
| 1023 | """ |
|---|
| [1072] | 1024 | substream = None |
|---|
| [1118] | 1025 | |
|---|
| [1072] | 1026 | def yield_parts(string): |
|---|
| 1027 | for idx, part in enumerate(regex.split(string)): |
|---|
| 1028 | if idx % 2: |
|---|
| 1029 | yield self.values[part] |
|---|
| 1030 | elif part: |
|---|
| 1031 | yield (TEXT, |
|---|
| 1032 | part.replace('\[', '[').replace('\]', ']'), |
|---|
| 1033 | (None, -1, -1) |
|---|
| 1034 | ) |
|---|
| 1035 | |
|---|
| [671] | 1036 | parts = parse_msg(string) |
|---|
| [1072] | 1037 | parts_counter = {} |
|---|
| [671] | 1038 | for order, string in parts: |
|---|
| [1072] | 1039 | parts_counter.setdefault(order, []).append(None) |
|---|
| 1040 | |
|---|
| 1041 | while parts: |
|---|
| 1042 | order, string = parts.pop(0) |
|---|
| 1043 | if len(parts_counter[order]) == 1: |
|---|
| 1044 | events = self.events[order] |
|---|
| 1045 | else: |
|---|
| 1046 | events = [self.events[order].pop(0)] |
|---|
| 1047 | parts_counter[order].pop() |
|---|
| 1048 | |
|---|
| 1049 | for event in events: |
|---|
| 1050 | if event[0] is SUB_START: |
|---|
| 1051 | substream = [] |
|---|
| 1052 | elif event[0] is SUB_END: |
|---|
| 1053 | # Yield a substream which might have directives to be |
|---|
| 1054 | # applied to it (after translation events) |
|---|
| 1055 | yield SUB, (self.subdirectives[order], substream), event[2] |
|---|
| 1056 | substream = None |
|---|
| 1057 | elif event[0] is TEXT: |
|---|
| 1058 | if string: |
|---|
| 1059 | for part in yield_parts(string): |
|---|
| 1060 | if substream is not None: |
|---|
| 1061 | substream.append(part) |
|---|
| 1062 | else: |
|---|
| 1063 | yield part |
|---|
| 1064 | # String handled, reset it |
|---|
| 1065 | string = None |
|---|
| 1066 | elif event[0] is START: |
|---|
| 1067 | if substream is not None: |
|---|
| 1068 | substream.append(event) |
|---|
| 1069 | else: |
|---|
| 1070 | yield event |
|---|
| 1071 | if string: |
|---|
| 1072 | for part in yield_parts(string): |
|---|
| 1073 | if substream is not None: |
|---|
| 1074 | substream.append(part) |
|---|
| 1075 | else: |
|---|
| 1076 | yield part |
|---|
| 1077 | # String handled, reset it |
|---|
| 1078 | string = None |
|---|
| 1079 | elif event[0] is END: |
|---|
| 1080 | if string: |
|---|
| 1081 | for part in yield_parts(string): |
|---|
| 1082 | if substream is not None: |
|---|
| 1083 | substream.append(part) |
|---|
| 1084 | else: |
|---|
| 1085 | yield part |
|---|
| 1086 | # String handled, reset it |
|---|
| 1087 | string = None |
|---|
| 1088 | if substream is not None: |
|---|
| 1089 | substream.append(event) |
|---|
| 1090 | else: |
|---|
| 1091 | yield event |
|---|
| 1092 | elif event[0] is EXPR: |
|---|
| 1093 | # These are handled on the strings itself |
|---|
| 1094 | continue |
|---|
| [901] | 1095 | else: |
|---|
| [1072] | 1096 | if string: |
|---|
| 1097 | for part in yield_parts(string): |
|---|
| 1098 | if substream is not None: |
|---|
| 1099 | substream.append(part) |
|---|
| 1100 | else: |
|---|
| 1101 | yield part |
|---|
| 1102 | # String handled, reset it |
|---|
| 1103 | string = None |
|---|
| 1104 | if substream is not None: |
|---|
| 1105 | substream.append(event) |
|---|
| 1106 | else: |
|---|
| 1107 | yield event |
|---|
| [671] | 1108 | |
|---|
| [1115] | 1109 | |
|---|
| [1072] | 1110 | def parse_msg(string, regex=re.compile(r'(?:\[(\d+)\:)|(?<!\\)\]')): |
|---|
| [862] | 1111 | """Parse a translated message using Genshi mixed content message |
|---|
| 1112 | formatting. |
|---|
| [1072] | 1113 | |
|---|
| [862] | 1114 | >>> parse_msg("See [1:Help].") |
|---|
| 1115 | [(0, 'See '), (1, 'Help'), (0, '.')] |
|---|
| [1072] | 1116 | |
|---|
| [862] | 1117 | >>> parse_msg("See [1:our [2:Help] page] for details.") |
|---|
| 1118 | [(0, 'See '), (1, 'our '), (2, 'Help'), (1, ' page'), (0, ' for details.')] |
|---|
| [1072] | 1119 | |
|---|
| [862] | 1120 | >>> parse_msg("[2:Details] finden Sie in [1:Hilfe].") |
|---|
| 1121 | [(2, 'Details'), (0, ' finden Sie in '), (1, 'Hilfe'), (0, '.')] |
|---|
| [1072] | 1122 | |
|---|
| [862] | 1123 | >>> parse_msg("[1:] Bilder pro Seite anzeigen.") |
|---|
| 1124 | [(1, ''), (0, ' Bilder pro Seite anzeigen.')] |
|---|
| [1072] | 1125 | |
|---|
| [862] | 1126 | :param string: the translated message string |
|---|
| 1127 | :return: a list of ``(order, string)`` tuples |
|---|
| 1128 | :rtype: `list` |
|---|
| 1129 | """ |
|---|
| 1130 | parts = [] |
|---|
| 1131 | stack = [0] |
|---|
| 1132 | while True: |
|---|
| 1133 | mo = regex.search(string) |
|---|
| 1134 | if not mo: |
|---|
| 1135 | break |
|---|
| 1136 | |
|---|
| 1137 | if mo.start() or stack[-1]: |
|---|
| 1138 | parts.append((stack[-1], string[:mo.start()])) |
|---|
| 1139 | string = string[mo.end():] |
|---|
| 1140 | |
|---|
| 1141 | orderno = mo.group(1) |
|---|
| 1142 | if orderno is not None: |
|---|
| 1143 | stack.append(int(orderno)) |
|---|
| 1144 | else: |
|---|
| 1145 | stack.pop() |
|---|
| 1146 | if not stack: |
|---|
| 1147 | break |
|---|
| 1148 | |
|---|
| 1149 | if string: |
|---|
| 1150 | parts.append((stack[-1], string)) |
|---|
| 1151 | |
|---|
| 1152 | return parts |
|---|
| 1153 | |
|---|
| [902] | 1154 | |
|---|
| [672] | 1155 | def extract_from_code(code, gettext_functions): |
|---|
| 1156 | """Extract strings from Python bytecode. |
|---|
| 1157 | |
|---|
| 1158 | >>> from genshi.template.eval import Expression |
|---|
| 1159 | >>> expr = Expression('_("Hello")') |
|---|
| [1115] | 1160 | >>> list(extract_from_code(expr, GETTEXT_FUNCTIONS)) |
|---|
| [672] | 1161 | [('_', u'Hello')] |
|---|
| [1072] | 1162 | |
|---|
| [672] | 1163 | >>> expr = Expression('ngettext("You have %(num)s item", ' |
|---|
| 1164 | ... '"You have %(num)s items", num)') |
|---|
| [1115] | 1165 | >>> list(extract_from_code(expr, GETTEXT_FUNCTIONS)) |
|---|
| [676] | 1166 | [('ngettext', (u'You have %(num)s item', u'You have %(num)s items', None))] |
|---|
| [672] | 1167 | |
|---|
| [676] | 1168 | :param code: the `Code` object |
|---|
| 1169 | :type code: `genshi.template.eval.Code` |
|---|
| [672] | 1170 | :param gettext_functions: a sequence of function names |
|---|
| [690] | 1171 | :since: version 0.5 |
|---|
| [672] | 1172 | """ |
|---|
| [676] | 1173 | def _walk(node): |
|---|
| [988] | 1174 | if isinstance(node, _ast.Call) and isinstance(node.func, _ast.Name) \ |
|---|
| 1175 | and node.func.id in gettext_functions: |
|---|
| [676] | 1176 | strings = [] |
|---|
| [716] | 1177 | def _add(arg): |
|---|
| [1158] | 1178 | if isinstance(arg, _ast.Str) and isinstance(arg.s, unicode): |
|---|
| 1179 | strings.append(arg.s) |
|---|
| 1180 | elif isinstance(arg, _ast.Str): |
|---|
| [988] | 1181 | strings.append(unicode(arg.s, 'utf-8')) |
|---|
| 1182 | elif arg: |
|---|
| [676] | 1183 | strings.append(None) |
|---|
| [716] | 1184 | [_add(arg) for arg in node.args] |
|---|
| [988] | 1185 | _add(node.starargs) |
|---|
| 1186 | _add(node.kwargs) |
|---|
| [676] | 1187 | if len(strings) == 1: |
|---|
| 1188 | strings = strings[0] |
|---|
| [672] | 1189 | else: |
|---|
| [676] | 1190 | strings = tuple(strings) |
|---|
| [988] | 1191 | yield node.func.id, strings |
|---|
| 1192 | elif node._fields: |
|---|
| 1193 | children = [] |
|---|
| 1194 | for field in node._fields: |
|---|
| 1195 | child = getattr(node, field, None) |
|---|
| 1196 | if isinstance(child, list): |
|---|
| 1197 | for elem in child: |
|---|
| 1198 | children.append(elem) |
|---|
| 1199 | elif isinstance(child, _ast.AST): |
|---|
| 1200 | children.append(child) |
|---|
| 1201 | for child in children: |
|---|
| [676] | 1202 | for funcname, strings in _walk(child): |
|---|
| 1203 | yield funcname, strings |
|---|
| 1204 | return _walk(code.ast) |
|---|
| [672] | 1205 | |
|---|
| [902] | 1206 | |
|---|
| [634] | 1207 | def extract(fileobj, keywords, comment_tags, options): |
|---|
| 1208 | """Babel extraction method for Genshi templates. |
|---|
| 1209 | |
|---|
| 1210 | :param fileobj: the file-like object the messages should be extracted from |
|---|
| 1211 | :param keywords: a list of keywords (i.e. function names) that should be |
|---|
| 1212 | recognized as translation functions |
|---|
| 1213 | :param comment_tags: a list of translator tags to search for and include |
|---|
| 1214 | in the results |
|---|
| 1215 | :param options: a dictionary of additional options (optional) |
|---|
| 1216 | :return: an iterator over ``(lineno, funcname, message, comments)`` tuples |
|---|
| 1217 | :rtype: ``iterator`` |
|---|
| 1218 | """ |
|---|
| 1219 | template_class = options.get('template_class', MarkupTemplate) |
|---|
| 1220 | if isinstance(template_class, basestring): |
|---|
| 1221 | module, clsname = template_class.split(':', 1) |
|---|
| 1222 | template_class = getattr(__import__(module, {}, {}, [clsname]), clsname) |
|---|
| 1223 | encoding = options.get('encoding', None) |
|---|
| 1224 | |
|---|
| [710] | 1225 | extract_text = options.get('extract_text', True) |
|---|
| 1226 | if isinstance(extract_text, basestring): |
|---|
| 1227 | extract_text = extract_text.lower() in ('1', 'on', 'yes', 'true') |
|---|
| 1228 | |
|---|
| [634] | 1229 | ignore_tags = options.get('ignore_tags', Translator.IGNORE_TAGS) |
|---|
| 1230 | if isinstance(ignore_tags, basestring): |
|---|
| 1231 | ignore_tags = ignore_tags.split() |
|---|
| 1232 | ignore_tags = [QName(tag) for tag in ignore_tags] |
|---|
| [710] | 1233 | |
|---|
| [634] | 1234 | include_attrs = options.get('include_attrs', Translator.INCLUDE_ATTRS) |
|---|
| 1235 | if isinstance(include_attrs, basestring): |
|---|
| 1236 | include_attrs = include_attrs.split() |
|---|
| 1237 | include_attrs = [QName(attr) for attr in include_attrs] |
|---|
| 1238 | |
|---|
| 1239 | tmpl = template_class(fileobj, filename=getattr(fileobj, 'name', None), |
|---|
| 1240 | encoding=encoding) |
|---|
| [1105] | 1241 | tmpl.loader = None |
|---|
| [1072] | 1242 | |
|---|
| [710] | 1243 | translator = Translator(None, ignore_tags, include_attrs, extract_text) |
|---|
| [1072] | 1244 | if hasattr(tmpl, 'add_directives'): |
|---|
| 1245 | tmpl.add_directives(Translator.NAMESPACE, translator) |
|---|
| [929] | 1246 | for message in translator.extract(tmpl.stream, gettext_functions=keywords): |
|---|
| 1247 | yield message |
|---|