Python extract.extract函数代码示例

  def extract(self):
    catalog_obj = catalog.Catalog()
    path = os.path.join(self.root, 'messages.pot')
    template = self.pod.open_file(path, mode='w')
    extracted = []

    # Extracts messages from views.
    pod_files = self.pod.list_dir('/')
    for pod_path in pod_files:
      if os.path.splitext(pod_path)[-1] in _TRANSLATABLE_EXTENSIONS:
        content = self.pod.read_file(pod_path)
        import cStringIO
        fp = cStringIO.StringIO()
        fp.write(content)
        fp.seek(0)
        import tokenize
        try:
          messages = extract.extract('python', fp)
          for message in messages:
            lineno, string, comments, context = message
            catalog_obj.add(string, None, [(pod_path, lineno)], auto_comments=comments, context=context)
        except tokenize.TokenError:
          print 'Problem extracting: {}'.format(pod_path)
          raise

    # TODO(jeremydw): Extract messages from content.

    # Writes to PO template.
    pofile.write_po(template, catalog_obj, width=80, no_location=True, omit_header=True, sort_output=True, sort_by_file=True)
    logging.info('Extracted {} messages from {} files to: {}'.format(len(extracted), len(pod_files), template))
    template.close()
    return catalog_obj

def test_template_string_standard_usage():
    buf = BytesIO(b"msg1 = gettext(`Very template, wow`)")
    messages = list(
        extract.extract('javascript', buf, {"gettext": None}, [], {})
    )

    assert messages == [(1, 'Very template, wow', [], None, ())]

    def _extract(self, app):
        catalog = Catalog(domain="django", charset="utf8")
        files = {}
        for dirpath, dirnames, filenames in filtered_walk(app.path):
            for filename in filenames:
                filename = os.path.join(dirpath, filename)
                if ACCEPTABLE_FILENAMES_RE.match(filename):
                    rel_filename = filename[len(os.path.commonprefix((app.path, filename))) + 1:].replace(os.sep, "/")
                    files[rel_filename] = filename
        self.log.info("%s: %d translatable files found", app.label, len(files))
        extractors = self.get_extractors()
        for rel_filename, filename in sorted(files.items()):
            extractor_tup = extractors.get(os.path.splitext(filename)[1])
            if not extractor_tup:
                self.log.warning("Not sure how to extract messages from %s", filename)
                continue
            extractor, options = extractor_tup

            with open(filename, "rb") as fp:
                for (lineno, message, comments, context) in extract(extractor, fp, options=options):
                    catalog.add(message, locations=[(rel_filename, 0)], auto_comments=comments)
        if len(catalog):
            pot_path = self._get_pot_path(app)
            with open(pot_path, "w") as outf:
                pofile.write_po(outf, catalog, width=1000, omit_header=True, sort_output=True)
                self.log.info("%s: %d messages in %s", app.label, len(catalog), pot_path)
        return catalog

def test_dotted_keyword_extract():
    buf = BytesIO(b"msg1 = com.corporate.i18n.formatMessage('Insert coin to continue')")
    messages = list(
        extract.extract('javascript', buf, {"com.corporate.i18n.formatMessage": None}, [], {})
    )

    assert messages == [(1, 'Insert coin to continue', [], None, ())]

def test_template_string_tag_usage():
    buf = BytesIO(b"function() { if(foo) msg1 = i18n`Tag template, wow`; }")
    messages = list(
        extract.extract('javascript', buf, {"i18n": None}, [], {})
    )

    assert messages == [(1, 'Tag template, wow', [], None, ())]

def fake_extract_from_dir(filename, fileobj, method, options, keywords, comment_tags):
    """We use Babel's exctract_from_dir() to pull out our gettext
    strings.  In the tests, I don't have a directory of files, I have StringIO
    objects.  So, we fake the original function with this one."""
    for lineno, message, comments, context in extract(method, fileobj, keywords,
                                                      comment_tags, options):

        yield filename, lineno, message, comments, context

def extract_from_file(filename):
    for pattern, method in MAPPING:
        if filename.endswith(pattern):
            with open(filename, "rb") as in_file:
                for lineno, message, comments, context in extract(method, in_file, keywords=KEYWORDS):
                    lineno = 0  # Avoid messy diffs
                    yield filename, lineno, message, comments
            break

    def test_future(self):
        buf = BytesIO(br"""
# -*- coding: utf-8 -*-
from __future__ import unicode_literals
nbsp = _('\xa0')
""")
        messages = list(extract.extract('python', buf,
                                        extract.DEFAULT_KEYWORDS, [], {}))
        assert messages[0][1] == u'\xa0'

 def test_warn_if_empty_string_msgid_found_in_context_aware_extraction_method(self):
     buf = BytesIO(b"\nmsg = pgettext('ctxt', '')\n")
     stderr = sys.stderr
     sys.stderr = StringIO()
     try:
         messages = extract.extract('python', buf)
         self.assertEqual([], list(messages))
         assert 'warning: Empty msgid.' in sys.stderr.getvalue()
     finally:
         sys.stderr = stderr

    def test_simple_extract(self):
        buf = BytesIO(
            b"""\
msg1 = _('simple')
msg2 = gettext('simple')
msg3 = ngettext('s', 'p', 42)
        """
        )
        messages = list(extract.extract("javascript", buf, extract.DEFAULT_KEYWORDS, [], {}))

        self.assertEqual([(1, "simple", [], None), (2, "simple", [], None), (3, ("s", "p"), [], None)], messages)

    def test_simple_extract(self):
        buf = StringIO("""\
msg1 = _('simple')
msg2 = gettext('simple')
msg3 = ngettext('s', 'p', 42)
        """)
        messages = \
            list(extract.extract('javascript', buf, extract.DEFAULT_KEYWORDS,
                                 [], {}))

        self.assertEqual([(1, 'simple', []),
                          (2, 'simple', []),
                          (3, ('s', 'p'), [])], messages)

def test_simple_extract():
    buf = BytesIO(b"""\
msg1 = _('simple')
msg2 = gettext('simple')
msg3 = ngettext('s', 'p', 42)
    """)
    messages = \
        list(extract.extract('javascript', buf, extract.DEFAULT_KEYWORDS,
                             [], {}))

    assert messages == [(1, 'simple', [], None, ()),
                        (2, 'simple', [], None, ()),
                        (3, ('s', 'p'), [], None, ())]

    def test_empty_string_msgid(self):
        buf = BytesIO(b"""\
msg = _('')
""")
        stderr = sys.stderr
        sys.stderr = StringIO()
        try:
            messages = \
                list(extract.extract('python', buf, extract.DEFAULT_KEYWORDS,
                                     [], {}))
            self.assertEqual([], messages)
            assert 'warning: Empty msgid.' in sys.stderr.getvalue()
        finally:
            sys.stderr = stderr

    def test_different_signatures(self):
        buf = BytesIO(b"""
foo = _('foo', 'bar')
n = ngettext('hello', 'there', n=3)
n = ngettext(n=3, 'hello', 'there')
n = ngettext(n=3, *messages)
n = ngettext()
n = ngettext('foo')
""")
        messages = \
            list(extract.extract('python', buf, extract.DEFAULT_KEYWORDS, [],
                                 {}))
        self.assertEqual(len(messages), 2)
        self.assertEqual(u'foo', messages[0][1])
        self.assertEqual((u'hello', u'there'), messages[1][1])

 def babel_extract_terms(fname, path, root, extract_method="python", trans_type='code',
                            extra_comments=None, extract_keywords={'_': None}):
     module, fabsolutepath, _, display_path = verified_module_filepaths(fname, path, root)
     extra_comments = extra_comments or []
     if module:
         src_file = open(fabsolutepath, 'r')
         try:
             for lineno, message, comments in extract.extract(extract_method, src_file,
                                                              keywords=extract_keywords):
                 push_translation(module, trans_type, display_path, lineno,
                                  encode(message), comments + extra_comments)
         except Exception:
             _logger.exception("Failed to extract terms from %s", fabsolutepath)
         finally:
             src_file.close()