def test_convert_to_pymorphy2(self, tmpdir):

        # import logging
        # from pymorphy2.opencorpora_dict.compile import logger
        # logger.setLevel(logging.DEBUG)
        # logger.addHandler(logging.StreamHandler())

        try:
            assert_can_create()
        except NotImplementedError as e:
            raise pytest.skip(e)

        # create a dictionary
        out_path = str(tmpdir.join('dicts'))
        options = {
            'min_paradigm_popularity': 0,
            'min_ending_freq': 0,
            'paradigm_prefixes': lang.ru.PARADIGM_PREFIXES,
        }
        convert_to_pymorphy2(self.XML_PATH,
                             out_path,
                             source_name='toy',
                             language_code='ru',
                             overwrite=True,
                             compile_options=options)

        # use it
        morph = pymorphy2.MorphAnalyzer(out_path)
        assert morph.tag('ёжиться') == [morph.TagClass('INFN,impf,intr')]

        # tag simplification should work
        assert morph.tag("ёж")[0] == morph.tag("ванька-встанька")[0]
    def test_convert_to_pymorphy2(self, tmpdir):

        # import logging
        # from pymorphy2.opencorpora_dict.compile import logger
        # logger.setLevel(logging.DEBUG)
        # logger.addHandler(logging.StreamHandler())

        try:
            assert_can_create()
        except NotImplementedError as e:
            raise pytest.skip(e)

        # create a dictionary
        out_path = str(tmpdir.join('dicts'))
        options = {
            'min_paradigm_popularity': 0,
            'min_ending_freq': 0,
            'paradigm_prefixes': lang.ru.PARADIGM_PREFIXES,
        }
        convert_to_pymorphy2(self.XML_PATH, out_path,
                             source_name='toy', language_code='ru',
                             overwrite=True, compile_options=options)

        # use it
        morph = pymorphy2.MorphAnalyzer(out_path)
        assert morph.tag('ёжиться') == [morph.TagClass('INFN,impf,intr')]

        # tag simplification should work
        assert morph.tag("ёж")[0] == morph.tag("ванька-встанька")[0]
Example #3
0
    def test_convert_to_pymorphy2(self, tmpdir):

        # import logging
        # from pymorphy2.opencorpora_dict.compile import logger
        # logger.setLevel(logging.DEBUG)
        # logger.addHandler(logging.StreamHandler())

        try:
            assert_can_create()
        except NotImplementedError as e:
            raise pytest.skip(e)

        # create a dictionary
        out_path = str(tmpdir.join('dicts'))
        options = {
            'min_paradigm_popularity': 0,
            'min_ending_freq': 0,
        }
        convert_to_pymorphy2(self.XML_PATH,
                             out_path,
                             overwrite=True,
                             prediction_options=options)

        # use it
        morph = pymorphy2.MorphAnalyzer(out_path,
                                        probability_estimator_cls=None)
        assert morph.tag('ёжиться') == [morph.TagClass('INFN,impf,intr')]
Example #4
0
    def test_convert_to_pymorphy2_with_allowed_link_types(self, tmpdir):

        try:
            assert_can_create()
        except NotImplementedError as e:
            raise pytest.skip(e)

        # create a dictionary
        out_path = str(tmpdir.join('dicts'))
        options = {
            'min_paradigm_popularity': 0,
            'min_ending_freq': 0,
            'paradigm_prefixes': lang.ru.PARADIGM_PREFIXES,
        }
        convert_to_pymorphy2(self.XML_PATH, out_path,
                             source_name='toy', language_code='ru',
                             overwrite=True, compile_options=options,
                             links_matching_path=self.ALLOWED_LINK_TYPES_XML_PATH)

        # use it
        morph = pymorphy2.MorphAnalyzer(out_path)
        assert morph.tag('ёжиться') == [morph.TagClass('INFN,impf,intr')]

        # tag simplification should work
        assert morph.tag("ёж")[0] == morph.tag("ванька-встанька")[0]

        # Init tags should be handled correctly
        assert 'Init' in morph.tag("Ц")[0]
        assert 'Init' not in morph.tag("ц")[0]

        # normalization tests
        assert morph.normal_forms('абсурднее') == ['абсурдный']
        assert morph.normal_forms('а') == ['а']
    def test_convert_to_pymorphy2(self, tmpdir):

        # import logging
        # from pymorphy2.opencorpora_dict.compile import logger
        # logger.setLevel(logging.DEBUG)
        # logger.addHandler(logging.StreamHandler())

        try:
            assert_can_create()
        except NotImplementedError as e:
            raise pytest.skip(e)

        # create a dictionary
        out_path = str(tmpdir.join('dicts'))
        options = {
            'min_paradigm_popularity': 0,
            'min_ending_freq': 0,
        }
        convert_to_pymorphy2(self.XML_PATH, out_path, overwrite=True,
                             prediction_options=options)

        # use it
        morph = pymorphy2.MorphAnalyzer(out_path)
        assert morph.tag('ёжиться') == [morph.TagClass('INFN,impf,intr')]