Python AddonIndexer.extract_document方法代碼示例

# 需要導入模塊: from olympia.addons.indexers import AddonIndexer [as 別名]
# 或者: from olympia.addons.indexers.AddonIndexer import extract_document [as 別名]

#.........這裏部分代碼省略.........
        """Make sure that the `index` setting is a true/false boolean.

        Old versions of ElasticSearch allowed 'no' and 'yes' strings,
        this changed with ElasticSearch 5.x.
        """
        doc_name = self.indexer.get_doctype_name()
        assert doc_name

        mapping_properties = self.indexer.get_mapping()[doc_name]['properties']

        assert all(
            isinstance(prop['index'], bool)
            for prop in mapping_properties.values()
            if 'index' in prop)

        # Make sure our version_mapping is setup correctly too.
        props = mapping_properties['current_version']['properties']

        assert all(
            isinstance(prop['index'], bool)
            for prop in props.values() if 'index' in prop)

        # As well as for current_version.files
        assert all(
            isinstance(prop['index'], bool)
            for prop in props['files']['properties'].values()
            if 'index' in prop)

    def _extract(self):
        qs = Addon.unfiltered.filter(id__in=[self.addon.pk])
        for t in self.transforms:
            qs = qs.transform(t)
        self.addon = list(qs)[0]
        return self.indexer.extract_document(self.addon)

    def test_extract_attributes(self):
        extracted = self._extract()

        # Like test_mapping() above, but for the extraction process:
        # Make sure the method does not return fields we did not expect to be
        # present, or omitted fields we want.
        assert set(extracted.keys()) == set(
            self.expected_fields(include_nullable=False))

        # Check base fields values. Other tests below check the dynamic ones.
        for field_name in self.simple_fields:
            assert extracted[field_name] == getattr(self.addon, field_name)

        assert extracted['app'] == [FIREFOX.id]
        assert extracted['boost'] == self.addon.average_daily_users ** .2 * 4
        assert extracted['category'] == [1, 22, 71]  # From fixture.
        assert extracted['current_version']
        assert extracted['listed_authors'] == [
            {'name': u'55021 التطب', 'id': 55021, 'username': '55021',
             'is_public': True}]
        assert extracted['platforms'] == [PLATFORM_ALL.id]
        assert extracted['ratings'] == {
            'average': self.addon.average_rating,
            'count': self.addon.total_ratings,
            'text_count': self.addon.text_ratings_count,
        }
        assert extracted['tags'] == []
        assert extracted['has_eula'] is True
        assert extracted['has_privacy_policy'] is True
        assert extracted['is_featured'] is False
        assert extracted['colors'] is None

# 需要導入模塊: from olympia.addons.indexers import AddonIndexer [as 別名]
# 或者: from olympia.addons.indexers.AddonIndexer import extract_document [as 別名]

#.........這裏部分代碼省略.........
            'app', 'appversion', 'boost', 'category', 'current_beta_version',
            'current_version', 'description', 'has_eula', 'has_privacy_policy',
            'has_theme_rereview', 'has_version', 'listed_authors', 'name',
            'name_sort', 'platforms', 'previews', 'public_stats', 'ratings',
            'summary', 'tags',
        ]

        # Fields that need to be present in the mapping, but might be skipped
        # for extraction because they can be null.
        nullable_fields = ['persona']

        # For each translated field that needs to be indexed, we store one
        # version for each language-specific analyzer we have.
        _indexed_translated_fields = ('name', 'description', 'summary')
        analyzer_fields = list(chain.from_iterable(
            [['%s_%s' % (field, analyzer) for analyzer in SEARCH_ANALYZER_MAP]
             for field in _indexed_translated_fields]))

        # It'd be annoying to hardcode `analyzer_fields`, so we generate it,
        # but to make sure the test is correct we still do a simple check of
        # the length to make sure we properly flattened the list.
        assert len(analyzer_fields) == (len(SEARCH_ANALYZER_MAP) *
                                        len(_indexed_translated_fields))

        # Each translated field that we want to return to the API.
        raw_translated_fields = [
            '%s_translations' % field for field in
            ['name', 'description', 'homepage', 'summary', 'support_email',
             'support_url']]

        # Return a list with the base fields and the dynamic ones added.
        fields = (cls.simple_fields + complex_fields + analyzer_fields +
                  raw_translated_fields)
        if include_nullable:
            fields += nullable_fields
        return fields

    def test_mapping(self):
        doc_name = self.indexer.get_doctype_name()
        assert doc_name

        mapping_properties = self.indexer.get_mapping()[doc_name]['properties']

        # Make sure the get_mapping() method does not return fields we did
        # not expect to be present, or omitted fields we want.
        assert set(mapping_properties.keys()) == set(self.expected_fields())

        # Make sure default_locale and translated fields are not indexed.
        assert mapping_properties['default_locale']['index'] == 'no'
        name_translations = mapping_properties['name_translations']
        assert name_translations['properties']['lang']['index'] == 'no'
        assert name_translations['properties']['string']['index'] == 'no'

        # Make sure nothing inside 'persona' is indexed, it's only there to be
        # returned back to the API directly.
        for field in mapping_properties['persona']['properties'].values():
            assert field['index'] == 'no'

        # Make sure current_version mapping is set.
        assert mapping_properties['current_version']['properties']
        version_mapping = mapping_properties['current_version']['properties']
        expected_version_keys = (
            'id', 'compatible_apps', 'files', 'reviewed', 'version')
        assert set(version_mapping.keys()) == set(expected_version_keys)

        # Make sure files mapping is set inside current_version.
        files_mapping = version_mapping['files']['properties']
        expected_file_keys = ('id', 'created', 'filename', 'hash', 'platform',
                              'size', 'status')
        assert set(files_mapping.keys()) == set(expected_file_keys)

    def _extract(self):
        qs = Addon.unfiltered.filter(id__in=[self.addon.pk]).no_cache()
        for t in self.transforms:
            qs = qs.transform(t)
        self.addon = list(qs)[0]
        return self.indexer.extract_document(self.addon)

    def test_extract_attributes(self):
        extracted = self._extract()

        # Like test_mapping() above, but for the extraction process:
        # Make sure the method does not return fields we did not expect to be
        # present, or omitted fields we want.
        assert set(extracted.keys()) == set(
            self.expected_fields(include_nullable=False))

        # Check base fields values. Other tests below check the dynamic ones.
        for field_name in self.simple_fields:
            assert extracted[field_name] == getattr(self.addon, field_name)

        assert extracted['app'] == [FIREFOX.id]
        assert extracted['appversion'] == {
            FIREFOX.id: {
                'min': 2000000200100L,
                'max': 4000000200100L,
                'max_human': '4.0',
                'min_human': '2.0',
            }
        }

# 需要導入模塊: from olympia.addons.indexers import AddonIndexer [as 別名]
# 或者: from olympia.addons.indexers.AddonIndexer import extract_document [as 別名]

#.........這裏部分代碼省略.........
        """Make sure that the `index` setting is a true/false boolean.

        Old versions of ElasticSearch allowed 'no' and 'yes' strings,
        this changed with ElasticSearch 5.x.
        """
        doc_name = self.indexer.get_doctype_name()
        assert doc_name

        mapping_properties = self.indexer.get_mapping()[doc_name]['properties']

        assert all(
            isinstance(prop['index'], bool)
            for prop in mapping_properties.values()
            if 'index' in prop)

        # Make sure our version_mapping is setup correctly too.
        props = mapping_properties['current_version']['properties']

        assert all(
            isinstance(prop['index'], bool)
            for prop in props.values() if 'index' in prop)

        # As well as for current_version.files
        assert all(
            isinstance(prop['index'], bool)
            for prop in props['files']['properties'].values()
            if 'index' in prop)

    def _extract(self):
        qs = Addon.unfiltered.filter(id__in=[self.addon.pk]).no_cache()
        for t in self.transforms:
            qs = qs.transform(t)
        self.addon = list(qs)[0]
        return self.indexer.extract_document(self.addon)

    def test_extract_attributes(self):
        extracted = self._extract()

        # Like test_mapping() above, but for the extraction process:
        # Make sure the method does not return fields we did not expect to be
        # present, or omitted fields we want.
        assert set(extracted.keys()) == set(
            self.expected_fields(include_nullable=False))

        # Check base fields values. Other tests below check the dynamic ones.
        for field_name in self.simple_fields:
            assert extracted[field_name] == getattr(self.addon, field_name)

        assert extracted['app'] == [FIREFOX.id]
        assert extracted['boost'] == self.addon.average_daily_users ** .2 * 4
        assert extracted['category'] == [1, 22, 71]  # From fixture.
        assert extracted['current_beta_version'] is None
        assert extracted['current_version']
        assert extracted['has_theme_rereview'] is None
        assert extracted['latest_unlisted_version'] is None
        assert extracted['listed_authors'] == [
            {'name': u'55021 التطب', 'id': 55021, 'username': '55021',
             'is_public': True}]
        assert extracted['platforms'] == [PLATFORM_ALL.id]
        assert extracted['ratings'] == {
            'average': self.addon.average_rating,
            'count': self.addon.total_ratings,
            'text_count': self.addon.text_ratings_count,
        }
        assert extracted['tags'] == []
        assert extracted['has_eula'] is True