Python Solr.more_like_this方法代码示例

# 需要导入模块: from pysolr import Solr [as 别名]
# 或者: from pysolr.Solr import more_like_this [as 别名]

#.........这里部分代码省略.........
        if highlight is True:
            kwargs['hl'] = 'true'
            kwargs['hl.fragsize'] = '200'
        
        if getattr(settings, 'HAYSTACK_INCLUDE_SPELLING', False) is True:
            kwargs['spellcheck'] = 'true'
            kwargs['spellcheck.collate'] = 'true'
            kwargs['spellcheck.count'] = 1
        
        if facets is not None:
            kwargs['facet'] = 'on'
            kwargs['facet.field'] = facets
        
        if date_facets is not None:
            kwargs['facet'] = 'on'
            kwargs['facet.date'] = date_facets.keys()
            
            for key, value in date_facets.items():
                # Date-based facets in Solr kinda suck.
                kwargs["f.%s.facet.date.start" % key] = self.conn._from_python(value.get('start_date'))
                kwargs["f.%s.facet.date.end" % key] = self.conn._from_python(value.get('end_date'))
                kwargs["f.%s.facet.date.gap" % key] = value.get('gap')
        
        if query_facets is not None:
            kwargs['facet'] = 'on'
            kwargs['facet.query'] = ["%s:%s" % (field, value) for field, value in query_facets.items()]
        
        if narrow_queries is not None:
            kwargs['fq'] = list(narrow_queries)
        
        raw_results = self.conn.search(query_string, **kwargs)
        return self._process_results(raw_results, highlight=highlight)
    
    def more_like_this(self, model_instance, additional_query_string=None, start_offset=0, end_offset=None, **kwargs):
        index = self.site.get_index(model_instance.__class__)
        field_name = index.get_content_field()
        params = {
            'fl': '*,score',
        }
        
        if start_offset is not None:
            params['start'] = start_offset
        
        if end_offset is not None:
            params['rows'] = end_offset
        
        if additional_query_string:
            params['fq'] = additional_query_string
        
        raw_results = self.conn.more_like_this("id:%s" % self.get_identifier(model_instance), field_name, **params)
        return self._process_results(raw_results)
    
    def _process_results(self, raw_results, highlight=False):
        from haystack import site
        results = []
        hits = raw_results.hits
        facets = {}
        spelling_suggestion = None
        
        if hasattr(raw_results, 'facets'):
            facets = {
                'fields': raw_results.facets.get('facet_fields', {}),
                'dates': raw_results.facets.get('facet_dates', {}),
                'queries': raw_results.facets.get('facet_queries', {}),
            }
            

# 需要导入模块: from pysolr import Solr [as 别名]
# 或者: from pysolr.Solr import more_like_this [as 别名]

#.........这里部分代码省略.........

                for facet in stats[k]:
                    kwargs['f.%s.stats.facet' % k] = facet

        if within is not None:
            from haystack.utils.geo import generate_bounding_box

            kwargs.setdefault('fq', [])
            ((min_lat, min_lng), (max_lat, max_lng)) = generate_bounding_box(within['point_1'], within['point_2'])
            # Bounding boxes are min, min TO max, max. Solr's wiki was *NOT*
            # very clear on this.
            bbox = '%s:[%s,%s TO %s,%s]' % (within['field'], min_lat, min_lng, max_lat, max_lng)
            kwargs['fq'].append(bbox)

        if dwithin is not None:
            kwargs.setdefault('fq', [])
            lng, lat = dwithin['point'].get_coords()
            geofilt = '{!geofilt pt=%s,%s sfield=%s d=%s}' % (lat, lng, dwithin['field'], dwithin['distance'].km)
            kwargs['fq'].append(geofilt)

        # Check to see if the backend should try to include distances
        # (Solr 4.X+) in the results.
        if self.distance_available and distance_point:
            # In early testing, you can't just hand Solr 4.X a proper bounding box
            # & request distances. To enable native distance would take calculating
            # a center point & a radius off the user-provided box, which kinda
            # sucks. We'll avoid it for now, since Solr 4.x's release will be some
            # time yet.
            # kwargs['fl'] += ' _dist_:geodist()'
            pass

        return kwargs

    def more_like_this(self, model_instance, additional_query_string=None,
                       start_offset=0, end_offset=None, models=None,
                       limit_to_registered_models=None, result_class=None, **kwargs):
        from haystack import connections

        # Deferred models will have a different class ("RealClass_Deferred_fieldname")
        # which won't be in our registry:
        model_klass = model_instance._meta.concrete_model

        index = connections[self.connection_alias].get_unified_index().get_index(model_klass)
        field_name = index.get_content_field()
        params = {
            'fl': '*,score',
        }

        if start_offset is not None:
            params['start'] = start_offset

        if end_offset is not None:
            params['rows'] = end_offset

        narrow_queries = set()

        if limit_to_registered_models is None:
            limit_to_registered_models = getattr(settings, 'HAYSTACK_LIMIT_TO_REGISTERED_MODELS', True)

        if models and len(models):
            model_choices = sorted(get_model_ct(model) for model in models)
        elif limit_to_registered_models:
            # Using narrow queries, limit the results to only models handled
            # with the current routers.
            model_choices = self.build_models_list()
        else:

# 需要导入模块: from pysolr import Solr [as 别名]
# 或者: from pysolr.Solr import more_like_this [as 别名]

#.........这里部分代码省略.........
    def test_search(self):
        results = self.solr.search('doc')
        self.assertEqual(len(results), 3)

        results = self.solr.search('example')
        self.assertEqual(len(results), 2)

        results = self.solr.search('nothing')
        self.assertEqual(len(results), 0)

        # Advanced options.
        results = self.solr.search('doc', **{
            'debug': 'true',
            'hl': 'true',
            'hl.fragsize': 8,
            'facet': 'on',
            'facet.field': 'popularity',
            'spellcheck': 'true',
            'spellcheck.collate': 'true',
            'spellcheck.count': 1,
            # TODO: Can't get these working in my test setup.
            # 'group': 'true',
            # 'group.field': 'id',
        })
        self.assertEqual(len(results), 3)
        self.assertTrue('explain' in results.debug)
        self.assertEqual(results.highlighting, {u'doc_4': {}, u'doc_2': {}, u'doc_1': {}})
        self.assertEqual(results.spellcheck, {})
        self.assertEqual(results.facets['facet_fields']['popularity'], ['10', 2, '7', 1, '2', 0, '8', 0])
        self.assertTrue(results.qtime is not None)
        # TODO: Can't get these working in my test setup.
        # self.assertEqual(results.grouped, '')

    def test_more_like_this(self):
        results = self.solr.more_like_this('id:doc_1', 'text')
        self.assertEqual(len(results), 0)

    def test_suggest_terms(self):
        results = self.solr.suggest_terms('title', '')
        self.assertEqual(len(results), 1)
        self.assertEqual(results, {'title': [('doc', 3), ('another', 2), ('example', 2), ('1', 1), ('2', 1), ('boring', 1), ('rock', 1), ('thing', 1)]})

    def test__build_doc(self):
        doc = {
            'id': 'doc_1',
            'title': 'Example doc ☃ 1',
            'price': 12.59,
            'popularity': 10,
        }
        doc_xml = force_unicode(ET.tostring(self.solr._build_doc(doc), encoding='utf-8'))
        self.assertTrue('<field name="title">Example doc ☃ 1</field>' in doc_xml)
        self.assertTrue('<field name="id">doc_1</field>' in doc_xml)
        self.assertEqual(len(doc_xml), 152)

    def test_add(self):
        self.assertEqual(len(self.solr.search('doc')), 3)
        self.assertEqual(len(self.solr.search('example')), 2)

        self.solr.add([
            {
                'id': 'doc_6',
                'title': 'Newly added doc',
            },
            {
                'id': 'doc_7',
                'title': 'Another example doc',

# 需要导入模块: from pysolr import Solr [as 别名]
# 或者: from pysolr.Solr import more_like_this [as 别名]

#.........这里部分代码省略.........
        if date_facets is not None:
            kwargs['facet'] = 'on'
            kwargs['facet.date'] = date_facets.keys()
            
            for key, value in date_facets.items():
                # Date-based facets in Solr kinda suck.
                kwargs["f.%s.facet.date.start" % key] = self.conn._from_python(value.get('start_date'))
                kwargs["f.%s.facet.date.end" % key] = self.conn._from_python(value.get('end_date'))
                gap_string = value.get('gap_by').upper()
                
                if value.get('gap_amount') != 1:
                    gap_string = "%d%sS" % (value.get('gap_amount'), gap_string)
                
                kwargs["f.%s.facet.date.gap" % key] = "/%s" % gap_string
        
        if query_facets is not None:
            kwargs['facet'] = 'on'
            kwargs['facet.query'] = ["%s:%s" % (field, value) for field, value in query_facets.items()]
        
        if narrow_queries is not None:
            kwargs['fq'] = list(narrow_queries)
        
        if collapse_field is not None:
            #http://wiki.apache.org/solr/FieldCollapsing
            #&collapse.field=url&collapse.max=1&collapse.type=normal
            kwargs['collapse'] = 'true'
            kwargs['collapse.type'] = collapse_type
            kwargs['collapse.max'] = collapse_max# assumes defaults to an int
            kwargs['collapse.field'] = collapse_field
        
        raw_results = self.conn.search(query_string, **kwargs)
        return self._process_results(raw_results, highlight=highlight)
    
    def more_like_this(self, model_instance, additional_query_string=None, start_offset=0, end_offset=None, **kwargs):
        index = self.site.get_index(model_instance.__class__)
        field_name = index.get_content_field()
        params = {
            'fl': '*,score',
        }
        
        if start_offset is not None:
            params['start'] = start_offset
        
        if end_offset is not None:
            params['rows'] = end_offset
        
        if additional_query_string:
            params['fq'] = additional_query_string
        
        raw_results = self.conn.more_like_this("id:%s" % self.get_identifier(model_instance), field_name, **params)
        return self._process_results(raw_results)
    
    def _process_results(self, raw_results, highlight=False):
        from haystack import site
        results = []
        hits = raw_results.hits
        facets = {}
        spelling_suggestion = None
        
        if hasattr(raw_results, 'facets'):
            facets = {
                'fields': raw_results.facets.get('facet_fields', {}),
                'dates': raw_results.facets.get('facet_dates', {}),
                'queries': raw_results.facets.get('facet_queries', {}),
            }
            

# 需要导入模块: from pysolr import Solr [as 别名]
# 或者: from pysolr.Solr import more_like_this [as 别名]

#.........这里部分代码省略.........
        sort_by=None,
        start_offset=0,
        end_offset=None,
        fields="",
        highlight=False,
        facets=None,
        date_facets=None,
        query_facets=None,
        narrow_queries=None,
    ):
        if len(query_string) == 0:
            return []

        kwargs = {"fl": "* score"}

        if fields:
            kwargs["fl"] = fields

        if sort_by is not None:
            kwargs["sort"] = sort_by

        if start_offset is not None:
            kwargs["start"] = start_offset

        if end_offset is not None:
            kwargs["rows"] = end_offset

        if highlight is True:
            kwargs["hl"] = "true"
            kwargs["hl.fragsize"] = "200"

        if facets is not None:
            kwargs["facet"] = "on"
            kwargs["facet.field"] = facets

        if date_facets is not None:
            kwargs["facet"] = "on"
            kwargs["facet.date"] = date_facets.keys()

            for key, value in date_facets.items():
                # Date-based facets in Solr kinda suck.
                kwargs["f.%s.facet.date.start" % key] = self.conn._from_python(value.get("start_date"))
                kwargs["f.%s.facet.date.end" % key] = self.conn._from_python(value.get("end_date"))
                kwargs["f.%s.facet.date.gap" % key] = value.get("gap")

        if query_facets is not None:
            kwargs["facet"] = "on"
            kwargs["facet.query"] = ["%s:%s" % (field, value) for field, value in query_facets.items()]

        if narrow_queries is not None:
            kwargs["fq"] = list(narrow_queries)

        raw_results = self.conn.search(query_string, **kwargs)
        return self._process_results(raw_results, highlight=highlight)

    def more_like_this(self, model_instance):
        from haystack.sites import site, NotRegistered

        index = site.get_index(model_instance.__class__)
        field_name = index.get_content_field()
        raw_results = self.conn.more_like_this("id:%s" % self.get_identifier(model_instance), field_name, fl="*,score")
        return self._process_results(raw_results)

    def _process_results(self, raw_results, highlight=False):
        results = []
        facets = {}

        if hasattr(raw_results, "facets"):
            facets = {
                "fields": raw_results.facets.get("facet_fields", {}),
                "dates": raw_results.facets.get("facet_dates", {}),
                "queries": raw_results.facets.get("facet_queries", {}),
            }

            for key in ["fields"]:
                for facet_field in facets[key]:
                    # Convert to a two-tuple, as Solr's json format returns a list of
                    # pairs.
                    facets[key][facet_field] = zip(facets[key][facet_field][::2], facets[key][facet_field][1::2])

        for raw_result in raw_results.docs:
            app_label, module_name = raw_result["django_ct_s"].split(".")
            additional_fields = {}

            for key, value in raw_result.items():
                additional_fields[str(key)] = self.conn._to_python(value)

            del (additional_fields["django_ct_s"])
            del (additional_fields["django_id_s"])
            del (additional_fields["score"])

            if raw_result["id"] in getattr(raw_results, "highlighting", {}):
                additional_fields["highlighted"] = raw_results.highlighting[raw_result["id"]]

            result = SearchResult(
                app_label, module_name, raw_result["django_id_s"], raw_result["score"], **additional_fields
            )
            results.append(result)

        return {"results": results, "hits": raw_results.hits, "facets": facets}

# 需要导入模块: from pysolr import Solr [as 别名]
# 或者: from pysolr.Solr import more_like_this [as 别名]

#.........这里部分代码省略.........
        self.assertEqual(results.highlighting, {u'doc_4': {}, u'doc_2': {}, u'doc_1': {}})
        self.assertEqual(results.spellcheck, {})
        self.assertEqual(results.facets['facet_fields']['popularity'], ['10', 2, '7', 1, '2', 0, '8', 0])
        self.assertTrue(results.qtime is not None)
        # TODO: Can't get these working in my test setup.
        # self.assertEqual(results.grouped, '')

    def test_search_with_nested_q(self):
        nested_q = self.solr.create_nested_q('edismax', 'blue', **{
                'qf': 'description comments'
        })
        results = self.solr.search('pony AND {}'.format(nested_q))
        
        self.assertSameIDs(results, ['sn6', 'sn2', 'sn1'])

    def test_disjunction_max(self):
        results = self.solr.disjunction_max('blue', 'description comments')
        
        self.assertSameIDs(results, ['sn6', 'sn4', 'sn2', 'sn1'])

    def test_disjunction_max_with_nested_q(self):
        nested_q = self.solr.create_nested_q('edismax', 'blue', **{
                'qf': 'description comments'
        })
        results = self.solr.disjunction_max('unicorn AND {}'.format(nested_q), 'cat name')
        
        self.assertSameIDs(results, ['sn6', 'sn4', 'sn2'])

    def test_spatial_search(self):
        results = self.solr.spatial_search('pony', 'store', '54.33131,10.12135', '100')
        
        self.assertSameIDs(results, ['sn6', 'sn3', 'sn2'])

    def test_more_like_this(self):
        results = self.solr.more_like_this('id:doc_1', 'text')
        self.assertEqual(len(results), 0)

    def test_suggest_terms(self):
        results = self.solr.suggest_terms('title', '')
        self.assertEqual(len(results), 1)
        self.assertEqual(results, {'title': [('doc', 3), ('another', 2), ('example', 2), ('1', 1), ('2', 1), ('boring', 1), ('rock', 1), ('thing', 1)]})

    def test__build_doc(self):
        doc = {
            'id': 'doc_1',
            'title': 'Example doc ☃ 1',
            'price': 12.59,
            'popularity': 10,
        }
        doc_xml = force_unicode(ET.tostring(self.solr._build_doc(doc), encoding='utf-8'))
        self.assertTrue('<field name="title">Example doc ☃ 1</field>' in doc_xml)
        self.assertTrue('<field name="id">doc_1</field>' in doc_xml)
        self.assertEqual(len(doc_xml), 152)

    def test_add(self):
        self.assertEqual(len(self.solr.search('doc')), 3)
        self.assertEqual(len(self.solr.search('example')), 2)

        self.solr.add([
            {
                'id': 'doc_6',
                'title': 'Newly added doc',
            },
            {
                'id': 'doc_7',
                'title': 'Another example doc',

# 需要导入模块: from pysolr import Solr [as 别名]
# 或者: from pysolr.Solr import more_like_this [as 别名]

#.........这里部分代码省略.........
                min_lng,
                max_lat,
                max_lng,
            )
            kwargs["fq"].append(bbox)

        if dwithin is not None:
            kwargs.setdefault("fq", [])
            lng, lat = dwithin["point"].coords
            geofilt = "{!geofilt pt=%s,%s sfield=%s d=%s}" % (
                lat,
                lng,
                dwithin["field"],
                dwithin["distance"].km,
            )
            kwargs["fq"].append(geofilt)

        # Check to see if the backend should try to include distances
        # (Solr 4.X+) in the results.
        if self.distance_available and distance_point:
            # In early testing, you can't just hand Solr 4.X a proper bounding box
            # & request distances. To enable native distance would take calculating
            # a center point & a radius off the user-provided box, which kinda
            # sucks. We'll avoid it for now, since Solr 4.x's release will be some
            # time yet.
            # kwargs['fl'] += ' _dist_:geodist()'
            pass

        if extra_kwargs:
            kwargs.update(extra_kwargs)

        return kwargs

    def more_like_this(
        self,
        model_instance,
        additional_query_string=None,
        start_offset=0,
        end_offset=None,
        models=None,
        limit_to_registered_models=None,
        result_class=None,
        **kwargs
    ):
        from haystack import connections

        # Deferred models will have a different class ("RealClass_Deferred_fieldname")
        # which won't be in our registry:
        model_klass = model_instance._meta.concrete_model

        index = (
            connections[self.connection_alias]
            .get_unified_index()
            .get_index(model_klass)
        )
        field_name = index.get_content_field()
        params = {"fl": "*,score"}

        if start_offset is not None:
            params["start"] = start_offset

        if end_offset is not None:
            params["rows"] = end_offset

        narrow_queries = set()

# 需要导入模块: from pysolr import Solr [as 别名]
# 或者: from pysolr.Solr import more_like_this [as 别名]

#.........这里部分代码省略.........

        results = self.solr.search("example")
        self.assertEqual(len(results), 2)

        results = self.solr.search("nothing")
        self.assertEqual(len(results), 0)

        # Advanced options.
        results = self.solr.search(
            "doc",
            **{
                "debug": "true",
                "hl": "true",
                "hl.fragsize": 8,
                "facet": "on",
                "facet.field": "popularity",
                "spellcheck": "true",
                "spellcheck.collate": "true",
                "spellcheck.count": 1,
                # TODO: Can't get these working in my test setup.
                # 'group': 'true',
                # 'group.field': 'id',
            }
        )
        self.assertEqual(len(results), 3)
        self.assertTrue("explain" in results.debug)
        self.assertEqual(results.highlighting, {"doc_4": {}, "doc_2": {}, "doc_1": {}})
        self.assertEqual(results.spellcheck, {})
        self.assertEqual(results.facets["facet_fields"]["popularity"], ["10", 2, "7", 1, "2", 0, "8", 0])
        self.assertTrue(results.qtime is not None)
        # TODO: Can't get these working in my test setup.
        # self.assertEqual(results.grouped, '')

    def test_more_like_this(self):
        results = self.solr.more_like_this("id:doc_1", "text")
        self.assertEqual(len(results), 0)

    def test_suggest_terms(self):
        results = self.solr.suggest_terms("title", "")
        self.assertEqual(len(results), 1)
        self.assertEqual(
            results,
            {
                "title": [
                    ("doc", 3),
                    ("another", 2),
                    ("example", 2),
                    ("1", 1),
                    ("2", 1),
                    ("boring", 1),
                    ("rock", 1),
                    ("thing", 1),
                ]
            },
        )

    def test__build_doc(self):
        doc = {"id": "doc_1", "title": "Example doc ☃ 1", "price": 12.59, "popularity": 10}
        doc_xml = force_unicode(ET.tostring(self.solr._build_doc(doc), encoding="utf-8"))
        self.assertTrue('<field name="title">Example doc ☃ 1</field>' in doc_xml)
        self.assertTrue('<field name="id">doc_1</field>' in doc_xml)
        self.assertEqual(len(doc_xml), 152)

    def test_add(self):
        self.assertEqual(len(self.solr.search("doc")), 3)
        self.assertEqual(len(self.solr.search("example")), 2)

# 需要导入模块: from pysolr import Solr [as 别名]
# 或者: from pysolr.Solr import more_like_this [as 别名]

#.........这里部分代码省略.........

        results = self.solr.search('nothing')
        self.assertEqual(len(results), 0)

        # Advanced options.
        results = self.solr.search('doc', **{
            'debug': 'true',
            'hl': 'true',
            'hl.fragsize': 8,
            'facet': 'on',
            'facet.field': 'popularity',
            'spellcheck': 'true',
            'spellcheck.collate': 'true',
            'spellcheck.count': 1,
            # TODO: Can't get these working in my test setup.
            # 'group': 'true',
            # 'group.field': 'id',
        })
        self.assertEqual(len(results), 3)
        self.assertTrue('explain' in results.debug)
        self.assertEqual(results.highlighting, {u'doc_4': {}, u'doc_2': {}, u'doc_1': {}})
        self.assertEqual(results.spellcheck, {})
        self.assertEqual(results.facets['facet_fields']['popularity'], ['10', 2, '7', 1, '2', 0, '8', 0])
        self.assertTrue(results.qtime is not None)
        # TODO: Can't get these working in my test setup.
        # self.assertEqual(results.grouped, '')

        # search should support custom handlers
        with self.assertRaises(SolrError):
            self.solr.search('doc', handler='fakehandler')
        args, kwargs = self.solr._send_request.call_args
        self.assertTrue(args[1].startswith('fakehandler'))

    def test_more_like_this(self):
        results = self.solr.more_like_this('id:doc_1', 'text')
        self.assertEqual(len(results), 0)
        # more_like_this should default to 'mlt' handler
        args, kwargs = self.solr._send_request.call_args
        self.assertTrue(args[1].startswith('mlt/?'))

        # more_like_this should support custom handlers
        with self.assertRaises(SolrError):
            self.solr.more_like_this('id:doc_1', 'text', handler='fakehandler')
        args, kwargs = self.solr._send_request.call_args
        self.assertTrue(args[1].startswith('fakehandler'))

    def test_suggest_terms(self):
        results = self.solr.suggest_terms('title', '')
        self.assertEqual(len(results), 1)
        self.assertEqual(results, {'title': [('doc', 3), ('another', 2), ('example', 2), ('1', 1), ('2', 1), ('boring', 1), ('rock', 1), ('thing', 1)]})
        # suggest_terms should default to 'mlt' handler
        args, kwargs = self.solr._send_request.call_args
        self.assertTrue(args[1].startswith('terms/?'))

        # suggest_terms should support custom handlers
        with self.assertRaises(SolrError):
            self.solr.suggest_terms('title', '', handler='fakehandler')
        args, kwargs = self.solr._send_request.call_args
        self.assertTrue(args[1].startswith('fakehandler'))

    def test__build_doc(self):
        doc = {
            'id': 'doc_1',
            'title': 'Example doc ☃ 1',
            'price': 12.59,
            'popularity': 10,