Support multi_match query type (#41)

* Support multi_match queries

luqum/__init__.py

@@ -1,4 +1,4 @@
 # -*- coding: utf-8 -*-
 
-__version__ = '0.7.5'
+__version__ = '0.8.0'
 __version_info__ = tuple(__version__.split('.'))

luqum/elasticsearch/tree.py

@@ -38,8 +38,10 @@ def __init__(self, no_analyze=None, method='term', fields=[], _name=None, field_
     def json(self):
         field = self.field
         inner_json = dict(self.field_options.get(field, {}))
-        inner_json.pop('type', None)  # remove "type" key
-        if self.method == 'query_string':
+        result = inner_json.pop('match_type', None)  # remove "match_type" key
+        if not result:  # conditionally remove "type" (for backward compatibility)
+            inner_json.pop('type', None)
+        if self.method in ['query_string', 'multi_match']:
             json = {self.method: inner_json}
         else:
             json = {self.method: {field: inner_json}}
@@ -50,7 +52,7 @@ def json(self):
             value = getattr(self, key, None)
             if value is not None:
                 if key == 'q':
-                    if self.method.startswith('match'):
+                    if 'match' in self.method:
                         inner_json['query'] = value
                         if self.method == 'match':
                             inner_json['zero_terms_query'] = self.zero_terms_query
@@ -94,7 +96,9 @@ def method(self):
             if self._value_has_wildcard_char():
                 return 'query_string'
             elif self._method.startswith("match"):
-                return self.field_options.get(self.field, {}).get("type", self._method)
+                options = self.field_options.get(self.field, {})
+                # Support the type opiton for backward compatibility
+                return options.get("match_type", options.get("type", self._method))
         return self._method
 
 

luqum/elasticsearch/visitor.py

@@ -79,7 +79,7 @@ def __init__(self, default_operator=SHOULD, default_field='text',
           None, will accept all non nested fields or object fields as sub fields.
         :param dict field_options: allows you to give defaults options for each fields.
           They will be applied unless, overwritten by generated parameters.
-          For match query, the `type` parameter modifies the query type.
+          For match query, the `match_type` parameter modifies the type of match query.
         :param bool match_word_as_phrase: if True,
           word expressions are matched using `match_phrase` instead of `match`.
           This options mainly keeps stability with 0.6 version.

luqum/tests/test_es_integration.py

@@ -4,7 +4,7 @@
 import elasticsearch_dsl
 from elasticsearch import Elasticsearch
 from elasticsearch.helpers import bulk
-from elasticsearch_dsl import Date, Index, Integer, Nested, Object, Search
+from elasticsearch_dsl import Date, Index, Integer, Nested, Object, Search, analyzer
 from elasticsearch_dsl.connections import connections
 from luqum.elasticsearch import ElasticsearchQueryBuilder, SchemaAnalyzer
 from luqum.parser import parser
@@ -37,7 +37,12 @@ class Illustrator(InnerDoc):
 
 
 class Book(Document):
-    title = Text()
+    title = Text(fields={
+        "no_vowels": Text(
+            analyzer=analyzer("no_vowels", "pattern", pattern="[\Waeiouy]"),
+            search_analyzer="standard"
+        )
+    })
     edition = Text()
     author = Object(properties={"name": Text(), "birthdate": Date()})
     publication_date = Date()
@@ -88,9 +93,16 @@ def setUpClass(cls):
         cls.search = Search(using=client, index="bk")
         MESSAGES_SCHEMA = {"mappings": Book._doc_type.mapping.to_dict()}
         schema_analizer = SchemaAnalyzer(MESSAGES_SCHEMA)
-        cls.es_builder = ElasticsearchQueryBuilder(
-            **schema_analizer.query_builder_options()
-        )
+
+        builder_options = schema_analizer.query_builder_options()
+        builder_options['field_options'] = {
+            'title.no_vowels': {
+                'match_type': 'multi_match',
+                'type': 'most_fields',
+                'fields': ('title', 'title.no_vowels')
+            }
+        }
+        cls.es_builder = ElasticsearchQueryBuilder(**builder_options)
         add_data()
 
     def _ask_luqum(self, req):
@@ -214,6 +226,12 @@ def test_complex_search(self):
             ["Harry Potter and the Order of the Phoenix"],
         )
 
+    def test_subfield_multi_match_search(self):
+        self.assertListEqual(
+            self._ask_luqum("title.no_vowels:Potter AND title.no_vowels:x"),
+            ["Harry Potter and the Order of the Phoenix"],
+        )
+
     @classmethod
     def tearDownClass(cls):
         if ES6: