graphiti-core 0.21.0rc2__py3-none-any.whl → 0.21.0rc4__py3-none-any.whl

graphiti_core/driver/driver.py

@@ -311,3 +311,12 @@ class GraphDriver(ABC):
                 return success if failed == 0 else success
 
         return 0
+
+    def build_fulltext_query(
+        self, query: str, group_ids: list[str] | None = None, max_query_length: int = 128
+    ) -> str:
+        """
+        Specific fulltext query builder for database providers.
+        Only implemented by providers that need custom fulltext query building.
+        """
+        raise NotImplementedError(f'build_fulltext_query not implemented for {self.provider}')

graphiti_core/driver/falkordb_driver.py

@@ -36,6 +36,42 @@ from graphiti_core.utils.datetime_utils import convert_datetimes_to_strings
 
 logger = logging.getLogger(__name__)
 
+STOPWORDS = [
+    'a',
+    'is',
+    'the',
+    'an',
+    'and',
+    'are',
+    'as',
+    'at',
+    'be',
+    'but',
+    'by',
+    'for',
+    'if',
+    'in',
+    'into',
+    'it',
+    'no',
+    'not',
+    'of',
+    'on',
+    'or',
+    'such',
+    'that',
+    'their',
+    'then',
+    'there',
+    'these',
+    'they',
+    'this',
+    'to',
+    'was',
+    'will',
+    'with',
+]
+
 
 class FalkorDriverSession(GraphDriverSession):
     provider = GraphProvider.FALKORDB
@@ -167,3 +203,78 @@ class FalkorDriver(GraphDriver):
         cloned = FalkorDriver(falkor_db=self.client, database=database)
 
         return cloned
+
+    def sanitize(self, query: str) -> str:
+        """
+        Replace FalkorDB special characters with whitespace.
+        Based on FalkorDB tokenization rules: ,.<>{}[]"':;!@#$%^&*()-+=~
+        """
+        # FalkorDB separator characters that break text into tokens
+        separator_map = str.maketrans(
+            {
+                ',': ' ',
+                '.': ' ',
+                '<': ' ',
+                '>': ' ',
+                '{': ' ',
+                '}': ' ',
+                '[': ' ',
+                ']': ' ',
+                '"': ' ',
+                "'": ' ',
+                ':': ' ',
+                ';': ' ',
+                '!': ' ',
+                '@': ' ',
+                '#': ' ',
+                '$': ' ',
+                '%': ' ',
+                '^': ' ',
+                '&': ' ',
+                '*': ' ',
+                '(': ' ',
+                ')': ' ',
+                '-': ' ',
+                '+': ' ',
+                '=': ' ',
+                '~': ' ',
+                '?': ' ',
+            }
+        )
+        sanitized = query.translate(separator_map)
+        # Clean up multiple spaces
+        sanitized = ' '.join(sanitized.split())
+        return sanitized
+
+    def build_fulltext_query(
+        self, query: str, group_ids: list[str] | None = None, max_query_length: int = 128
+    ) -> str:
+        """
+        Build a fulltext query string for FalkorDB using RedisSearch syntax.
+        FalkorDB uses RedisSearch-like syntax where:
+        - Field queries use @ prefix: @field:value
+        - Multiple values for same field: (@field:value1|value2)
+        - Text search doesn't need @ prefix for content fields
+        - AND is implicit with space: (@group_id:value) (text)
+        - OR uses pipe within parentheses: (@group_id:value1|value2)
+        """
+        if group_ids is None or len(group_ids) == 0:
+            group_filter = ''
+        else:
+            group_values = '|'.join(group_ids)
+            group_filter = f'(@group_id:{group_values})'
+
+        sanitized_query = self.sanitize(query)
+
+        # Remove stopwords from the sanitized query
+        query_words = sanitized_query.split()
+        filtered_words = [word for word in query_words if word.lower() not in STOPWORDS]
+        sanitized_query = ' | '.join(filtered_words)
+
+        # If the query is too long return no query
+        if len(sanitized_query.split(' ')) + len(group_ids or '') >= max_query_length:
+            return ''
+
+        full_query = group_filter + ' (' + sanitized_query + ')'
+
+        return full_query

graphiti_core/graph_queries.py

@@ -71,12 +71,41 @@ def get_range_indices(provider: GraphProvider) -> list[LiteralString]:
 
 def get_fulltext_indices(provider: GraphProvider) -> list[LiteralString]:
     if provider == GraphProvider.FALKORDB:
-        return [
-            """CREATE FULLTEXT INDEX FOR (e:Episodic) ON (e.content, e.source, e.source_description, e.group_id)""",
-            """CREATE FULLTEXT INDEX FOR (n:Entity) ON (n.name, n.summary, n.group_id)""",
-            """CREATE FULLTEXT INDEX FOR (n:Community) ON (n.name, n.group_id)""",
-            """CREATE FULLTEXT INDEX FOR ()-[e:RELATES_TO]-() ON (e.name, e.fact, e.group_id)""",
-        ]
+        from typing import cast
+
+        from graphiti_core.driver.falkordb_driver import STOPWORDS
+
+        # Convert to string representation for embedding in queries
+        stopwords_str = str(STOPWORDS)
+
+        # Use type: ignore to satisfy LiteralString requirement while maintaining single source of truth
+        return cast(
+            list[LiteralString],
+            [
+                f"""CALL db.idx.fulltext.createNodeIndex(
+                                                {{
+                                                    label: 'Episodic',
+                                                    stopwords: {stopwords_str}
+                                                }},
+                                                'content', 'source', 'source_description', 'group_id'
+                                                )""",
+                f"""CALL db.idx.fulltext.createNodeIndex(
+                                                {{
+                                                    label: 'Entity',
+                                                    stopwords: {stopwords_str}
+                                                }},
+                                                'name', 'summary', 'group_id'
+                                                )""",
+                f"""CALL db.idx.fulltext.createNodeIndex(
+                                                {{
+                                                    label: 'Community',
+                                                    stopwords: {stopwords_str}
+                                                }},
+                                                'name', 'group_id'
+                                                )""",
+                """CREATE FULLTEXT INDEX FOR ()-[e:RELATES_TO]-() ON (e.name, e.fact, e.group_id)""",
+            ],
+        )
 
     if provider == GraphProvider.KUZU:
         return [

graphiti_core/graphiti.py

@@ -456,12 +456,12 @@ class Graphiti:
             start = time()
             now = utc_now()
 
-            # if group_id is None, use the default group id by the provider
-            group_id = group_id or get_default_group_id(self.driver.provider)
             validate_entity_types(entity_types)
 
             validate_excluded_entity_types(excluded_entity_types, entity_types)
             validate_group_id(group_id)
+            # if group_id is None, use the default group id by the provider
+            group_id = group_id or get_default_group_id(self.driver.provider)
 
             previous_episodes = (
                 await self.retrieve_episodes(

graphiti_core/helpers.py

@@ -54,7 +54,7 @@ def get_default_group_id(provider: GraphProvider) -> str:
     For most databases, the default group id is an empty string, while there are database types that require a specific default group id.
     """
     if provider == GraphProvider.FALKORDB:
-        return '_'
+        return '\\_'
     else:
         return ''
 
@@ -116,7 +116,7 @@ async def semaphore_gather(
     return await asyncio.gather(*(_wrap_coroutine(coroutine) for coroutine in coroutines))
 
 
-def validate_group_id(group_id: str) -> bool:
+def validate_group_id(group_id: str | None) -> bool:
     """
     Validate that a group_id contains only ASCII alphanumeric characters, dashes, and underscores.
 

graphiti_core/search/search_utils.py

@@ -92,6 +92,8 @@ def fulltext_query(query: str, group_ids: list[str] | None, driver: GraphDriver)
         if len(query.split(' ')) > MAX_QUERY_LENGTH:
             return ''
         return query
+    elif driver.provider == GraphProvider.FALKORDB:
+        return driver.build_fulltext_query(query, group_ids, MAX_QUERY_LENGTH)
     group_ids_filter_list = (
         [driver.fulltext_syntax + f'group_id:"{g}"' for g in group_ids]
         if group_ids is not None

graphiti_core/utils/bulk_utils.py

@@ -136,12 +136,14 @@ async def add_nodes_and_edges_bulk_tx(
         entity_data: dict[str, Any] = {
             'uuid': node.uuid,
             'name': node.name,
-            'name_embedding': node.name_embedding,
             'group_id': node.group_id,
             'summary': node.summary,
             'created_at': node.created_at,
         }
 
+        if not bool(driver.aoss_client):
+            entity_data['name_embedding'] = node.name_embedding
+
         entity_data['labels'] = list(set(node.labels + ['Entity']))
         if driver.provider == GraphProvider.KUZU:
             attributes = convert_datetimes_to_strings(node.attributes) if node.attributes else {}
@@ -161,7 +163,6 @@ async def add_nodes_and_edges_bulk_tx(
             'target_node_uuid': edge.target_node_uuid,
             'name': edge.name,
             'fact': edge.fact,
-            'fact_embedding': edge.fact_embedding,
             'group_id': edge.group_id,
             'episodes': edge.episodes,
             'created_at': edge.created_at,
@@ -170,6 +171,9 @@ async def add_nodes_and_edges_bulk_tx(
             'invalid_at': edge.invalid_at,
         }
 
+        if not bool(driver.aoss_client):
+            edge_data['fact_embedding'] = edge.fact_embedding
+
         if driver.provider == GraphProvider.KUZU:
             attributes = convert_datetimes_to_strings(edge.attributes) if edge.attributes else {}
             edge_data['attributes'] = json.dumps(attributes)
@@ -195,21 +199,29 @@ async def add_nodes_and_edges_bulk_tx(
     else:
         await tx.run(get_episode_node_save_bulk_query(driver.provider), episodes=episodes)
         await tx.run(
-            get_entity_node_save_bulk_query(driver.provider, nodes),
+            get_entity_node_save_bulk_query(
+                driver.provider, nodes, has_aoss=bool(driver.aoss_client)
+            ),
             nodes=nodes,
-            has_aoss=bool(driver.aoss_client),
         )
         await tx.run(
             get_episodic_edge_save_bulk_query(driver.provider),
             episodic_edges=[edge.model_dump() for edge in episodic_edges],
         )
         await tx.run(
-            get_entity_edge_save_bulk_query(driver.provider),
+            get_entity_edge_save_bulk_query(driver.provider, has_aoss=bool(driver.aoss_client)),
             entity_edges=edges,
-            has_aoss=bool(driver.aoss_client),
         )
 
-        if driver.aoss_client:
+        if bool(driver.aoss_client):
+            for node_data, entity_node in zip(nodes, entity_nodes, strict=True):
+                if node_data.get('uuid') == entity_node.uuid:
+                    node_data['name_embedding'] = entity_node.name_embedding
+
+            for edge_data, entity_edge in zip(edges, entity_edges, strict=True):
+                if edge_data.get('uuid') == entity_edge.uuid:
+                    edge_data['fact_embedding'] = entity_edge.fact_embedding
+
             await driver.save_to_aoss(EPISODE_INDEX_NAME, episodes)
             await driver.save_to_aoss(ENTITY_INDEX_NAME, nodes)
             await driver.save_to_aoss(ENTITY_EDGE_INDEX_NAME, edges)

{graphiti_core-0.21.0rc2.dist-info → graphiti_core-0.21.0rc4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: graphiti-core
-Version: 0.21.0rc2
+Version: 0.21.0rc4
 Summary: A temporal graph building library
 Project-URL: Homepage, https://help.getzep.com/graphiti/graphiti/overview
 Project-URL: Repository, https://github.com/getzep/graphiti

{graphiti_core-0.21.0rc2.dist-info → graphiti_core-0.21.0rc4.dist-info}/RECORD

@@ -1,10 +1,10 @@
 graphiti_core/__init__.py,sha256=e5SWFkRiaUwfprYIeIgVIh7JDedNiloZvd3roU-0aDY,55
 graphiti_core/edges.py,sha256=PhJm_s28cHLEaIqcw66wP16hOq4P4bVQbC_sESHQkXU,20919
 graphiti_core/errors.py,sha256=cH_v9TPgEPeQE6GFOHIg5TvejpUCBddGarMY2Whxbwc,2707
-graphiti_core/graph_queries.py,sha256=9DWMiFTB-OmodMDaOws0lwzgiD7EUDNO7mAFJ1nxusE,6624
-graphiti_core/graphiti.py,sha256=UPa85sdCdO4xnl68EsFYnbkqTgta2LA-XwP7hzcAsyg,42072
+graphiti_core/graph_queries.py,sha256=ZWMqAo5pwb8PO5ddg4zZ0ArhHWuWV42g3R9ULIxsHOs,8058
+graphiti_core/graphiti.py,sha256=Uxppvzjc-jFO6JCRE3PUlU5B5zUo-jBrC02uETb0-x0,42072
 graphiti_core/graphiti_types.py,sha256=C_p2XwScQlCzo7ets097TrSLs9ATxPZQ4WCsxDS7QHc,1066
-graphiti_core/helpers.py,sha256=6q_wpiOW3_j28EfZ7FgWW7Hl5pONj_5zvVXZGW9FxTU,5175
+graphiti_core/helpers.py,sha256=q8kbL9gz8igdlh-oMUS-ylUyeMlXZb-ccf-HQkrES_0,5184
 graphiti_core/nodes.py,sha256=wYLQcVEXvQMxTpTc9LWSoPTzzaoUOm0rl07c9wS1XSY,30323
 graphiti_core/py.typed,sha256=vlmmzQOt7bmeQl9L3XJP4W6Ry0iiELepnOrinKz5KQg,79
 graphiti_core/cross_encoder/__init__.py,sha256=hry59vz21x-AtGZ0MJ7ugw0HTwJkXiddpp_Yqnwsen0,723
@@ -13,8 +13,8 @@ graphiti_core/cross_encoder/client.py,sha256=KLsbfWKOEaAV3adFe3XZlAeb-gje9_sVKCV
 graphiti_core/cross_encoder/gemini_reranker_client.py,sha256=hmITG5YIib52nrKvINwRi4xTfAO1U4jCCaEVIwImHw0,6208
 graphiti_core/cross_encoder/openai_reranker_client.py,sha256=WHMl6Q6gEslR2EzjwpFSZt2Kh6bnu8alkLvzmi0MDtg,4674
 graphiti_core/driver/__init__.py,sha256=kCWimqQU19airu5gKwCmZtZuXkDfaQfKSUhMDoL-rTA,626
-graphiti_core/driver/driver.py,sha256=5YbpyUq7L1pf8s5R2FYv6B3KabTCmRvhSAHiVyRvt5o,10433
-graphiti_core/driver/falkordb_driver.py,sha256=JsNBRQHBVENA8eqAngD-8dw1aTH1ZKUtE1on8sd7owY,6431
+graphiti_core/driver/driver.py,sha256=EO9Aj5O2vpH7iyvQQcE5uJGQ8eA-_i6f8NwfAlW8r74,10831
+graphiti_core/driver/falkordb_driver.py,sha256=Q-dImfK4O2bkikqFzo0Wg2g7iFFRSuzy_c6u82tX6-M,9361
 graphiti_core/driver/kuzu_driver.py,sha256=RcWu8E0CCdofrFe34NmCeqfuhaZr_7ZN5jqDkI3VQMI,5453
 graphiti_core/driver/neo4j_driver.py,sha256=E93PdOZaH7wzEbIfoiDSYht49jr6zSzvMMyo1INGEOw,4096
 graphiti_core/driver/neptune_driver.py,sha256=akNLHhFHPEeQu-xO3PM51RomklntT6k5eA2CQ4AFbCc,10311
@@ -60,11 +60,11 @@ graphiti_core/search/search_config.py,sha256=v_rUHsu1yo5OuPfEm21lSuXexQs-o8qYwSS
 graphiti_core/search/search_config_recipes.py,sha256=4GquRphHhJlpXQhAZOySYnCzBWYoTwxlJj44eTOavZQ,7443
 graphiti_core/search/search_filters.py,sha256=DOAmYkc6A0z20EZId5fJZj1RvLz4WeQcoPANk9k-Sh8,10304
 graphiti_core/search/search_helpers.py,sha256=wj3ARlCNnZixNNntgCdAqzGoE4de4lW3r4rSG-3WyGw,2877
-graphiti_core/search/search_utils.py,sha256=HqNYbyQklAq5GoOUo_W8Xut1GMGVYlNv0kE0rh03KHo,76827
+graphiti_core/search/search_utils.py,sha256=ak1aBeKNuxS7szydNHwva2ABWSRlQ0S_v8ZOx7k0wc4,76958
 graphiti_core/telemetry/__init__.py,sha256=5kALLDlU9bb2v19CdN7qVANsJWyfnL9E60J6FFgzm3o,226
 graphiti_core/telemetry/telemetry.py,sha256=47LrzOVBCcZxsYPsnSxWFiztHoxYKKxPwyRX0hnbDGc,3230
 graphiti_core/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-graphiti_core/utils/bulk_utils.py,sha256=XyjWg_S6sQ1hM8HfHHoDnDn5yAWO1Ja0fYGFMfS7XV8,17071
+graphiti_core/utils/bulk_utils.py,sha256=9XWXqjxiu2ydKMLKQRTbvzO6cO1o1HRjjpmaf5Ym51k,17633
 graphiti_core/utils/datetime_utils.py,sha256=J-zYSq7-H-2n9hYOXNIun12kM10vNX9mMATGR_egTmY,1806
 graphiti_core/utils/maintenance/__init__.py,sha256=vW4H1KyapTl-OOz578uZABYcpND4wPx3Vt6aAPaXh78,301
 graphiti_core/utils/maintenance/community_operations.py,sha256=XMiokEemn96GlvjkOvbo9hIX04Fea3eVj408NHG5P4o,11042
@@ -74,7 +74,7 @@ graphiti_core/utils/maintenance/node_operations.py,sha256=r9ilkA01eq1z-nF8P_s1EX
 graphiti_core/utils/maintenance/temporal_operations.py,sha256=IIaVtShpVkOYe6haxz3a1x3v54-MzaEXG8VsxFUNeoY,3582
 graphiti_core/utils/maintenance/utils.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 graphiti_core/utils/ontology_utils/entity_types_utils.py,sha256=4eVgxLWY6Q8k9cRJ5pW59IYF--U4nXZsZIGOVb_yHfQ,1285
-graphiti_core-0.21.0rc2.dist-info/METADATA,sha256=Pqw2f7ySYgCDyyuWDBvT4KLrL-SCJXa2aFHlNzIJzJo,26933
-graphiti_core-0.21.0rc2.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-graphiti_core-0.21.0rc2.dist-info/licenses/LICENSE,sha256=KCUwCyDXuVEgmDWkozHyniRyWjnWUWjkuDHfU6o3JlA,11325
-graphiti_core-0.21.0rc2.dist-info/RECORD,,
+graphiti_core-0.21.0rc4.dist-info/METADATA,sha256=W_bFyM6HCQ8N7-tFNyjGHuRafGDYj89EtQ4tngZPUKc,26933
+graphiti_core-0.21.0rc4.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+graphiti_core-0.21.0rc4.dist-info/licenses/LICENSE,sha256=KCUwCyDXuVEgmDWkozHyniRyWjnWUWjkuDHfU6o3JlA,11325
+graphiti_core-0.21.0rc4.dist-info/RECORD,,