Adds a fail_if_exist option to index creation functions (#180)

* Updated changelog * Added a fail_if_exists option to create_fulltext_index and create_vector_index * Updated test_indexes.py unit tests * Added tests to test_indexes.py to test fail_if_exists parameters * Replace double quotes with single quotes in f strings * Updated changelog
neo4j · Oct 15, 2024 · 102d799 · 102d799
1 parent 59c6207
commit 102d799
Show file tree

Hide file tree

Showing 3 changed files with 55 additions and 8 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -2,6 +2,12 @@
 
 ## Next
 
+### Added
+- Introduced a fail_if_exist option to index creation functions to control behavior when an index already exists.
+
+### Changed
+- Comprehensive rewrite of the README to improve clarity and provide detailed usage examples.
+
 ## 1.0.0
 
 ### Fixed

diff --git a/src/neo4j_graphrag/indexes.py b/src/neo4j_graphrag/indexes.py
@@ -38,6 +38,7 @@ def create_vector_index(
     embedding_property: str,
     dimensions: int,
     similarity_fn: Literal["euclidean", "cosine"],
+    fail_if_exists: bool = False,
     neo4j_database: Optional[str] = None,
 ) -> None:
     """
@@ -46,7 +47,6 @@ def create_vector_index(
 
     See Cypher manual on `creating vector indexes <https://neo4j.com/docs/cypher-manual/current/indexes/semantic-indexes/vector-indexes/#create-vector-index>`_.
 
-    Important: This operation will fail if an index with the same name already exists.
     Ensure that the index name provided is unique within the database context.
 
     Example:
@@ -72,6 +72,7 @@ def create_vector_index(
             embedding_property="vectorProperty",
             dimensions=1536,
             similarity_fn="euclidean",
+            fail_if_exists=False,
         )
 
 
@@ -83,6 +84,7 @@ def create_vector_index(
         dimensions (int): Vector embedding dimension
         similarity_fn (str): case-insensitive values for the vector similarity function:
             ``euclidean`` or ``cosine``.
+        fail_if_exists (bool): If True raise an error if the index already exists. Defaults to False.
         neo4j_database (Optional[str]): The name of the Neo4j database. If not provided, this defaults to "neo4j" in the database (`see reference to documentation <https://neo4j.com/docs/operations-manual/current/database-administration/#manage-databases-default>`_).
 
     Raises:
@@ -105,7 +107,7 @@ def create_vector_index(
 
     try:
         query = (
-            f"CREATE VECTOR INDEX $name FOR (n:{label}) ON n.{embedding_property} OPTIONS "
+            f"CREATE VECTOR INDEX $name {'' if fail_if_exists else 'IF NOT EXISTS'} FOR (n:{label}) ON n.{embedding_property} OPTIONS "
             "{ indexConfig: { `vector.dimensions`: toInteger($dimensions), `vector.similarity_function`: $similarity_fn } }"
         )
         logger.info(f"Creating vector index named '{name}'")
@@ -123,6 +125,7 @@ def create_fulltext_index(
     name: str,
     label: str,
     node_properties: list[str],
+    fail_if_exists: bool = False,
     neo4j_database: Optional[str] = None,
 ) -> None:
     """
@@ -131,7 +134,6 @@ def create_fulltext_index(
 
     See Cypher manual on `creating fulltext indexes <https://neo4j.com/docs/cypher-manual/current/indexes/semantic-indexes/full-text-indexes/#create-full-text-indexes>`_.
 
-    Important: This operation will fail if an index with the same name already exists.
     Ensure that the index name provided is unique within the database context.
 
     Example:
@@ -155,6 +157,7 @@ def create_fulltext_index(
             INDEX_NAME,
             label="Document",
             node_properties=["vectorProperty"],
+            fail_if_exists=False,
         )
 
 
@@ -163,6 +166,7 @@ def create_fulltext_index(
         name (str): The unique name of the index.
         label (str): The node label to be indexed.
         node_properties (list[str]): The node properties to create the fulltext index on.
+        fail_if_exists (bool): If True raise an error if the index already exists. Defaults to False.
         neo4j_database (Optional[str]): The name of the Neo4j database. If not provided, this defaults to "neo4j" in the database (`see reference to documentation <https://neo4j.com/docs/operations-manual/current/database-administration/#manage-databases-default>`_).
 
     Raises:
@@ -180,7 +184,7 @@ def create_fulltext_index(
 
     try:
         query = (
-            "CREATE FULLTEXT INDEX $name "
+            f"CREATE FULLTEXT INDEX $name {'' if fail_if_exists else 'IF NOT EXISTS'} "
             f"FOR (n:`{label}`) ON EACH "
             f"[{', '.join(['n.`' + prop + '`' for prop in node_properties])}]"
         )

diff --git a/tests/unit/test_indexes.py b/tests/unit/test_indexes.py
@@ -30,7 +30,7 @@
 
 def test_create_vector_index_happy_path(driver: MagicMock) -> None:
     create_query = (
-        "CREATE VECTOR INDEX $name FOR (n:People) ON n.name OPTIONS "
+        "CREATE VECTOR INDEX $name IF NOT EXISTS FOR (n:People) ON n.name OPTIONS "
         "{ indexConfig: { `vector.dimensions`: toInteger($dimensions), `vector.similarity_function`: $similarity_fn } }"
     )
 
@@ -43,9 +43,26 @@ def test_create_vector_index_happy_path(driver: MagicMock) -> None:
     )
 
 
+def test_create_vector_index_fail_if_exists(driver: MagicMock) -> None:
+    create_query = (
+        "CREATE VECTOR INDEX $name  FOR (n:People) ON n.name OPTIONS "
+        "{ indexConfig: { `vector.dimensions`: toInteger($dimensions), `vector.similarity_function`: $similarity_fn } }"
+    )
+
+    create_vector_index(
+        driver, "my-index", "People", "name", 2048, "cosine", fail_if_exists=True
+    )
+
+    driver.execute_query.assert_called_once_with(
+        create_query,
+        {"name": "my-index", "dimensions": 2048, "similarity_fn": "cosine"},
+        database_=None,
+    )
+
+
 def test_create_vector_index_ensure_escaping(driver: MagicMock) -> None:
     create_query = (
-        "CREATE VECTOR INDEX $name FOR (n:People) ON n.name OPTIONS "
+        "CREATE VECTOR INDEX $name IF NOT EXISTS FOR (n:People) ON n.name OPTIONS "
         "{ indexConfig: { `vector.dimensions`: toInteger($dimensions), `vector.similarity_function`: $similarity_fn } }"
     )
 
@@ -120,7 +137,7 @@ def test_create_fulltext_index_happy_path(driver: MagicMock) -> None:
     label = "node-label"
     text_node_properties = ["property-1", "property-2"]
     create_query = (
-        "CREATE FULLTEXT INDEX $name "
+        "CREATE FULLTEXT INDEX $name IF NOT EXISTS "
         f"FOR (n:`{label}`) ON EACH "
         f"[{', '.join(['n.`' + property + '`' for property in text_node_properties])}]"
     )
@@ -134,6 +151,26 @@ def test_create_fulltext_index_happy_path(driver: MagicMock) -> None:
     )
 
 
+def test_create_fulltext_index_fail_if_exists(driver: MagicMock) -> None:
+    label = "node-label"
+    text_node_properties = ["property-1", "property-2"]
+    create_query = (
+        "CREATE FULLTEXT INDEX $name  "
+        f"FOR (n:`{label}`) ON EACH "
+        f"[{', '.join(['n.`' + property + '`' for property in text_node_properties])}]"
+    )
+
+    create_fulltext_index(
+        driver, "my-index", label, text_node_properties, fail_if_exists=True
+    )
+
+    driver.execute_query.assert_called_once_with(
+        create_query,
+        {"name": "my-index"},
+        database_=None,
+    )
+
+
 def test_create_fulltext_index_raises_error_with_neo4j_client_error(
     driver: MagicMock,
 ) -> None:
@@ -159,7 +196,7 @@ def test_create_fulltext_index_ensure_escaping(driver: MagicMock) -> None:
     label = "node-label"
     text_node_properties = ["property-1", "property-2"]
     create_query = (
-        "CREATE FULLTEXT INDEX $name "
+        "CREATE FULLTEXT INDEX $name IF NOT EXISTS "
         f"FOR (n:`{label}`) ON EACH "
         f"[{', '.join(['n.`' + property + '`' for property in text_node_properties])}]"
     )