Text2Cypher custom prompt: doc, example and bug fix (#229)

* Doc + bug fix

* Do not change the behavior, just document they said

* Use same order for patched functions and check order of mocked object

docs/source/api.rst

@@ -338,12 +338,21 @@ RagTemplate
 
 .. autoclass:: neo4j_graphrag.generation.prompts.RagTemplate
     :members:
+    :exclude-members: format
 
 ERExtractionTemplate
 --------------------
 
 .. autoclass:: neo4j_graphrag.generation.prompts.ERExtractionTemplate
     :members:
+    :exclude-members: format
+
+Text2CypherTemplate
+--------------------
+
+.. autoclass:: neo4j_graphrag.generation.prompts.Text2CypherTemplate
+    :members:
+    :exclude-members: format
 
 
 ****

examples/README.md

@@ -58,7 +58,8 @@ are listed in [the last section of this file](#customize).
 
 - [Control result format for VectorRetriever](customize/retrievers/result_formatter_vector_retriever.py)
 - [Control result format for VectorCypherRetriever](customize/retrievers/result_formatter_vector_cypher_retriever.py)
-
+- [Use pre-filters](customize/retrievers/use_pre_filters.py)
+- [Text2Cypher: use a custom prompt](customize/retrievers/text2cypher_custom_prompt.py)
 
 ### LLMs
 
@@ -74,7 +75,7 @@ are listed in [the last section of this file](#customize).
 
 ### Prompts
 
-- [Using a custom prompt](old/graphrag_custom_prompt.py)
+- [Using a custom prompt for RAG](customize/answer/custom_prompt.py)
 
 
 ### Embedders

examples/customize/retrievers/text2cypher_custom_prompt.py

@@ -0,0 +1,76 @@
+"""The example shows how to provide a custom prompt to Text2CypherRetriever.
+
+Example using the OpenAILLM, hence the OPENAI_API_KEY needs to be set in the
+environment for this example to run.
+"""
+
+import neo4j
+from neo4j_graphrag.llm import OpenAILLM
+from neo4j_graphrag.retrievers import Text2CypherRetriever
+from neo4j_graphrag.schema import get_schema
+
+# Define database credentials
+URI = "neo4j+s://demo.neo4jlabs.com"
+AUTH = ("recommendations", "recommendations")
+DATABASE = "recommendations"
+
+# Create LLM object
+llm = OpenAILLM(model_name="gpt-4o", model_params={"temperature": 0})
+
+# (Optional) Specify your own Neo4j schema
+# (also see get_structured_schema and get_schema functions)
+neo4j_schema = """
+Node properties:
+User {name: STRING}
+Person {name: STRING, born: INTEGER}
+Movie {tagline: STRING, title: STRING, released: INTEGER}
+Relationship properties:
+ACTED_IN {roles: LIST}
+DIRECTED {}
+REVIEWED {summary: STRING, rating: INTEGER}
+The relationships:
+(:Person)-[:ACTED_IN]->(:Movie)
+(:Person)-[:DIRECTED]->(:Movie)
+(:User)-[:REVIEWED]->(:Movie)
+"""
+
+prompt = """Task: Generate a Cypher statement for querying a Neo4j graph database from a user input.
+
+Do not use any properties or relationships not included in the schema.
+Do not include triple backticks ``` or any additional text except the generated Cypher statement in your response.
+
+Always filter movies that have not already been reviewed by the user with name: '{user_name}' using for instance:
+(m:Movie)<-[:REVIEWED]-(:User {{name: <the_user_name>}})
+
+Schema:
+{schema}
+
+Input:
+{query_text}
+
+Cypher query:
+"""
+
+with neo4j.GraphDatabase.driver(URI, auth=AUTH) as driver:
+    # Initialize the retriever
+    retriever = Text2CypherRetriever(
+        driver=driver,
+        llm=llm,
+        neo4j_schema=neo4j_schema,
+        # here we provide a custom prompt
+        custom_prompt=prompt,
+        neo4j_database=DATABASE,
+    )
+
+    # Generate a Cypher query using the LLM, send it to the Neo4j database, and return the results
+    query_text = "Which movies did Hugo Weaving star in?"
+    print(
+        retriever.search(
+            query_text=query_text,
+            prompt_params={
+                # you have to specify all placeholder except the {query_text} one
+                "schema": get_schema(driver),
+                "user_name": "the user asking question",
+            },
+        )
+    )

examples/retrieve/text2cypher_search.py

@@ -22,13 +22,11 @@
 Movie {tagline: STRING, title: STRING, released: INTEGER}
 Relationship properties:
 ACTED_IN {roles: LIST}
+DIRECTED {}
 REVIEWED {summary: STRING, rating: INTEGER}
 The relationships:
 (:Person)-[:ACTED_IN]->(:Movie)
 (:Person)-[:DIRECTED]->(:Movie)
-(:Person)-[:PRODUCED]->(:Movie)
-(:Person)-[:WROTE]->(:Movie)
-(:Person)-[:FOLLOWS]->(:Person)
 (:Person)-[:REVIEWED]->(:Movie)
 """
 

src/neo4j_graphrag/retrievers/text2cypher.py

@@ -48,7 +48,7 @@ class Text2CypherRetriever(Retriever):
     """
     Allows for the retrieval of records from a Neo4j database using natural language.
     Converts a user's natural language query to a Cypher query using an LLM,
-    then retrieves records from a Neo4j database using the generated Cypher query
+    then retrieves records from a Neo4j database using the generated Cypher query.
 
     Args:
         driver (neo4j.Driver): The Neo4j Python driver.
@@ -98,23 +98,23 @@ def __init__(
         self.examples = validated_data.examples
         self.result_formatter = validated_data.result_formatter
         self.custom_prompt = validated_data.custom_prompt
-        try:
+        if validated_data.custom_prompt:
+            neo4j_schema = ""
+        else:
             if (
-                not validated_data.custom_prompt
-            ):  # don't need schema for a custom prompt
-                self.neo4j_schema = (
-                    validated_data.neo4j_schema_model.neo4j_schema
-                    if validated_data.neo4j_schema_model
-                    else get_schema(validated_data.driver_model.driver)
-                )
+                validated_data.neo4j_schema_model
+                and validated_data.neo4j_schema_model.neo4j_schema
+            ):
+                neo4j_schema = validated_data.neo4j_schema_model.neo4j_schema
             else:
-                self.neo4j_schema = ""
-
-        except (Neo4jError, DriverError) as e:
-            error_message = getattr(e, "message", str(e))
-            raise SchemaFetchError(
-                f"Failed to fetch schema for Text2CypherRetriever: {error_message}"
-            ) from e
+                try:
+                    neo4j_schema = get_schema(validated_data.driver_model.driver)
+                except (Neo4jError, DriverError) as e:
+                    error_message = getattr(e, "message", str(e))
+                    raise SchemaFetchError(
+                        f"Failed to fetch schema for Text2CypherRetriever: {error_message}"
+                    ) from e
+        self.neo4j_schema = neo4j_schema
 
     def get_search_results(
         self, query_text: str, prompt_params: Optional[Dict[str, Any]] = None
@@ -142,12 +142,10 @@ def get_search_results(
 
         if prompt_params is not None:
             # parse the schema and examples inputs
-            examples_to_use = prompt_params.get("examples") or (
+            examples_to_use = prompt_params.pop("examples", None) or (
                 "\n".join(self.examples) if self.examples else ""
             )
-            schema_to_use = prompt_params.get("schema") or self.neo4j_schema
-            prompt_params.pop("examples", None)
-            prompt_params.pop("schema", None)
+            schema_to_use = prompt_params.pop("schema", None) or self.neo4j_schema
         else:
             examples_to_use = "\n".join(self.examples) if self.examples else ""
             schema_to_use = self.neo4j_schema

tests/unit/retrievers/test_text2cypher.py

@@ -19,6 +19,7 @@
 from neo4j.exceptions import CypherSyntaxError, Neo4jError
 from neo4j_graphrag.exceptions import (
     RetrieverInitializationError,
+    SchemaFetchError,
     SearchValidationError,
     Text2CypherRetrievalError,
 )
@@ -39,8 +40,8 @@ def test_t2c_retriever_initialization(driver: MagicMock, llm: MagicMock) -> None
 @patch("neo4j_graphrag.retrievers.base.Retriever._verify_version")
 @patch("neo4j_graphrag.retrievers.text2cypher.get_schema")
 def test_t2c_retriever_schema_retrieval(
-    _verify_version_mock: MagicMock,
     get_schema_mock: MagicMock,
+    _verify_version_mock: MagicMock,
     driver: MagicMock,
     llm: MagicMock,
 ) -> None:
@@ -51,13 +52,13 @@ def test_t2c_retriever_schema_retrieval(
 @patch("neo4j_graphrag.retrievers.base.Retriever._verify_version")
 @patch("neo4j_graphrag.retrievers.text2cypher.get_schema")
 def test_t2c_retriever_schema_retrieval_failure(
-    _verify_version_mock: MagicMock,
     get_schema_mock: MagicMock,
+    _verify_version_mock: MagicMock,
     driver: MagicMock,
     llm: MagicMock,
 ) -> None:
     get_schema_mock.side_effect = Neo4jError
-    with pytest.raises(Neo4jError):
+    with pytest.raises(SchemaFetchError):
         Text2CypherRetriever(driver, llm)
 
 
@@ -310,3 +311,31 @@ def test_t2c_retriever_with_custom_prompt_bad_prompt_params(
     llm.invoke.assert_called_once_with(
         """This is a custom prompt. test ['example A', 'example B']"""
     )
+
+
+@patch("neo4j_graphrag.retrievers.base.Retriever._verify_version")
+@patch("neo4j_graphrag.retrievers.text2cypher.get_schema")
+def test_t2c_retriever_with_custom_prompt_and_schema(
+    get_schema_mock: MagicMock,
+    _verify_version_mock: MagicMock,
+    driver: MagicMock,
+    llm: MagicMock,
+    neo4j_record: MagicMock,
+) -> None:
+    prompt = "This is a custom prompt. {query_text} {schema}"
+    query = "test"
+
+    driver.execute_query.return_value = (
+        [neo4j_record],
+        None,
+        None,
+    )
+
+    retriever = Text2CypherRetriever(driver=driver, llm=llm, custom_prompt=prompt)
+    retriever.search(
+        query_text=query,
+        prompt_params={},
+    )
+
+    get_schema_mock.assert_not_called()
+    llm.invoke.assert_called_once_with("""This is a custom prompt. test """)