Zǻi$ddlmZddlZddlZddlmZmZmZmZddl Z ddl m Z m Z m Z ddlmZddlmZmZmZmZddlmZddlmZdd lmZdd lmZdd lmZmZm Z m!Z!m"Z"m#Z#m$Z$ejJe&Z'dd Z(Gd deZ)y)) annotationsN)AnyCallableDictOptional)CypherSyntaxError DriverError Neo4jError)ValidationError)RetrieverInitializationErrorSchemaFetchErrorSearchValidationErrorText2CypherRetrievalError)Text2CypherTemplate) LLMInterface) Retriever) get_schema)LLMModelNeo4jDriverModelNeo4jSchemaModelRawSearchResultRetrieverResultItemText2CypherRetrieverModelText2CypherSearchModelcd}tj||tj}|r|dn|}tjdd|}tjdd|}tjdd|}|S)aExtract and format Cypher query from text, handling code blocks and special characters. This function performs two main operations: 1. Extracts Cypher code from within triple backticks (```), if present 2. Automatically adds backtick quotes around multi-word identifiers: - Node labels (e.g., ":Data Science" becomes ":`Data Science`") - Property keys (e.g., "first name:" becomes "`first name`:") - Relationship types (e.g., "[:WORKS WITH]" becomes "[:`WORKS WITH`]") Args: text (str): Raw text that may contain Cypher code, either within triple backticks or as plain text. Returns: str: Properly formatted Cypher query with correct backtick quoting. z ```(.*?)```rzB:\s*(?!`\s*)(\s*)([a-zA-Z0-9_]+(?:\s+[a-zA-Z0-9_]+)+)(?!\s*`)(\s*)z:`\2`z=([,{]\s*)(?!`)([a-zA-Z0-9_]+(?:\s+[a-zA-Z0-9_]+)+)(?!`)(\s*:)z\1`\2`\3zV(\[\s*[a-zA-Z0-9_]*\s*:\s*)(?!`)([a-zA-Z0-9_]+(?:\s+[a-zA-Z0-9_]+)+)(?!`)(\s*(?:\]|-)))refindallDOTALLsub)textpatternmatches cypher_querys c/opt/lhia/marcimex/agent/venv/lib/python3.12/site-packages/neo4j_graphrag/retrievers/text2cypher.pyextract_cypherr%0s}$Gjj$ 2G!(71:dL66ML 66HL 66aL cdeZdZdZ d dfd Z d ddZxZS)Text2CypherRetrievera 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. Args: driver (neo4j.Driver): The Neo4j Python driver. llm (neo4j_graphrag.generation.llm.LLMInterface): LLM object to generate the Cypher query. neo4j_schema (Optional[str]): Neo4j schema used to generate the Cypher query. examples (Optional[list[str], optional): Optional user input/query pairs for the LLM to use as examples. custom_prompt (Optional[str]): Optional custom prompt to use instead of auto generated prompt. Will include the neo4j_schema for schema and examples for examples prompt parameters, if they are provided. Raises: RetrieverInitializationError: If validation of the input arguments fail. c  t|}t|} |r t|nd} t|| | ||||} t|!| jj| j| jj|_ | j|_| j|_| j |_| j rJ| j"r4| j"j$r| j"j$}||_yd}||_y| j"r4| j"j$r| j"j$}||_y t'| jj}||_y#t$r} t | j | d} ~ wwxYw#t(t*f$r*} t-| dt/| } t1d| | d} ~ wwxYw)N)driver)llm) neo4j_schema) driver_model llm_modelneo4j_schema_modelexamplesresult_formatter custom_promptneo4j_databasemessagez1Failed to fetch schema for Text2CypherRetriever: )rrrrr r errorssuper__init__r-r*r3r.r+r0r1r2r/r,rr r getattrstrr )selfr*r+r,r0r1r2r3r-r.r/validated_datae error_message __class__s r$r8zText2CypherRetriever.__init__ks B+6:L S)I?K l;QU 7)##5!!1+-N   ' ' . .0M0M "++//&// . ? ?+99  ' '11"55BB-@@MM ") " )11"55BB-@@MM )#-n.I.I.P.P#QL )A B.qxxz: A B6#K0$+Ay#a&$AM*KM?[s/:E7F"7 FFF"G1%GGc t|}t |j }|^|j ddxs)|jrdj|jnd}|j ddxs |j}n?|jrdj|jnd}|j}t}|jd|||jd|}tjd | |jj!|} t#| j$} tjd | |j&j)| |j*t,j.j0 \} } } t9| d | iS#t$r}t|j|d}~wwxYw#t2$r}t5d |j6|d}~wwxYw)aZConverts query_text to a Cypher query using an LLM. Retrieve records from a Neo4j database using the generated Cypher query. Args: query_text (str): The natural language query used to search the Neo4j database. prompt_params (Dict[str, Any]): additional values to inject into the custom prompt, if it is provided. If the schema or examples parameter is specified, it will overwrite the corresponding value passed during initialization. Example: {'schema': 'this is the graph schema'} Raises: SearchValidationError: If validation of the input arguments fail. Text2CypherRetrievalError: If the LLM fails to generate a correct Cypher query. Returns: RawSearchResult: The results of the search query as a list of neo4j.Record and an optional metadata dict ) query_textN)templater0 r4schema)rDr0rAzText2CypherRetriever prompt: %sz%Text2CypherRetriever Cypher query: %s)query_ database_routing_zFailed to get search result: cypher)recordsmetadata)rr rr6rr2popr0joinr,dictformatrAloggerdebugr+invoker%contentr* execute_queryr3neo4jRoutingControlREADrrr5r) r;rA prompt_paramsr<r=prompt_templateexamples_to_use schema_to_useprompt llm_result t2c_queryrI_s r$get_search_resultsz'Text2CypherRetriever.get_search_resultss" ;3zJN.t7I7IJ  $+// DA,0MM $--(r *--h=RARARM:>--dii 6RO --M FM'''  $%00     6? 0J&z'9'9:I LL@) L KK55 ----226MGQ)  O ;' 3 : ;D! +/ {;  s0 F:B F> F;F66F;> G%G  G%)NNNNN)r*z neo4j.Driverr+rr, Optional[str]r0zOptional[list[str]]r1z7Optional[Callable[[neo4j.Record], RetrieverResultItem]]r2rar3rareturnNone)N)rAr:rXzOptional[Dict[str, Any]]rbr)__name__ __module__ __qualname____doc__r8r` __classcell__)r?s@r$r(r(Zs('+(, '+(,;););)$ ;) & ;)  ;)%;)&;) ;)|JN? ? .F? ? r&r()r r:rbr:)* __future__rloggingrtypingrrrrrUneo4j.exceptionsrr r pydanticr neo4j_graphrag.exceptionsr r rr!neo4j_graphrag.generation.promptsrneo4j_graphrag.llmrneo4j_graphrag.retrievers.baserneo4j_graphrag.schemarneo4j_graphrag.typesrrrrrrr getLoggerrdrPr%r(rKr&r$rusr# 00 GG$ B+4,   8 $'TM 9M r&