o ZhZ@sddlmZddlZddlZddlmZmZddlmZm Z m Z m Z m Z m Z mZddlmZddlmZddlmZerHddlmZddlmZeZeGd d d ZGd d d eZdS) ) annotationsN)asdict dataclass) TYPE_CHECKINGAnyDictIterableListOptionalTupleDocument) Embeddings) VectorStore)DocumentCollectionc@s:eZdZUdZded<ded<ded<ded <d ed <d S) CollectionConfigaConfiguration for a `Zep Collection`. If the collection does not exist, it will be created. Attributes: name (str): The name of the collection. description (Optional[str]): An optional description of the collection. metadata (Optional[Dict[str, Any]]): Optional metadata for the collection. embedding_dimensions (int): The number of dimensions for the embeddings in the collection. This should match the Zep server configuration if auto-embed is true. is_auto_embedded (bool): A flag indicating whether the collection is automatically embedded by Zep. strname Optional[str] descriptionOptional[Dict[str, Any]]metadataintembedding_dimensionsboolis_auto_embeddedN)__name__ __module__ __qualname____doc____annotations__r!r![/var/www/html/lang_env/lib/python3.10/site-packages/langchain_community/vectorstores/zep.pyrs  rcseZdZdZdddddZfddZed[ddZd\ddZd\ddZ  d]d^dd Z  d]d_d%d&Z  d]d_d'd(Z  )d`dad1d2Z  )d`dad3d4Z 5 dbdcd6d7Z 5 dbddd9d:Z 5 dbddd;d<Z 5 dbddd=d>Z 5 dbdcd?d@Z 5 dbdedBdCZ 5 dbdedDdEZ 5 F G dfdgdKdLZ 5 F G dfdgdMdNZ 5 F G dfdhdOdPZ 5 F G dfdhdQdRZe   S S  didjdUdVZdkdldXdYZZS)mZepVectorStorea`Zep` vector store. It provides methods for adding texts or documents to the store, searching for similar documents, and deleting documents. Search scores are calculated using cosine similarity normalized to [0, 1]. Args: api_url (str): The URL of the Zep API. collection_name (str): The name of the collection in the Zep store. api_key (Optional[str]): The API key for the Zep API. config (Optional[CollectionConfig]): The configuration for the collection. Required if the collection does not already exist. embedding (Optional[Embeddings]): Optional embedding function to use to embed the texts. Required if the collection is not auto-embedded. Napi_keyconfig embeddingcollection_namerapi_urlr%rr&Optional[CollectionConfig]r'Optional[Embeddings]returnNonecst|s tdzddlm}Wn tytdw|||d|_||_|r5|j|jkr5|j|_||_ | |_ ||_ dS)Nzs&    zZepVectorStore.__init__cCs|jS)z/Access the query embedding object if available.)r8)r9r!r!r" embeddingsaszZepVectorStore.embeddingsrcCsTddlm}z |jj|j}W|S|y)td|jd|}Y|Sw)z; Load the collection from the Zep backend. r) NotFoundErrorz Collection z$ not found. Creating new collection.) r2r=r4documentZget_collectionr(loggerinfo_create_collection)r9r= collectionr!r!r"r6fs    zZepVectorStore._load_collectioncCs,|jstd|jjjdit|j}|S)z= Create a new collection in the Zep backend. zCCollection config must be specified when creating a new collection.Nr!)r5r1r4r>Zadd_collectionr)r9rBr!r!r"rAvs z!ZepVectorStore._create_collectiontexts Iterable[str] metadatasOptional[List[Dict[Any, Any]]] document_idsOptional[List[str]]List[ZepDocument]c Csddlm}d}|jr|jjr|jdurtjdddn,|jdurG|jt|}|jrF|jj t |dkrFt d|jj dt |dn g}t |D]!\}}| |||r\||nd|rc||nd|rj||nddqN|S) Nrr z{The collection is set to auto-embed and an embedding function is present. Ignoring the embedding function.) stacklevelzkThe embedding dimensions of the collection and the embedding function do not match. Collection dimensions: z, Embedding dimensions: )contentrZ document_idr')zep_python.documentr r7rr8warningswarnZembed_documentslistrlenr1 enumerateappend) r9rCrErG ZepDocumentr< documentsidr!r!r"_generate_documents_to_addsD    z)ZepVectorStore._generate_documents_to_addOptional[List[Dict[str, Any]]]kwargsr List[str]cKs,|jstd||||}|j|}|S)aRun more texts through the embeddings and add to the vectorstore. Args: texts: Iterable of strings to add to the vectorstore. metadatas: Optional list of metadatas associated with the texts. document_ids: Optional list of document ids associated with the texts. kwargs: vectorstore specific parameters Returns: List of ids from adding the texts into the vectorstore. Return docs most similar to query using specified search type. similarityrcrmmrsearch_type of ? not allowed. Expected search_type to be 'similarity' or 'mmr'.)similarity_searchmax_marginal_relevance_searchr1r9rarbrrcrZr!r!r"searchs  zZepVectorStore.searchcs^|dkr|j|f||d|IdHS|dkr'|j|f||d|IdHStd|d)rerfrgNrhrirj)asimilarity_searchamax_marginal_relevance_searchr1rmr!r!r"asearchs(    zZepVectorStore.asearchcKs&|j|f||d|}dd|DS)"Return docs most similar to query.rgcSg|]\}}|qSr!r!.0doc_r!r!r" z4ZepVectorStore.similarity_search..(_similarity_search_with_relevance_scoresr9rarcrrZresultsr!r!r"rks z ZepVectorStore.similarity_searchList[Tuple[Document, float]]cKs|j|f||d|S)z$Run similarity search with distance.rgr{)r9rarcrrZr!r!r"similarity_search_with_scores z+ZepVectorStore.similarity_search_with_scorecKsl|jstd|jjs"|jr"|j|}|jjd|||d|}n |jj|f||d|}dd|DS)a Default similarity search with relevance scores. Modify if necessary in subclass. Return docs and relevance scores in the range [0, 1]. 0 is dissimilar, 1 is most similar. Args: query: input text k: Number of Documents to return. Defaults to 4. metadata: Optional, metadata filter **kwargs: kwargs to be passed to similarity search. Should include: score_threshold: Optional, a floating point value between 0 to 1 and filter the resulting set of retrieved docs Returns: List of Tuples of (doc, similarity_score) r\r'limitrrrcS&g|]}t|j|jd|jpdfqSZ page_contentrgr rLrZscorervrwr!r!r"ryKzKZepVectorStore._similarity_search_with_relevance_scores..Nr!)r7r1rr8 embed_queryrnr9rarcrrZ query_vectorr~r!r!r"r|"s* z7ZepVectorStore._similarity_search_with_relevance_scorescsz|jstd|jjs&|jr&|j|}|jjd|||d|IdH}n|jj|f||d|IdH}dd|DS)rsr\rNrcSrrrrr!r!r"rynrzKZepVectorStore.asimilarity_search_with_relevance_scores..r!)r7r1rr8rrqrr!r!r"(asimilarity_search_with_relevance_scoresVs,   z7ZepVectorStore.asimilarity_search_with_relevance_scorescs.|j||fd|i|IdH}dd|DS)rsrNcSrtr!r!rur!r!r"ryrzz5ZepVectorStore.asimilarity_search..)rr}r!r!r"roys  z!ZepVectorStore.asimilarity_search List[float]cKs6|jstd|jjd|||d|}dd|DS)aFReturn docs most similar to embedding vector. Args: embedding: Embedding to look up documents similar to. k: Number of Documents to return. Defaults to 4. metadata: Optional, metadata filter Returns: List of Documents most similar to the query vector. r\rcSg|] }t|j|jdqSrr rLrrr!r!r"ryz>ZepVectorStore.similarity_search_by_vector..Nr!r7r1rnr9r'rcrrZr~r!r!r"similarity_search_by_vectorsz*ZepVectorStore.similarity_search_by_vectorcs8|jstd|jjd|||d|}dd|DS)z-Return docs most similar to embedding vector.r\rcSrrrrr!r!r"ryrz?ZepVectorStore.asimilarity_search_by_vector..Nr!rrr!r!r"asimilarity_search_by_vectorsz+ZepVectorStore.asimilarity_search_by_vector?fetch_k lambda_multfloatc Ksx|jstd|jjs$|jr$|j|}|jjd|||d|d|}n|jj|f||d|d|\}}dd|DS) aReturn docs selected using the maximal marginal relevance. Maximal marginal relevance optimizes for similarity to query AND diversity among selected documents. Args: query: Text to look up documents similar to. k: Number of Documents to return. Defaults to 4. fetch_k: Number of Documents to fetch to pass to MMR algorithm. Zep determines this automatically and this parameter is ignored. lambda_mult: Number between 0 and 1 that determines the degree of diversity among the results with 0 corresponding to maximum diversity and 1 to minimum diversity. Defaults to 0.5. metadata: Optional, metadata to filter the resulting set of retrieved docs Returns: List of Documents selected by maximal marginal relevance. r\rhr'rrrb mmr_lambdarrrbrcSrrrrvrWr!r!r"ryz@ZepVectorStore.max_marginal_relevance_search..Nr!)r7r1rr8rrnZsearch_return_query_vector r9rarcrrrrZrr~r!r!r"rls4    z,ZepVectorStore.max_marginal_relevance_searchc s|jstd|jjs(|jr(|j|}|jjd|||d|d|IdH}n|jj|f||d|d|IdH\}}dd|DS) :Return docs selected using the maximal marginal relevance.r\rhrNrcSrrrrr!r!r"ryrzAZepVectorStore.amax_marginal_relevance_search..r!)r7r1rr8rrqZasearch_return_query_vectorrr!r!r"rps6    z-ZepVectorStore.amax_marginal_relevance_searchcKs:|jstd|jjd|||d|d|}dd|DS)aReturn docs selected using the maximal marginal relevance. Maximal marginal relevance optimizes for similarity to query AND diversity among selected documents. Args: embedding: Embedding to look up documents similar to. k: Number of Documents to return. Defaults to 4. fetch_k: Number of Documents to fetch to pass to MMR algorithm. Zep determines this automatically and this parameter is ignored. lambda_mult: Number between 0 and 1 that determines the degree of diversity among the results with 0 corresponding to maximum diversity and 1 to minimum diversity. Defaults to 0.5. metadata: Optional, metadata to filter the resulting set of retrieved docs Returns: List of Documents selected by maximal marginal relevance. r\rhrcSrrrrr!r!r"ryKrzJZepVectorStore.max_marginal_relevance_search_by_vector..Nr!rr9r'rcrrrrZr~r!r!r"'max_marginal_relevance_search_by_vector!s z6ZepVectorStore.max_marginal_relevance_search_by_vectorcsB|jstd|jjd|||d|d|IdH}dd|DS)rr\rhrNcSrrrrr!r!r"ryerzKZepVectorStore.amax_marginal_relevance_search_by_vector..r!)r7r1rqrr!r!r"(amax_marginal_relevance_search_by_vectorMs   z7ZepVectorStore.amax_marginal_relevance_search_by_vectorOptional[List[dict]]c Ks"||||||d} | ||| S)a Class method that returns a ZepVectorStore instance initialized from texts. If the collection does not exist, it will be created. Args: texts (List[str]): The list of texts to add to the vectorstore. embedding (Optional[Embeddings]): Optional embedding function to use to embed the texts. metadatas (Optional[List[Dict[str, Any]]]): Optional list of metadata associated with the texts. collection_name (str): The name of the collection in the Zep store. api_url (str): The URL of the Zep API. api_key (Optional[str]): The API key for the Zep API. config (Optional[CollectionConfig]): The configuration for the collection. kwargs: Additional parameters specific to the vectorstore. Returns: ZepVectorStore: An instance of ZepVectorStore. r$)r^) clsrCr'rEr(r)r%r&rZZvecstorer!r!r" from_textsgs  zZepVectorStore.from_textsidscKsH|dus t|dkrtd|jdurtd|D]}|j|qdS)zDelete by Zep vector UUIDs. Parameters ---------- ids : Optional[List[str]] The UUIDs of the vectors to delete. Raises ------ ValueError If no UUIDs are provided. NrzNo uuids provided to delete.zNo collection name provided.)rQr1r7Zdelete_document)r9rrZur!r!r"deletes zZepVectorStore.delete) r(rr)rr%rr&r*r'r+r,r-)r,r+)r,r)NN)rCrDrErFrGrHr,rI) rCrDrErYrGrHrZrr,r[)Nr`) rarrbrrrrcrrZrr,rd)rrN) rarrcrrrrZrr,rd) rarrcrrrrZrr,r) r'rrcrrrrZrr,rd)rrrrN)rarrcrrrrrrrrZrr,rd)r'rrcrrrrrrrrZrr,rd)NNrrNN)rCr[r'r+rErr(rr)rr%rr&r*rZrr,r#)N)rrHrZrr,r-)rrrrr0propertyr<r6rArXr^r_rnrqrkrr|rrorrrlrprr classmethodrr __classcell__r!r!r:r"r#,s#    -       7 &  %  ; ) / )r#) __future__rloggingrN dataclassesrrtypingrrrrr r r Zlangchain_core.documentsr Zlangchain_core.embeddingsrZlangchain_core.vectorstoresrrMrTr getLoggerr?rr#r!r!r!r"s $