a bgJ5@sddlmZddlZddlZddlmZmZmZmZm Z m Z m Z ddl m Z ddlmZddlmZddlmZdZd ZGd d d eZdS) ) annotationsN)AnyDictIterableListOptionalTupleTypeDocument) Embeddings)get_from_dict_or_env) VectorStoreiZlangchain_pg_embeddingc @seZdZdZeeddfddddddd d d d Zd d ddZedd ddZ e ddeedfddddddddddd ddZ dddddd dddZ dHddddddd d!Z dIddd#dd$d%d&d'ZdJd(dd#dd$d)d*d+ZdKddd#d,d-d.d/ZdLd(dd#d,d0d1d2Ze deeddfd3dddddddddd4 d5d6Ze deeddfd7ddddddddd8 d9d:Ze eedfd3ddddddd;ddd?d@dAZe eeddfd3d$ddddddddB dCdDZe dddddddEdFdGZdS)MHologresa`Hologres API` vector store. - `connection_string` is a hologres connection string. - `embedding_function` any embedding function implementing `langchain.embeddings.base.Embeddings` interface. - `ndims` is the number of dimensions of the embedding output. - `table_name` is the name of the table to store embeddings and data. (default: langchain_pg_embedding) - NOTE: The table will be created when initializing the store (if not exists) So, make sure the user has the right permissions to create tables. - `pre_delete_table` if True, will delete the table if it exists. (default: False) - Useful for testing. FNstrr intboolzOptional[logging.Logger]None)connection_stringembedding_functionndims table_namepre_delete_tableloggerreturncCs:||_||_||_||_||_|p*tt|_| dSN) rrrrrlogging getLogger__name__r __post_init__)selfrrrrrrr!w/var/www/html/cobodadashboardai.evdpl.com/venv/lib/python3.9/site-packages/langchain_community/vectorstores/hologres.py__init__ s zHologres.__init__)rcCs0ddlm}||j|j|jddi|jd|_dS)z' Initialize the store. r)HologresVectordocumenttext)rrZ table_schemarN)Zhologres_vectorr$rrrrstorage)r r$r!r!r"r1s zHologres.__post_init__cCs|jSr)r)r r!r!r" embeddingsAszHologres.embeddingsz List[str]zList[List[float]]zOptional[List[dict]]zOptional[List[str]]r) textsr(r metadatasidsrrrkwargsrc Ksb|durdd|D}|s(dd|D}|| } || ||||d} | jf||||d| | S)NcSsg|]}ttqSr!ruuiduuid4.0_r!r!r" Sz#Hologres.__from..cSsg|]}iqSr!r!r0r!r!r"r3Vr4)rrrrr)r)r(r*r+)get_connection_stringadd_embeddings) clsr)r(rr*r+rrrr,rstorer!r!r"Z__fromEs$  zHologres.__fromz Iterable[str]z List[dict])r)r(r*r+r,rc KsZz$dd|D}|j||||Wn0tyT}z|j|WYd}~n d}~00dS)a/Add embeddings to the vectorstore. Args: texts: Iterable of strings to add to the vectorstore. embeddings: List of list of embedding vectors. metadatas: List of metadatas associated with the texts. kwargs: vectorstore specific parameters cSsg|] }d|iqS)r%r!r1tr!r!r"r3yr4z+Hologres.add_embeddings..N)r'Zupsert_vectors Exceptionr exception)r r)r(r*r+r,Z schema_dataser!r!r"r6hs zHologres.add_embeddings)r)r*r+r,rcKsT|durdd|D}|jt|}|s8dd|D}|j||||fi||S)arRun more texts through the embeddings and add to the vectorstore. Args: texts: Iterable of strings to add to the vectorstore. metadatas: Optional list of metadatas associated with the texts. kwargs: vectorstore specific parameters Returns: List of ids from adding the texts into the vectorstore. NcSsg|]}ttqSr!r-r0r!r!r"r3r4z&Hologres.add_texts..cSsg|]}iqSr!r!r0r!r!r"r3r4)rembed_documentslistr6)r r)r*r+r,r(r!r!r" add_texts~szHologres.add_textszOptional[dict]zList[Document])querykfilterr,rcKs|jj|d}|j|||dS)a^Run similarity search with Hologres with distance. Args: query (str): Query text to search for. k (int): Number of results to return. Defaults to 4. filter (Optional[Dict[str, str]]): Filter by metadata. Defaults to None. Returns: List of Documents most similar to the query. )r& embeddingrCrD)r embed_querysimilarity_search_by_vector)r rBrCrDr,rFr!r!r"similarity_searchs zHologres.similarity_searchz List[float])rFrCrDr,rcKs|j|||d}dd|DS)akReturn docs most similar to embedding vector. Args: embedding: Embedding to look up documents similar to. k: Number of Documents to return. Defaults to 4. filter (Optional[Dict[str, str]]): Filter by metadata. Defaults to None. Returns: List of Documents most similar to the query vector. rEcSsg|] \}}|qSr!r!)r1docr2r!r!r"r3r4z8Hologres.similarity_search_by_vector..)&similarity_search_with_score_by_vector)r rFrCrDr,Zdocs_and_scoresr!r!r"rHsz$Hologres.similarity_search_by_vectorzList[Tuple[Document, float]])rBrCrDrcCs |j|}|j|||d}|S)abReturn docs most similar to query. Args: query: Text to look up documents similar to. k: Number of Documents to return. Defaults to 4. filter (Optional[Dict[str, str]]): Filter by metadata. Defaults to None. Returns: List of Documents most similar to the query and score for each rE)rrGrK)r rBrCrDrFdocsr!r!r"similarity_search_with_scores  z%Hologres.similarity_search_with_score)rFrCrDrcCs(|jj||dg|d}dd|D}|S)Nr%)rCZselect_columnsZmetadata_filterscSs(g|] }t|d|dd|dfqS)r%metadata) page_contentrNZdistancer )r1resultr!r!r"r3szCHologres.similarity_search_with_score_by_vector..)r'search)r rFrCrDresultsrLr!r!r"rKs  z/Hologres.similarity_search_with_score_by_vectorzType[Hologres]) r7r)rFr*rrr+rr,rc Ks0|t|} |j|| |f|||||d|S)aM Return VectorStore initialized from texts and embeddings. Hologres connection string is required "Either pass it as a parameter or set the HOLOGRES_CONNECTION_STRING environment variable. Create the connection string by calling HologresVector.connection_string_from_db_params r*r+rrr)r>r?_Hologres__from) r7r)rFr*rrr+rr,r(r!r!r" from_textss zHologres.from_textszList[Tuple[str, List[float]]]) text_embeddingsrFr*rrr+rr,rc Ks>dd|D} dd|D} |j| | |f|||||d|S)a~Construct Hologres wrapper from raw documents and pre- generated embeddings. Return VectorStore initialized from documents and embeddings. Hologres connection string is required "Either pass it as a parameter or set the HOLOGRES_CONNECTION_STRING environment variable. Create the connection string by calling HologresVector.connection_string_from_db_params Example: .. code-block:: python from langchain_community.vectorstores import Hologres from langchain_community.embeddings import OpenAIEmbeddings embeddings = OpenAIEmbeddings() text_embeddings = embeddings.embed_documents(texts) text_embedding_pairs = list(zip(texts, text_embeddings)) faiss = Hologres.from_embeddings(text_embedding_pairs, embeddings) cSsg|] }|dqS)rr!r9r!r!r"r37r4z,Hologres.from_embeddings..cSsg|] }|dqS)r!r9r!r!r"r38r4rS)rT) r7rVrFr*rrr+rr,r)r(r!r!r"from_embeddingss  zHologres.from_embeddings)r7rFrrrr,rcKs ||}||||||d}|S)z Get instance of an existing Hologres store.This method will return the instance of the store without inserting any new embeddings )rrrrr)r5)r7rFrrrr,rr8r!r!r"from_existing_indexFs zHologres.from_existing_indexzDict[str, Any])r,rcCst|ddd}|std|S)NrZHOLOGRES_CONNECTION_STRING)datakeyZenv_keyzHologres connection string is requiredEither pass it as a parameteror set the HOLOGRES_CONNECTION_STRING environment variable.Create the connection string by callingHologresVector.connection_string_from_db_params)r ValueError)r7r,rr!r!r"r5aszHologres.get_connection_string) r7 documentsrFrrr+pre_delete_collectionr,rc KsNdd|D}dd|D} ||} | |d<|jf|||| |||d|S)aQ Return VectorStore initialized from documents and embeddings. Hologres connection string is required "Either pass it as a parameter or set the HOLOGRES_CONNECTION_STRING environment variable. Create the connection string by calling HologresVector.connection_string_from_db_params cSsg|] }|jqSr!)rOr1dr!r!r"r3r4z+Hologres.from_documents..cSsg|] }|jqSr!)rNr_r!r!r"r3r4r)r)r^rFr*r+rr)r5rU) r7r]rFrrr+r^r,r)r*rr!r!r"from_documentsts zHologres.from_documents)hostportdatabaseuserpasswordrc Cs"d|d|d|d|d| S)z2Return connection string from database parameters.zdbname=z user=z password=z host=z port=r!)r7rbrcrdrerfr!r!r" connection_string_from_db_paramss z)Hologres.connection_string_from_db_params)NN)rAN)rAN)rAN)rAN)r __module__ __qualname____doc__ADA_TOKEN_COUNT_LANGCHAIN_DEFAULT_TABLE_NAMEr#rpropertyr( classmethodrTr6r@rIrHrMrKrUrXrYr5rargr!r!r!r"rsz$" $!"."$r) __future__rrr.typingrrrrrrr Zlangchain_core.documentsr Zlangchain_core.embeddingsr Zlangchain_core.utilsr Zlangchain_core.vectorstoresrrkrlrr!r!r!r"s $