3fiX5ddlmZddlZddlmZmZmZmZmZm Z m Z m Z m Z m Z mZddlZddlmZddlmZddlmZddlmZddlmZerdd lmZe d eeef Zej>e Z!d Z"ed ddGddeZ#y)) annotationsN) TYPE_CHECKINGAnyCallableDict GeneratorIterableListOptionalTupleTypeVarUnion) deprecated)Document) Embeddings) VectorStore)maximal_marginal_relevance) CollectionMongoDBDocumentType)bounddz0.0.25z1.0z*langchain_mongodb.MongoDBAtlasVectorSearch)sinceremovalalternative_importcfeZdZdZddddd ddZeddZdd Ze dd Z d dd Z dd Z d ddZ d ddZ d ddZ d ddZe d d!dZy )"MongoDBAtlasVectorSearcha`MongoDB Atlas Vector Search` vector store. To use, you should have both: - the ``pymongo`` python package installed - a connection string associated with a MongoDB Atlas Cluster having deployed an Atlas Search index Example: .. code-block:: python from langchain_community.vectorstores import MongoDBAtlasVectorSearch from langchain_community.embeddings.openai import OpenAIEmbeddings from pymongo import MongoClient mongo_client = MongoClient("") collection = mongo_client[""][""] embeddings = OpenAIEmbeddings() vectorstore = MongoDBAtlasVectorSearch(collection, embeddings) defaulttext embeddingcosine) index_nametext_key embedding_keyrelevance_score_fncX||_||_||_||_||_||_y)a Args: collection: MongoDB collection to add the texts to. embedding: Text embedding model to use. text_key: MongoDB field that will contain the text for each document. embedding_key: MongoDB field that will contain the embedding for each document. index_name: Name of the Atlas Search index. relevance_score_fn: The similarity score used for the index. Currently supported: Euclidean, cosine, and dot product. N) _collection _embedding _index_name _text_key_embedding_key_relevance_score_fn)self collectionrr!r"r#r$s l/var/www/auto_recruiter/arenv/lib/python3.12/site-packages/langchain_community/vectorstores/mongodb_atlas.py__init__z!MongoDBAtlasVectorSearch.__init__>s2,&#%!+#5 c|jSN)r'r,s r. embeddingsz#MongoDBAtlasVectorSearch.embeddings[s r0c|jdk(r |jS|jdk(r |jS|jdk(r |jSt d|j)N euclidean dotProductr z!No relevance score function for $)r+_euclidean_relevance_score_fn%_max_inner_product_relevance_score_fn_cosine_relevance_score_fnNotImplementedErrorr3s r._select_relevance_score_fnz3MongoDBAtlasVectorSearch._select_relevance_score_fn_sn  # #{ 255 5  % % 5== =  % % 122 2%3D4L4L3MN r0c  ddlm}ddlm}ddlm}|||d|d }|jd \} } || | } || |fi|S#t $r t dwxYw) aConstruct a `MongoDB Atlas Vector Search` vector store from a MongoDB connection URI. Args: connection_string: A valid MongoDB connection URI. namespace: A valid MongoDB namespace (database and collection). embedding: The text embedding model to use for the vector store. Returns: A new MongoDBAtlasVectorSearch instance. r)version) MongoClient) DriverInfozGCould not import pymongo, please install it with `pip install pymongo`. Langchain langchain)namer>)driver.)importlib.metadatar>pymongor?pymongo.driver_infor@ ImportErrorsplit) clsconnection_string namespacerkwargsr>r?r@clientdb_namecollection_namer-s r.from_connection_stringz/MongoDBAtlasVectorSearch.from_connection_stringks(  2 + 6 * ; 8LM $-??3#7 G__5 :y3F33 )  s AA'Nc |jdt}|xs d|D}g}g}g}tt||D][\} \} } |j | |j | | dz|zdk(s7|j |j ||g}g}]|r!|j |j |||S)a>Run 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. Returns: List of ids from adding the texts into the vectorstore. batch_sizec3 K|]}iywr2).0_s r. z5MongoDBAtlasVectorSearch.add_texts..s:M!2:Ms r)getDEFAULT_INSERT_BATCH_SIZE enumeratezipappendextend _insert_texts) r,texts metadatasrNrT _metadatas texts_batchmetadatas_batch result_idsirmetadatas r. add_textsz"MongoDBAtlasVectorSearch.add_textssZZ .GH -6-M:Mu:M   #,S -C#D % Ah   t $  " "8 ,A#q(!!$"4"4[/"RS "$  %    d00oN Or0c |sgS|jj|}t|||Dcgc]"\}}}|j||j|i|$}}}}|j j |}|jScc}}}wr2)r'embed_documentsr^r)r*r& insert_many inserted_ids) r,rbrcr4tmr to_insert insert_results r.raz&MongoDBAtlasVectorSearch._insert_textssI__44U; $'ui#D  1i^^Q 3 3Y D! D  ((44Y? )))  s'Bcp||j|dz||jd}|r||d<d|i}|ddddiiig}||j||jj |}g} |D]L} | j |j } | j d} | jt| | | fN| S) N ) queryVectorpath numCandidateslimitindexfilterz $vectorSearchz$setscorez$metavectorSearchScore) page_contentri) r*r(r`r& aggregatepopr)r_r) r,rk pre_filterpost_filter_pipelineparamsquerypipelinecursordocsresrr{s r._similarity_search_with_scorez6MongoDBAtlasVectorSearch._similarity_search_with_scores%''V%%   )F8  &)  g)<=> ?  + OO0 1!!++H5 LC774>>*DGGG$E KKtcBEJ K L r0cf|jj|}|j||||}|S)aReturn MongoDB documents most similar to the given query and their scores. Uses the vectorSearch operator available in MongoDB Atlas Search. For more: https://www.mongodb.com/docs/atlas/atlas-vector-search/vector-search-stage/ Args: query: Text to look up documents similar to. k: (Optional) number of documents to return. Defaults to 4. pre_filter: (Optional) dictionary of argument(s) to prefilter document fields on. post_filter_pipeline: (Optional) Pipeline of MongoDB aggregation stages following the vectorSearch stage. Returns: List of documents most similar to the query and their scores. rrr)r' embed_queryr)r,rrrrrrs r.similarity_search_with_scorez5MongoDBAtlasVectorSearch.similarity_search_with_scores@.OO//6 11 !!5 2   r0c |jd}|j||||}|rd|vr|D]\}} | |jd<|D cgc]\}} | c} }Scc} }w)aReturn MongoDB documents most similar to the given query. Uses the vectorSearch operator available in MongoDB Atlas Search. For more: https://www.mongodb.com/docs/atlas/atlas-vector-search/vector-search-stage/ Args: query: Text to look up documents similar to. k: (Optional) number of documents to return. Defaults to 4. pre_filter: (Optional) dictionary of argument(s) to prefilter document fields on. post_filter_pipeline: (Optional) Pipeline of MongoDB aggregation stages following the vectorSearch stage. Returns: List of documents most similar to the query and their scores. additionalrsimilarity_scorer{)r[rri) r,rrrrrNrdocs_and_scoresdocr{rXs r.similarity_searchz*MongoDBAtlasVectorSearch.similarity_searchsy0ZZ - ;; !!5 <  , :- . U(- W% ."12Q222s Ac >|jj|}|j||||} tt j || D cgc]\} } | j |j c} } ||} | D cgc] } | | d }} |Scc} } wcc} w)a Return documents 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: (Optional) number of documents to return. Defaults to 4. fetch_k: (Optional) number of documents to fetch before passing to MMR algorithm. Defaults to 20. 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. pre_filter: (Optional) dictionary of argument(s) to prefilter on document fields. post_filter_pipeline: (Optional) pipeline of MongoDB aggregation stages following the vectorSearch stage. Returns: List of documents selected by maximal marginal relevance. r)r lambda_multr)r'rrrnparrayrir*)r,rrfetch_krrrrNquery_embeddingrrrXmmr_doc_indexesrhmmr_docss r.max_marginal_relevance_searchz6MongoDBAtlasVectorSearch.max_marginal_relevance_search"s>//55e<11 !!5 2  5 HH_ %=A B63S\\$-- . B#   )881DGAJ88 C9s #B Bc Z| td|||fi|}|j|||S)aConstruct a `MongoDB Atlas Vector Search` vector store from raw documents. This is a user-friendly interface that: 1. Embeds documents. 2. Adds the documents to a provided MongoDB Atlas Vector Search index (Lucene) This is intended to be a quick way to get started. Example: .. code-block:: python from pymongo import MongoClient from langchain_community.vectorstores import MongoDBAtlasVectorSearch from langchain_community.embeddings import OpenAIEmbeddings mongo_client = MongoClient("") collection = mongo_client[""][""] embeddings = OpenAIEmbeddings() vectorstore = MongoDBAtlasVectorSearch.from_texts( texts, embeddings, metadatas=metadatas, collection=collection ) z*Must provide 'collection' named parameter.)rc) ValueErrorrj)rKrbrrcr-rN vectorstores r. from_textsz#MongoDBAtlasVectorSearch.from_textsQsAF  IJ J*i:6: ey9r0) r-zCollection[MongoDBDocumentType]rrr!strr"rr#rr$r)returnr)rzCallable[[float], float]) rLrrMrrrrNrrrr2)rbz Iterable[str]rczOptional[List[Dict[str, Any]]]rNrrr )rb List[str]rczList[Dict[str, Any]]rr )NN) rz List[float]rintrOptional[Dict]rOptional[List[Dict]]rList[Tuple[Document, float]]) rrrrrrrrrr) rrrrrrrrrNrrList[Document])rg?NN)rrrrrrrfloatrrrrrNrrr)NN) rbrrrrcrr-z)Optional[Collection[MongoDBDocumentType]]rNrrr)__name__ __module__ __qualname____doc__r/propertyr4r< classmethodrRrjrarrrrrrVr0r.rr$so 2$("*6366  6  66 6: #4#4#4 #4  #4 " #4#4P592   > * %)59  #  3  & F%)59  #  3  & F%)59 #3#3 #3# #3 3 #3  #3 #3P %)59-- - -  - # -3-- -^ +/@D &&&( & > &  & "&&r0r)$ __future__rloggingtypingrrrrrr r r r r rnumpyrlangchain_core._api.deprecationrlangchain_core.documentsrlangchain_core.embeddingsrlangchain_core.vectorstoresr&langchain_community.vectorstores.utilsrpymongo.collectionrrr getLoggerrloggerr\rrVr0r.rs"    6-03M-34S>J   8 $  C O{O  Or0