dh^0SSKJr SSKrSSKJr SSKJrJrJrJ r J r J r J r J r Jr SSKJr SSKJr \(a SSKJr SSKJr "S S \\5r\ "S \\\4S 9r\R4"\5rS r"SS\5rg)) annotationsN)Enum) TYPE_CHECKINGAnyDict GeneratorIterableListOptionalTypeVarUnion)Document) VectorStore) Embeddings) Collectionc(\rSrSrSrSrSrSrSrg)DocumentDBSimilarityTypez)DocumentDB Similarity Type as enumerator.cosine dotProduct euclideanN) __name__ __module__ __qualname____firstlineno____doc__COSDOTEUC__static_attributes__rc/var/www/html/shao/venv/lib/python3.13/site-packages/langchain_community/vectorstores/documentdb.pyrrs3 C C Cr"rDocumentDBDocumentType)boundc\rSrSrSrSSSS.SSjjr\SSj5rSS jr\ SS j5r SS jr S S jr S \ RSS4S!SjjrS"S#SjjrS$Sjr\ S%S&Sjj5rS"S'SjjrS"S(SjjrS)S*SjjrS+SS.S,SjjjrSrg)-DocumentDBVectorSearch,aD`Amazon DocumentDB (with MongoDB compatibility)` vector store. Please refer to the official Vector Search documentation for more details: https://docs.aws.amazon.com/documentdb/latest/developerguide/vector-search.html To use, you should have both: - the ``pymongo`` python package installed - a connection string and credentials associated with a DocumentDB cluster Example: . code-block:: python from langchain_community.vectorstores import DocumentDBVectorSearch from langchain_community.embeddings.openai import OpenAIEmbeddings from pymongo import MongoClient mongo_client = MongoClient("") collection = mongo_client[""][""] embeddings = OpenAIEmbeddings() vectorstore = DocumentDBVectorSearch(collection, embeddings) vectorSearchIndex textContent vectorContent) index_nametext_key embedding_keycjXlX lX0lX@lXPl[ R Ulg)aqConstructor for DocumentDBVectorSearch Args: collection: MongoDB collection to add the texts to. embedding: Text embedding model to use. index_name: Name of the Vector Search index. text_key: MongoDB field that will contain the text for each document. embedding_key: MongoDB field that will contain the embedding for each document. N) _collection _embedding _index_name _text_key_embedding_keyrr_similarity_type)self collection embeddingr-r.r/s r#__init__DocumentDBVectorSearch.__init__Bs.(&#%!+ 8 < <r"cUR$N)r2r7s r# embeddings!DocumentDBVectorSearch.embeddings]s r"cUR$)z=Returns the index name Returns: Returns the index name )r3r>s r#get_index_name%DocumentDBVectorSearch.get_index_nameasr"c SSKJn U"U5nURS5upxXgUn U"X40UD6$![a [S5ef=f)aFCreates an Instance of DocumentDBVectorSearch from a Connection String Args: connection_string: The DocumentDB cluster endpoint connection string namespace: The namespace (database.collection) embedding: The embedding utility **kwargs: Dynamic keyword arguments Returns: an instance of the vector store r) MongoClientzGCould not import pymongo, please install it with `pip install pymongo`..)pymongorE ImportErrorsplit) clsconnection_string namespacer9kwargsrEclientdb_namecollection_namer8s r#from_connection_string-DocumentDBVectorSearch.from_connection_stringjsg(  + **;<#,??3#7 __5 :3F33 )  s 3A cURR5nURnUHnURS5nXB:XdM g g)zVerifies if the specified index name during instance construction exists on the collection Returns: Returns True on success and False if no such index exists on the collection nameTF)r1 list_indexesr3pop)r7cursorr-rescurrent_index_names r# index_exists#DocumentDBVectorSearch.index_existssI!!..0%% C!$ !/ r"czUR5(a&URRUR5 gg)zEDeletes the index specified during instance construction if it existsN)rZr1 drop_indexr3r>s r# delete_index#DocumentDBVectorSearch.delete_indexs0         ' '(8(8 9 r"i@c X lURRURURS0SUUUUS.S./S.nURR nUR U5nU$)aXCreates an index using the index name specified at instance construction Args: dimensions: Number of dimensions for vector similarity. The maximum number of supported dimensions is 2000 similarity: Similarity algorithm to use with the HNSW index. Possible options are: - DocumentDBSimilarityType.COS (cosine distance), - DocumentDBSimilarityType.EUC (Euclidean distance), and - DocumentDBSimilarityType.DOT (dot product). m: Specifies the max number of connections for an HNSW index. Large impact on memory consumption. ef_construction: Specifies the size of the dynamic candidate list for constructing the graph for HNSW index. Higher values lead to more accurate results but slower indexing speed. Returns: An object describing the created index vectorhnsw)type similarity dimensionsmefConstruction)rTkey vectorOptions) createIndexesindexes)r6r1rTr3r5databasecommand)r7rgrfrhef_constructioncreate_index_commandscurrent_databasecreate_index_responsess r# create_index#DocumentDBVectorSearch.create_indexs@!+"--22!,, //: &&0&0*9 &  ! $ ++442B1I1I !2 &%r"Nc URS[5nU=(d SU5n/n/n/n[[X55H\un upUR U 5 UR U 5 U S-U-S:XdM8UR UR Xg55 /n/nM^ U(a UR UR Xg55 U$)N batch_sizec3$# UHn0v M g7fr=r).0_s r# 3DocumentDBVectorSearch.add_texts..s:Mu!2usr)getDEFAULT_INSERT_BATCH_SIZE enumeratezipappendextend _insert_texts) r7texts metadatasrMrw _metadatas texts_batchmetadatas_batch result_idsitextmetadatas r# add_texts DocumentDBVectorSearch.add_textss ZZ .GH -6-M:Mu:M   #,S-C#D A   t $  " "8 ,A#q(!!$"4"4["RS "$ $E    d00N Or"c U(d/$URRU5n[XU5VVVs/sH!upEnURX@RU0UEPM# nnnnUR R U5nUR$s snnnf)zUsed to Load Documents into the collection Args: texts: The list of documents strings to load metadatas: The list of metadata objects associated with each document Returns: )r2embed_documentsrr4r5r1 insert_many inserted_ids) r7rrr?trhr9 to_insert insert_results r#r$DocumentDBVectorSearch._insert_textssI__44U; $'u#D #Di^^Q 3 3Y D! D#D  ((44Y? )))  s'Bc RUc [S5eU"XB40UD6nURXS9 U$)Nz*Must provide 'collection' named parameter.)r) ValueErrorr)rJrr9rr8rM vectorstores r# from_texts!DocumentDBVectorSearch.from_textss<  IJ J*:6: e9r"c TUc [S5eUHnURU5 M g)Nz#No document ids provided to delete.T)rdelete_document_by_id)r7idsrM document_ids r#deleteDocumentDBVectorSearch.deletes. ;BC CK  & &{ 3r"cSSKJn Uc [S5eURR SU"U505 g![an[S5UeSnAff=f)zRRemoves a Specific Document by Id Args: document_id: The document identifier r)ObjectIdz>Unable to import bson, please install with `pip install bson`.Nz"No document id provided to delete._id) bson.objectidrrHrr1 delete_one)r7rres r#r,DocumentDBVectorSearch.delete_document_by_id&sa   .  AB B ##UH[,A$BC P  s: A AAc U(d0nSU0SSUURURUUS.00/nURRU5n/nUH6nUR UR 5n UR [XS95 M8 U$)aReturns a list of documents. Args: embeddings: The query vector k: the number of documents to return ef_search: Specifies the size of the dynamic candidate list that HNSW index uses during search. A higher value of efSearch provides better recall at cost of speed. filter (Optional[Dict[str, str]]): Filter by metadata. Defaults to None. Returns: A list of documents closest to the query vector z$matchz$search vectorSearch)rcpathrfkefSearch) page_contentr)r5r6r1 aggregaterVr4rr) r7r?r ef_searchfilterpipelinerWdocsrXrs r# _similarity_search_without_score7DocumentDBVectorSearch._similarity_search_without_score7s*F v "", $ 3 3&*&;&;$- %  * !!++H5C774>>*D KKdA B r")rc URRU5nURXbX4S9nUVs/sHoPM sn$s snf)N)r?rrr)r2 embed_queryr) r7queryrrrrMr?rdocs r#similarity_search(DocumentDBVectorSearch.similarity_searchgsL__007 44!)5  $$tt$$$s =)r1r2r5r3r6r4) r8z"Collection[DocumentDBDocumentType]r9rr-strr.rr/r)returnr)rr) rKrrLrr9rrMrrr()rbool)rNone) rgintrfrrhrrprrzdict[str, Any]r=)rz Iterable[str]rzOptional[List[Dict[str, Any]]]rMrrr )r List[str]rzList[Dict[str, Any]]rr )NN) rrr9rrzOptional[List[dict]]r8z,Optional[Collection[DocumentDBDocumentType]]rMrrr()rzOptional[List[str]]rMrrzOptional[bool])rz Optional[str]rr)(N) r?z List[float]rrrrrOptional[Dict[str, Any]]rList[Document])rr) rrrrrrrrrMrrr)rrrrrr:propertyr?rB classmethodrQrZr^rrrtrrrrrrrr!rr"r#r(r(,s&4.%,=6==  =  ==6 444 4  4 44>$:/G/K/K! <&<&-<&  <&  <&  <&B592   ,*0 +/CG   (  A      D(+/ .. . . ) .  .f % ,0 % %  % % ) % %  % %r"r() __future__rloggingenumrtypingrrrrr r r r r langchain_core.documentsrlangchain_core.vectorstoresrlangchain_core.embeddingsrpymongo.collectionrrrr$ getLoggerrloggerrr(rr"r#rsx"   .34-sD!!9c3hP   8 $H%[H%r"