o ZhD@sddlmZddlmZmZmZmZmZmZm Z m Z ddl Z ddl mZddlmZddlmZddlmZddlmZmZdd lmZdd d ZGdddeZdS)) annotations)AnyDictIterableListOptionalTupleUnioncastN)Document) Embeddings guard_import) VectorStore) AddableMixinDocstore)InMemoryDocstorereturnrcCstdS)z= Import usearch if available, otherwise raise error. usearch.indexr rr_/var/www/html/lang_env/lib/python3.10/site-packages/langchain_community/vectorstores/usearch.pydependable_usearch_importsrc@s\eZdZdZd%d d Z d&d'ddZ d(d)ddZ d(d*dd Ze !d+d,d#d$Z d S)-USearchzc`USearch` vector store. To use, you should have the ``usearch`` python package installed. embeddingr indexrdocstorerids List[str]cCs||_||_||_||_dS)z%Initialize with necessary components.N)rrrr)selfrrrrrrr__init__s zUSearch.__init__Ntexts Iterable[str] metadatasOptional[List[Dict]]&Optional[Union[np.ndarray, list[str]]]kwargsrc  st|jtstd|jd|jt|}g}t|D]\}}|r'||ni} |t || dqt |j dd|durOt fddt|D}n t|trYt |}|jt |t ||jtt|||j |ttt|S) alRun more texts through the embeddings and add to the vectorstore. Args: texts: Iterable of strings to add to the vectorstore. metadatas: Optional list of metadatas associated with the texts. ids: Optional list of unique IDs. Returns: List of ids from adding the texts into the vectorstore. zSIf trying to add texts, the underlying docstore should support adding items, which z does notZ page_contentmetadataNcsg|] \}}t|qSrstr.0id_Zlast_idrr Gsz%USearch.add_texts..) isinstancerr ValueErrorrembed_documentslist enumerateappendr intrnparrayradddictzipextendr rr+tolist) rr r"rr% embeddings documentsitextr'rr0r add_texts)s(    zUSearch.add_textsqueryr+kr8List[Tuple[Document, float]]c Cs||j|}|jt||}g}t|j|jD]"\}}|j t |}t |t s4t d|d||||fq|S)a Return docs most similar to query. Args: query: Text to look up documents similar to. k: Number of Documents to return. Defaults to 4. Returns: List of documents most similar to the query with distance. Could not find document for id , got )r embed_queryrsearchr9r:r=keysZ distancesrr+r2r r3r7) rrFrGquery_embeddingmatchesZdocs_with_scoresr.Zscoredocrrrsimilarity_search_with_scorePs  z$USearch.similarity_search_with_scoreList[Document]c Ksl|j|}|jt||}g}|jD]}|jt|}t |t s.t d|d|| |q|S)zReturn docs most similar to query. Args: query: Text to look up documents similar to. k: Number of Documents to return. Defaults to 4. Returns: List of Documents most similar to the query. rIrJ) rrKrrLr9r:rMrr+r2r r3r7) rrFrGr%rNrOdocsr.rPrrrsimilarity_searchjs    zUSearch.similarity_searchcosmetriccKs||}g}|durtddt|D}n t|tr"t|}t|D]\} } |r0|| ni} |t| | dq&tt t ||} t d} | j t |d|d}|t|t||||| ttt|S)aWConstruct USearch wrapper from raw documents. This is a user friendly interface that: 1. Embeds documents. 2. Creates an in memory docstore 3. Initializes the USearch database This is intended to be a quick way to get started. Example: .. code-block:: python from langchain_community.vectorstores import USearch from langchain_community.embeddings import OpenAIEmbeddings embeddings = OpenAIEmbeddings() usearch = USearch.from_texts(texts, embeddings) NcSsg|]\}}t|qSrr*r,rrrr1sz&USearch.from_texts..r&rr)ndimrV)r4r9r:r6r2r5r7r rr<r=rIndexlenr;r rr+r?)clsr rr"rrVr%r@rArBrCr'rZusearchrrrr from_textss   zUSearch.from_texts)rr rrrrrr)NN) r r!r"r#rr$r%rrr)rE)rFr+rGr8rrH)rFr+rGr8r%rrrR)NNrU)r rrr r"r#rr$rVr+r%rrr) __name__ __module__ __qualname____doc__rrDrQrT classmethodr[rrrrrs  *  r)rr) __future__rtypingrrrrrrr r numpyr9Zlangchain_core.documentsr Zlangchain_core.embeddingsr Zlangchain_core.utilsrZlangchain_core.vectorstoresrZ!langchain_community.docstore.baserrZ&langchain_community.docstore.in_memoryrrrrrrrs (