dha@SSKJr SSKrSSKrSSKJr SSKJrJrJ r J r J r J r J r JrJr SSKrSSKJr SSKJr SSKJr SSKJr \(aSSKr"S S \5rg) ) annotationsN)repeat) TYPE_CHECKINGAnyDictIterableListOptionalTupleTypeUnion)Document) Embeddings) VectorStore)maximal_marginal_relevancecj\rSrSrSrSSSjjr\SSj5rSSSjjr\ SSSjj5r SSjr SSS jjr SSS jjr SS S jjrS!S jrS"S#S jjrSS$Sjjr\S%S&Sjj5r\S'Sj5rS(S)SjjrS(S*SjjrS%S+SjjrSrg),SupabaseVectorStoreaJ`Supabase Postgres` vector store. It assumes you have the `pgvector` extension installed and a `match_documents` (or similar) function. For more details: https://integrations.langchain.com/vectorstores?integration_name=SupabaseVectorStore You can implement your own `match_documents` function in order to limit the search space to a subset of documents based on your own authorization or business logic. Note that the Supabase Python client does not yet support async operations. If you'd like to use `max_marginal_relevance_search`, please review the instructions below on modifying the `match_documents` function to return matched embeddings. Examples: .. code-block:: python from langchain_community.embeddings.openai import OpenAIEmbeddings from langchain_core.documents import Document from langchain_community.vectorstores import SupabaseVectorStore from supabase.client import create_client docs = [ Document(page_content="foo", metadata={"id": 1}), ] embeddings = OpenAIEmbeddings() supabase_client = create_client("my_supabase_url", "my_supabase_key") vector_store = SupabaseVectorStore.from_documents( docs, embeddings, client=supabase_client, table_name="documents", query_name="match_documents", chunk_size=500, ) To load from an existing table: .. code-block:: python from langchain_community.embeddings.openai import OpenAIEmbeddings from langchain_community.vectorstores import SupabaseVectorStore from supabase.client import create_client embeddings = OpenAIEmbeddings() supabase_client = create_client("my_supabase_url", "my_supabase_key") vector_store = SupabaseVectorStore( client=supabase_client, embedding=embeddings, table_name="documents", query_name="match_documents", ) NcSSKnXlX lU=(d SUlU=(d SUlU=(d SUlg![a [S5ef=f)z Initialize with supabase client.rNzXCould not import supabase python package. Please install it with `pip install supabase`. documentsmatch_documents)supabase ImportError_client _embedding table_name query_name chunk_size)selfclient embeddingrrrrs a/var/www/html/shao/venv/lib/python3.13/site-packages/langchain_community/vectorstores/supabase.py__init__SupabaseVectorStore.__init__Xs_   &/$3 $9(9$+ A  s AAcUR$N)r)r s r# embeddingsSupabaseVectorStore.embeddingsqs c U=(d. UVs/sH!n[[R"55PM# snnURX5nURR [ U55nURXvU5$s snfr')struuiduuid4_texts_to_documentsrembed_documentslist add_vectors)r texts metadatasidskwargs_docsvectorss r# add_textsSupabaseVectorStore.add_textsush77Ac$**,'7''9//11$u+>s33 8s'A?c  0U(d [S5eU(d [S5eURU5n UV s/sH!n [[R"55PM# nn UR X5n UR "XEXX40U D6 U"UUUUUS9$s sn f)z9Return VectorStore initialized from texts and embeddings.zSupabase client is required.z)Supabase document table_name is required.)r!r"rrr) ValueErrorr0r,r-r.r/ _add_vectors) clsr3r"r4r!rrrr5r6r(r7r8s r# from_textsSupabaseVectorStore.from_textss;< <HI I..u5 */0%Qs4::< %0&&u8   # EK !!!   1s'BcfURURURXX0R5$r')r>rrr)r r9rr5s r#r2SupabaseVectorStore.add_vectorss,   LL$//7sOO  r*c bURRU5nUR"U4X#S.UD6$N)kfilter)r embed_querysimilarity_search_by_vectorr queryrFrGr6vectors r#similarity_search%SupabaseVectorStore.similarity_searchs3,,U3//U!UfUUr*c fUR"U4X#S.UD6nUVVs/sHupgUPM nnnU$s snnfrE)1similarity_search_by_vector_with_relevance_scores) r r"rFrGr6resultdocr7rs r#rI/SupabaseVectorStore.similarity_search_by_vectorsJGG   -3 (..vVSSv ./s -c bURRU5nUR"U4X#S.UD6$rE)rrHrPrJs r#'similarity_search_with_relevance_scores;SupabaseVectorStore.similarity_search_with_relevance_scoress>,,U3EE   *0  r*c.[US9nU(aX#S'U$)N)query_embeddingrG)dict)r rKrGrets r# match_argsSupabaseVectorStore.match_argss#59 "M r*c U(arUR5H^upg[U[5(dMSU;dM$USnSRSU55n SUSU S3n U(a SUSU S3nM\U nM` UR X5n UR R URU 5n U(a%U RRS SUS35U lU RRS U5U lU R5n U RVs/sHWnURS 5(dM[URS 05URS S 5S9URSS54PMY nnUbHUVVs/sHunnUU:dMUU4PM nnn[U5S:Xa[R "SU35 U$s snfs snnf)Nz$in,c3># UHnS[U5S3v M g7f)'N)r,).0vs r# XSupabaseVectorStore.similarity_search_by_vector_with_relevance_scores..s)KAAc!fXQ-sz metadata->>z IN ()(z) and (andlimitcontentmetadatarj page_content similarityrzDNo relevant docs were retrieved using the relevance score threshold )items isinstancerYjoinr[rrpcrparamssetexecutedatagetrlenwarningswarn)r rKrFrGpostgrest_filterscore_thresholdkeyvalue in_values values_str new_filtermatch_documents_params query_builderressearch match_resultrRrns r#rPESupabaseVectorStore.similarity_search_by_vector_with_relevance_scoress $lln eT**u~ %e I!$)K)K!KJ#.se5 A!FJ(-./?.@ |ST+U(+5(-"&!? ((:PQ #0#7#7#;#;+,A.$M  -3377C ##%((  #zz)$ #ZZ B7!'Ir!: <-  #     &(4'3OC0"j!'3  < A% ""1!24 1  sG9AG G G c URX5nURRURU5nU(a%URR SSUS35UlURR SU5UlUR 5nURVs/sHnURS5(dM[URS05URSS5S9URS S 5[R"URS S5RS 5[RS S94PM n nU $s snf)Nrgrfrerhrirjrkrlrnror"z[]r^)sep)r[rrsrrtrurvrwrxrnp fromstringstripfloat32) r rKrFrGr|rrrrrs r#0similarity_search_by_vector_returning_embeddingsDSupabaseVectorStore.similarity_search_by_vector_returning_embeddingss0"&!? ((:PQ #0#7#7#;#;+,A.$M  -3377C ##%(( #zz)$ #ZZ B7!'Ir!: <- JJ{B/55d;RZZS #  "# s #EBEc xUc [05n[X5VVs/sHup#[X#S9PM nnnU$s snnf)z:Return list of Documents from list of texts and metadatas.)rmrj)rzipr)r3r4textrjr8s r#r/'SupabaseVectorStore._texts_to_documentsAsM  r I#&e"7 "7 $ :"7    s6c >[U5VVs/sH(upxXGX7RUX7RS.UEPM* n nn/n [S[ U 5U5Hn XX-n UR U5R U 5R5n [ U R5S:Xa [S5eU RV s/sH4oRS5(dM[U RS55PM6 nn U RU5 M U $s snnfs sn f)zAdd vectors to Supabase table.)idrir"rjrzError inserting: No rows addedr) enumeratermrjrangeryfrom_upsertrvrw Exceptionrxr,extend)r!rr9rr5rr6idxr"rowsid_listichunkrQs r#r> SupabaseVectorStore._add_vectorsQs(#,G"4 & #5 h$>66&%N33     #5 &  q#d)Z0AQ^,E\\*-44U;CCEF6;;1$ @AA.4[[H[EE$K#3quuT{#[CH NN3 11 & (Is.DDDc URX5nUVs/sHowSPM nnUVs/sHowSPM n n[[R"U/[RS9U UUS9n U V s/sHoU PM n n U $s snfs snfs sn f)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. 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. Returns: List of Documents selected by maximal marginal relevance. r)dtype)rF lambda_mult)rrrarrayr) r r"rFfetch_krr6rQ doc_tuplematched_documentsmatched_embeddings mmr_selectedrfiltered_documentss r#'max_marginal_relevance_search_by_vector;SupabaseVectorStore.max_marginal_relevance_search_by_vectorws0FF   query_embedding) AS similarity FROM docstore ORDER BY docstore.embedding <=> query_embedding LIMIT match_count; END; $$; ``` )r)rrHr)r rKrFrrr6r"r8s r#max_marginal_relevance_search1SupabaseVectorStore.max_marginal_relevance_searchs:rOO//6 ;; '<  r*c  Uc [S5eUVs/sHnSU0PM nnUHWnURRUR5R 5R SUS5R 5 MY gs snf)z=Delete by vector IDs. Args: ids: List of ids to delete. NzNo ids provided to delete.r)r=rrrdeleteeqrv)r r5r6rrrows r#rSupabaseVectorStore.deletes ;9: :  & b  & C LL  t / 6 6 8 ; ;D#d) L T T V& s B)rrrrr)rN) r!supabase.client.Clientr"rrr,rintrUnion[str, None]returnNone)rr)NN) r3 Iterable[str]r4zOptional[List[Dict[Any, Any]]]r5Optional[List[str]]r6rr List[str])NNrrrN)r?zType['SupabaseVectorStore']r3rr"rr4zOptional[List[dict]]r!z Optional[supabase.client.Client]r Optional[str]rrrrr5rr6rrz'SupabaseVectorStore')r9List[List[float]]rList[Document]r5rrr)N) rKr,rFrrGOptional[Dict[str, Any]]r6rrr) r" List[float]rFrrGrr6rrr) rKr,rFrrGrr6rrList[Tuple[Document, float]])rKrrGrrzDict[str, Any])NNN) rKrrFrrGrr|rr}zOptional[float]rr) rKrrFrrGrr|rrz(List[Tuple[Document, float, np.ndarray]]r')r3rr4z"Optional[Iterable[Dict[Any, Any]]]rr)r!rrr,r9rrrr5rrrr6rrr)rg?) r"rrFrrrrfloatr6rrr) rKr,rFrrrrrr6rrr)r5rr6rrr)__name__ __module__ __qualname____firstlineno____doc__r$propertyr(r: classmethodr@r2rMrIrUr[rPr staticmethodr/r>rrr__static_attributes__r*r#rrs8~'+ ,&,, ,  , % , ,2 59#' 4 42 4! 4  4  4 +/37$/'8#'! (! ! ! ( ! 1 ! " ! %! ! !! !  ! ! F " "     +/ VV V) V  V  V+/    )     $+/       )    &   *B ,0*.+/ << <) < ( < ) < &<D,0*. $$ $) $ ( $ 2 $L9=  5    #&####" #  #  ## ##P ("(" (" ("  ("  (" ("Z == = =  =  = =~WWr*r) __future__rr-rz itertoolsrtypingrrrrr r r r r numpyrlangchain_core.documentsrlangchain_core.embeddingsrlangchain_core.vectorstoresr&langchain_community.vectorstores.utilsrrrrr*r#rsG"    -03MVW+VWr*