@h-%SrSSKJr SSKJr SSKJr SSKJr SSK J r SSK J r J r SSKJr SS KJrJrJr SS KJr SS KJr SS jr\"S SSS9"SS\55rg)zLCombine documents by doing a first pass and then refining on more documents.) annotations)Any) deprecated) Callbacks)Document)BasePromptTemplateformat_documentPromptTemplate) ConfigDictFieldmodel_validator)BaseCombineDocumentsChain)LLMChainc[S/SS9$)N page_contentz{page_content})input_variablestemplater a/var/www/html/shao/venv/lib/python3.13/site-packages/langchain/chains/combine_documents/refine.py_get_default_document_promptrs >*:EU VVrz0.3.1z1.0zThis class is deprecated. Please see the migration guide here for a recommended replacement: https://python.langchain.com/docs/versions/migrating_chains/refine_docs_chain/)sinceremovalmessagech^\rSrSr%SrS\S'S\S'S\S'S\S'\"\S 9rS \S 'S r S \S'\ SU4Sjj5r \ "SSS9r \"SS9\SSj55r\"SS9\SSj55rS S!SjjrS S!SjjrS"SjrS#SjrS$Sjr\ S%Sj5rSrU=r$)&RefineDocumentsChainazCombine documents by doing a first pass and then refining on more documents. This algorithm first calls `initial_llm_chain` on the first document, passing that first document in with the variable name `document_variable_name`, and produces a new variable with the variable name `initial_response_name`. Then, it loops over every remaining document. This is called the "refine" step. It calls `refine_llm_chain`, passing in that document with the variable name `document_variable_name` as well as the previous response with the variable name `initial_response_name`. Example: .. code-block:: python from langchain.chains import RefineDocumentsChain, LLMChain from langchain_core.prompts import PromptTemplate from langchain_community.llms import OpenAI # This controls how each document will be formatted. Specifically, # it will be passed to `format_document` - see that function for more # details. document_prompt = PromptTemplate( input_variables=["page_content"], template="{page_content}" ) document_variable_name = "context" llm = OpenAI() # The prompt here should take as an input variable the # `document_variable_name` prompt = PromptTemplate.from_template( "Summarize this content: {context}" ) initial_llm_chain = LLMChain(llm=llm, prompt=prompt) initial_response_name = "prev_response" # The prompt here should take as an input variable the # `document_variable_name` as well as `initial_response_name` prompt_refine = PromptTemplate.from_template( "Here's your first summary: {prev_response}. " "Now add to it based on the following context: {context}" ) refine_llm_chain = LLMChain(llm=llm, prompt=prompt_refine) chain = RefineDocumentsChain( initial_llm_chain=initial_llm_chain, refine_llm_chain=refine_llm_chain, document_prompt=document_prompt, document_variable_name=document_variable_name, initial_response_name=initial_response_name, ) rinitial_llm_chainrefine_llm_chainstrdocument_variable_nameinitial_response_name)default_factoryrdocument_promptFboolreturn_intermediate_stepscJ>[TU]nUR(a/UQSPnU$)z"Expect input key. :meta private: intermediate_steps)super output_keysr')self _output_keys __class__s rr+ RefineDocumentsChain.output_keysds. w*  ) )@\@+?@LrTforbid)arbitrary_types_allowedextrabefore)modec(SU;a USUS'US U$)zFor backwards compatibility.return_refine_stepsr'r)clsvaluess rget_return_intermediate_steps2RefineDocumentsChain.get_return_intermediate_stepsts+ !F *289N2OF. /,- rcSU;a Sn[U5eUSRRnSU;a&[U5S:Xa USUS'U$Sn[U5eUSU;aSUSSU3n[U5eU$) z4Get default document variable name, if not provided.rz"initial_llm_chain must be providedr"rzWdocument_variable_name must be provided if there are multiple llm_chain input_variableszdocument_variable_name z- was not found in llm_chain input_variables: ) ValueErrorpromptrlen)r7r8msgllm_chain_variabless r"get_default_document_variable_name7RefineDocumentsChain.get_default_document_variable_name}s f ,6CS/ !$%89@@PP #6 1&'1,3Fq3I/0 9!o% , -5H H)&1I*J)KL;;N:OQ S/ ! rc 0UR"U40UD6nURR"SSU0UD6nU/nUSSHInURXu5n0UEUEnURR"SSU0UD6nUR U5 MK UR Xe5$)aCombine by mapping first chain over all, then stuffing into final chain. Args: docs: List of documents to combine callbacks: Callbacks to be passed through **kwargs: additional parameters to be passed to LLM calls (like other input variables besides the documents) Returns: The first element returned is the single string output. The second element returned is a dictionary of other keys to return. callbacksr<Nr)_construct_initial_inputsrpredict_construct_refine_inputsr append_construct_result r,docsrEkwargsinputsres refine_stepsdoc base_inputss r combine_docs!RefineDocumentsChain.combine_docss$//??$$,,KyKFKu 8C77AK. .v.F''//N)NvNC    $  %%l88rc`# UR"U40UD6nURR"SSU0UD6IShvN nU/nUSSHQnURXu5n0UEUEnURR"SSU0UD6IShvN nUR U5 MS UR Xe5$NrN,7f)aAsync combine by mapping a first chain over all, then stuffing into a final chain. Args: docs: List of documents to combine callbacks: Callbacks to be passed through **kwargs: additional parameters to be passed to LLM calls (like other input variables besides the documents) Returns: The first element returned is the single string output. The second element returned is a dictionary of other keys to return. rENr<r)rFrapredictrHr rIrJrKs r acombine_docs"RefineDocumentsChain.acombine_docss&//??**33RiR6RRu 8C77AK. .v.F--66UUfUUC    $  %%l88S Vs"5B.B*AB.?B,+B.,B.c<UR(aSU0nX#4$0nX#4$)Nr))r')r,rPrOextra_return_dicts rrJ&RefineDocumentsChain._construct_results2  ) )!5| D %%!# %%rc\UR[XR5URU0$N)r"r r%r#)r,rQrOs rrH-RefineDocumentsChain._construct_refine_inputss-  ' '>R>R)S  & &  rc SUSR0nURUSR5 URRVs0sHoDX4_M nnUR URR "S0UD60n0UEUE$s snf)Nrrr)rupdatemetadatar%rr"format)r,rLrM base_infok document_inforRs rrF.RefineDocumentsChain._construct_initial_inputss $T!W%9%9: a))*262F2F2V2VW2VQIL2V W  ' ')=)=)D)D)U})U )+((( Xs Bcg)Nrefine_documents_chainr)r,s r _chain_type RefineDocumentsChain._chain_types'rr)return list[str])r8dictrkrr])rLlist[Document]rErrMrrktuple[str, dict])rPrlrOr!rkro)rQrrOr!rkdict[str, Any])rLrnrMrrkrp)rkr!)__name__ __module__ __qualname____firstlineno____doc____annotations__r rr%r'propertyr+r model_configr classmethodr9rBrSrWrJrHrFri__static_attributes__ __classcell__)r.s@rrrs`0d /)QL*/4+O'S&+t+?  $L (#$(#$6 $999 9  9> $999 9  9:& ) ) )  )((rrN)rkr )ru __future__rtypingrlangchain_core._apirlangchain_core.callbacksrlangchain_core.documentsrlangchain_core.promptsrr langchain_core.prompts.promptr pydanticr r r'langchain.chains.combine_documents.baserlangchain.chains.llmrrrrrrrscR"*.-F877*W   Y K(4K(K(r