蘺h#dZddlmZddlmZmZmZmZddlm Z ddl m Z m Z ddl mZddlmZmZmZddlmZdd lmZdd lmZdd ZGddeZdS)zLCombine documents by doing a first pass and then refining on more documents.) annotations)AnyDictListTuple)Document)BasePromptTemplateformat_documentPromptTemplate)ExtraFieldroot_validator) Callbacks)BaseCombineDocumentsChain)LLMChainreturnr c&tdgdS)N page_contentz{page_content})input_variablestemplater q/var/www/html/services/rag-service/venv/lib/python3.11/site-packages/langchain/chains/combine_documents/refine.py_get_default_document_promptrs >*:EU V V VVrcTeZdZUdZded< ded< ded< ded< eeZd ed < d Zd ed < e d,fd Z GddZ e dd-dZ e dd-dZ d.d/d!Z d.d/d"Zd0d%Zd1d)Zd2d*Ze d3d+ZxZS)4RefineDocumentsChaina& Combine 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_factoryr document_promptFboolreturn_intermediate_stepsr List[str]cHtj}|jr|dgz}|S)z2Expect input key. :meta private: intermediate_steps)super output_keysr&)self _output_keys __class__s rr+z RefineDocumentsChain.output_keysZs1 ww*  ) A'+?*@@Lrc$eZdZdZejZdZdS)RefineDocumentsChain.Configz'Configuration for this pydantic object.TN)__name__ __module__ __qualname____doc__r forbidextraarbitrary_types_allowedrrrConfigr0es#55 "&rr8T)prevaluesrc*d|vr|d|d<|d=|S)zFor backwards compatibility.return_refine_stepsr&r)clsr:s rget_return_intermediate_stepsz2RefineDocumentsChain.get_return_intermediate_stepsks. !F * *289N2OF. /,- rcd|vr@|djj}t|dkr |d|d<nFtd|djj}|d|vrtd|dd||S)z4Get default document variable name, if not provided.r!rrzWdocument_variable_name must be provided if there are multiple llm_chain input_variableszdocument_variable_name z- was not found in llm_chain input_variables: )promptrlen ValueError)r=r:llm_chain_variabless r"get_default_document_variable_namez7RefineDocumentsChain.get_default_document_variable_namess $6 1 1"()<"="D"T &''1,,3Fq3I/00 9 #))<"="D"T ./7JJJ Uf5M.NUU?RUU rNdocsList[Document] callbacksrkwargsrTuple[str, dict]c |j|fi|}|jjdd|i|}|g}|ddD]G}|||}i||}|jjdd|i|}||H|||S)aCombine by mapping first chain over all, then stuffing into final chain. Args: docs: List of documents to combine callbacks: Callbacks to be passed through **kwargs: additional parameters to be passed to LLM calls (like other input variables besides the documents) Returns: The first element returned is the single string output. The second element returned is a dictionary of other keys to return. rHr@Nr)_construct_initial_inputsrpredict_construct_refine_inputsrappend_construct_result r,rFrHrIinputsres refine_stepsdoc base_inputss r combine_docsz!RefineDocumentsChain.combine_docss0/????,d$,KKyKFKKu 8 % %C77SAAK. .v.F/$'/NN)NvNNC    $ $ $ $%%lC888rc8K|j|fi|}|jjdd|i|d{V}|g}|ddD]M}|||}i||}|jjdd|i|d{V}||N|||S)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. rHNr@r)rLrapredictrNrrOrPrQs r acombine_docsz"RefineDocumentsChain.acombine_docss 0/????3D*3RRiR6RRRRRRRRu 8 % %C77SAAK. .v.F6-6UUUfUUUUUUUUC    $ $ $ $%%lC888rrTrSc&|jrd|i}ni}||fS)Nr))r&)r,rTrSextra_return_dicts rrPz&RefineDocumentsChain._construct_results-  ) #!5| D   " %%%rrUrDict[str, Any]cH|jt||j|j|iSN)r!r r$r")r,rUrSs rrNz-RefineDocumentsChain._construct_refine_inputss)  'd>R)S)S  &  rc d|dji|djfd|jjD}|j|jjdi|i}i||}|S)Nrrc"i|] }|| Srr).0k base_infos r zBRefineDocumentsChain._construct_initial_inputs..sWWWQIaLWWWrr)rupdatemetadatar$rr!format)r,rFrI document_inforVrRrds @rrLz.RefineDocumentsChain._construct_initial_inputss$T!W%9: a)***WWWW$2F2VWWW  ')D)=)D)U)U})U)U +K*6* rcdS)Nrefine_documents_chainr)r,s r _chain_typez RefineDocumentsChain._chain_types''r)rr')r:rrrr_)rFrGrHrrIrrrJ)rTr'rSr rrJ)rUrrSr rr])rFrGrIrrr])rr )r1r2r3r4__annotations__rrr$r&propertyr+r8rr>rErWrZrPrNrLrl __classcell__)r.s@rrrs00d /)QL*/%4+++OS&+++++? X'''''''' ^^*<@999994<@999994&&&&       (((X(((((rrN)rr )r4 __future__rtypingrrrrlangchain_core.documentsrlangchain_core.promptsr r langchain_core.prompts.promptr langchain_core.pydantic_v1r rrlangchain.callbacks.managerr'langchain.chains.combine_documents.baserlangchain.chains.llmrrrrrrrys6RR""""""))))))))))))------FFFFFFFF888888CCCCCCCCCC111111*)))))WWWW(((((4(((((r