o
    {qi-%                     @  s   d Z ddlmZ ddlmZ ddlmZ ddlmZ ddl	m
Z
 ddlmZmZ ddlmZ dd	lmZmZmZ dd
lmZ ddlmZ dddZeddddG dd deZdS )zLCombine documents by doing a first pass and then refining on more documents.    )annotations)Any)
deprecated)	Callbacks)Document)BasePromptTemplateformat_documentPromptTemplate)
ConfigDictFieldmodel_validator)BaseCombineDocumentsChain)LLMChainreturnr
   c                   C  s   t dgddS )Npage_contentz{page_content})input_variablestemplater	    r   r   p/home/app/PaddleOCR-VL/.venv_paddleocr/lib/python3.10/site-packages/langchain/chains/combine_documents/refine.py_get_default_document_prompt   s   r   z0.3.1z1.0zThis class is deprecated. Please see the migration guide here for a recommended replacement: https://python.langchain.com/docs/versions/migrating_chains/refine_docs_chain/)ZsinceZremovalmessagec                      s   e Zd ZU dZded< 	 ded< 	 ded< 	 ded< 	 eedZd	ed
< 	 dZded< 	 e	d6 fddZ
edddZedded7ddZedded7ddZ	d8d9d%d&Z	d8d9d'd(Zd:d+d,Zd;d0d1Zd<d2d3Ze	d=d4d5Z  ZS )>RefineDocumentsChaina&	  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,
            )
    r   initial_llm_chainrefine_llm_chainstrdocument_variable_nameinitial_response_name)default_factoryr   document_promptFboolreturn_intermediate_stepsr   	list[str]c                   s   t  j}| jrg |d}|S )z2Expect input key.

        :meta private:
        intermediate_steps)superoutput_keysr!   )selfZ_output_keys	__class__r   r   r%   d   s   z RefineDocumentsChain.output_keysTZforbid)Zarbitrary_types_allowedextrabefore)modevaluesdictr   c                 C  s   d|v r|d |d< |d= |S )zFor backwards compatibility.Zreturn_refine_stepsr!   r   )clsr,   r   r   r   get_return_intermediate_stepst   s   z2RefineDocumentsChain.get_return_intermediate_stepsc                 C  s|   d|vr
d}t ||d jj}d|vr(t|dkr"|d |d< |S d}t ||d |vr<d|d  d| }t ||S )	z4Get default document variable name, if not provided.r   z"initial_llm_chain must be providedr      r   zWdocument_variable_name must be provided if there are multiple llm_chain input_variableszdocument_variable_name z- was not found in llm_chain input_variables: )
ValueErrorpromptr   len)r.   r,   msgZllm_chain_variablesr   r   r   "get_default_document_variable_name}   s$   z7RefineDocumentsChain.get_default_document_variable_nameNdocslist[Document]	callbacksr   kwargstuple[str, dict]c           	      K  s   | j |fi |}| jjdd|i|}|g}|dd D ]}| ||}i ||}| jjdd|i|}|| q| ||S )a  Combine 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.
        r8   r0   Nr   )_construct_initial_inputsr   Zpredict_construct_refine_inputsr   append_construct_result	r&   r6   r8   r9   Zinputsresrefine_stepsdocbase_inputsr   r   r   combine_docs   s   z!RefineDocumentsChain.combine_docsc           	        s   | j |fi |}| jjdd|i|I dH }|g}|dd D ]!}| ||}i ||}| jjdd|i|I dH }|| q!| ||S )a  Async 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.
        r8   Nr0   r   )r;   r   Zapredictr<   r   r=   r>   r?   r   r   r   acombine_docs   s   z"RefineDocumentsChain.acombine_docsrA   r@   c                 C  s"   | j rd|i}||fS i }||fS )Nr#   )r!   )r&   rA   r@   Zextra_return_dictr   r   r   r>      s
   z&RefineDocumentsChain._construct_resultrB   r   dict[str, Any]c                 C  s   | j t|| j| j|iS N)r   r   r   r   )r&   rB   r@   r   r   r   r<      s   z-RefineDocumentsChain._construct_refine_inputsc                   sX   d|d j i  |d j  fdd| jjD }| j| jjdi |i}i ||S )Nr   r   c                   s   i | ]}| | qS r   r   ).0kZ	base_infor   r   
<dictcomp>   s    zBRefineDocumentsChain._construct_initial_inputs.<locals>.<dictcomp>r   )r   updatemetadatar   r   r   format)r&   r6   r9   Zdocument_inforC   r   rJ   r   r;      s   z.RefineDocumentsChain._construct_initial_inputsc                 C  s   dS )NZrefine_documents_chainr   )r&   r   r   r   _chain_type   s   z RefineDocumentsChain._chain_type)r   r"   )r,   r-   r   r   rG   )r6   r7   r8   r   r9   r   r   r:   )rA   r"   r@   r   r   r:   )rB   r   r@   r   r   rF   )r6   r7   r9   r   r   rF   )r   r   )__name__
__module____qualname____doc____annotations__r   r   r   r!   propertyr%   r   Zmodel_configr   classmethodr/   r5   rD   rE   r>   r<   r;   rO   __classcell__r   r   r'   r   r      sJ   
 
2



r   N)r   r
   )rS   
__future__r   typingr   Zlangchain_core._apir   Zlangchain_core.callbacksr   Zlangchain_core.documentsr   Zlangchain_core.promptsr   r   Zlangchain_core.prompts.promptr
   Zpydanticr   r   r   Z'langchain.chains.combine_documents.baser   Zlangchain.chains.llmr   r   r   r   r   r   r   <module>   s$    
	