3fiFdZddlmZddlZddlZddlZddlZddlZddlZddl m Z ddl m Z m Z mZmZddlmZddlmZmZddlZddlmZdd lmZmZmZdd lmZmZmZm Z m!Z!m"Z"dd l#m$Z$dd l%m&Z&dd l'm(Z(ddl)m*Z*ddl+m,Z,ddl-m.Z.ddl/m0Z0ddl1m2Z2m3Z3m4Z4ddl5m6Z6ddl7m8Z8ddl9m:Z:ddl;mZ>ddl?m@Z@mAZAddlBmCZCddlDmEZEddlFmGZGddlHmIZIddlJmKZKddlLmMZMddlNmOZOejeQZRGdd e<ZSGd!d"e<ZTGd#d$e*eezZUGd%d&e*eVeezZWGd'd(eSZXGd)d*eTZYed+eCd,-Gd.d/eSZZed+eCd,-Gd0d1eSZ[Gd2d3e8Z\eVeezezZ]eXeYzZ^Gd4d5eKZ_y)6zEChain that takes in an input and produces an action and action input.) annotationsN)abstractmethod) AsyncIteratorCallableIteratorSequence)Path)Anycast) deprecated) AgentAction AgentFinish AgentStep)AsyncCallbackManagerForChainRunAsyncCallbackManagerForToolRunBaseCallbackManagerCallbackManagerForChainRunCallbackManagerForToolRun Callbacks)OutputParserException)BaseLanguageModel) BaseMessage)BaseOutputParser)BasePromptTemplate)FewShotPromptTemplate)PromptTemplate)RunnableRunnableConfig ensure_config) AddableDict)BaseTool)get_color_mapping) BaseModel ConfigDictmodel_validator)Selfoverride)AGENT_DEPRECATION_WARNING)AgentExecutorIterator) AgentType) InvalidTool)Chain)LLMChain)asyncio_timeoutceZdZdZed dZddZe d ddZe d ddZ eed dZ ddZ e d ddZ edd Zedfd Zdd Zdd ZxZS)BaseSingleActionAgentzBase Single Action Agent class.cdgSReturn values of the agent.outputselfs \/var/www/auto_recruiter/arenv/lib/python3.12/site-packages/langchain_classic/agents/agent.py return_valuesz#BaseSingleActionAgent.return_values: zcy)Get allowed tools.Nr5r6s r8get_allowed_toolsz'BaseSingleActionAgent.get_allowed_tools?sr;c y)0Given input, decided what to do. Args: intermediate_steps: Steps the LLM has taken to date, along with observations. callbacks: Callbacks to run. **kwargs: User inputs. Returns: Action specifying what tool to use. Nr5r7intermediate_steps callbackskwargss r8planzBaseSingleActionAgent.planCr;c Kyw)6Async given input, decided what to do. Args: intermediate_steps: Steps the LLM has taken to date, along with observations. callbacks: Callbacks to run. **kwargs: User inputs. Returns: Action specifying what tool to use. Nr5rAs r8aplanzBaseSingleActionAgent.aplanVcyReturn the input keys.Nr5r6s r8 input_keysz BaseSingleActionAgent.input_keysirFr;c J|dk(rtddidSd|d}t|)Return response when agent has been stopped due to max iterations. Args: early_stopping_method: Method to use for early stopping. intermediate_steps: Steps the LLM has taken to date, along with observations. Returns: Agent finish object. Raises: ValueError: If `early_stopping_method` is not supported. forcer43Agent stopped due to iteration limit or time limit.'Got unsupported early_stopping_method ``r ValueErrorr7early_stopping_methodrB_msgs r8return_stopped_responsez-BaseSingleActionAgent.return_stopped_responsensA& !G +PQ 88M7NaPor;c t)aConstruct an agent from an LLM and tools. Args: llm: Language model to use. tools: Tools to use. callback_manager: Callback manager to use. kwargs: Additional arguments. Returns: Agent object. NotImplementedError)clsllmtoolscallback_managerrDs r8from_llm_and_toolsz(BaseSingleActionAgent.from_llm_and_toolss &"!r;ctz#Return Identifier of an agent type.r_r6s r8 _agent_typez!BaseSingleActionAgent._agent_type "!r;c t|} |j}t |t rt |j|d<|S|||d<|S#t$rd}Y@wxYw)zuReturn dictionary representation of agent. Returns: Dictionary representation of agent. N_type)super model_dumprhr` isinstancer*strvalue)r7rD_dictrk __class__s r8dictzBaseSingleActionAgent.dictss"$ $$E eY ' -E'N  "E'N # E s A AAc4t|tr t|n|}|j}|j dd|j }d|vrd|d}t ||jdk(r3|jd5}tj||d d d d y |jjd r3|jd5}tj||d d d d y |d}t|#1swYy xYw#1swYy xYw)zSave the agent. Args: file_path: Path to file to save the agent to. Example: ```python # If working with agent executor agent.agent.save(file_path="path/agent.yaml") ``` Tparentsexist_okrkAgent z does not support saving.jsonwindentNz.yamlz.ymlFdefault_flow_style must be json or yaml)rnror parentmkdirrsr`suffixopenjsondumpendswithyamlrX)r7 file_path save_pathdirectory_path agent_dictr\fs r8savezBaseSingleActionAgent.saves(2)S'ADOy "))TD9YY[ * $4& 89C%c* *   w &$ 3 *a2 3 3    & &'8 9$ C *aEB C CK45CS/ ! 3 3 C CDDD DciS#Return logging kwargs for tool run.r5r6s r8tool_run_logging_kwargsz-BaseSingleActionAgent.tool_run_logging_kwargs r;return list[str]rlist[str] | NoneNrBlist[tuple[AgentAction, str]]rCrrDr rAgentAction | AgentFinishrZrorBrr[r rr) rbrrcSequence[BaseTool]rdBaseCallbackManager | NonerDr rr0rrorDr r builtins.dictrz Path | strrNonerr)__name__ __module__ __qualname____doc__propertyr9r>rrErIrOr] classmethodrerhr'rsrr __classcell__rrs@r8r0r07s^)  $ 9    #   $ $ 9    #   $%%":   8 8< " """5 "  "  ""(""" "Dr;r0ceZdZdZed dZd dZe d ddZe d ddZ eed dZ ddZ eddZ e dfd Zdd Zdd ZxZS)BaseMultiActionAgentzBase Multi Action Agent class.cdgSr2r5r6s r8r9z"BaseMultiActionAgent.return_valuesr:r;cy)zHGet allowed tools. Returns: Allowed tools. Nr5r6s r8r>z&BaseMultiActionAgent.get_allowed_toolss r;c y)a5Given input, decided what to do. Args: intermediate_steps: Steps the LLM has taken to date, along with the observations. callbacks: Callbacks to run. **kwargs: User inputs. Returns: Actions specifying what tool to use. Nr5rAs r8rEzBaseMultiActionAgent.planrFr;c Kyw)a;Async given input, decided what to do. Args: intermediate_steps: Steps the LLM has taken to date, along with the observations. callbacks: Callbacks to run. **kwargs: User inputs. Returns: Actions specifying what tool to use. Nr5rAs r8rIzBaseMultiActionAgent.aplanrJrKcyrMr5r6s r8rOzBaseMultiActionAgent.input_keysrFr;c J|dk(rtddidSd|d}t|)rQrRr4z$Agent stopped due to max iterations.rTrUrVrWrYs r8r]z,BaseMultiActionAgent.return_stopped_responses:& !G +*PQSUV V78M7NaPor;ctrgr_r6s r8rhz BaseMultiActionAgent._agent_type1rir;c t|}tjt5t |j |d<ddd|S#1swY|SxYw)*Return dictionary representation of agent.rkN)rlrm contextlibsuppressr`rorhr7rDrqrrs r8rszBaseMultiActionAgent.dict6sO"$  !4 5 3 !1!12E'N 3  3 s A  Ac4t|tr t|n|}|j}d|vrd|d}t ||j }|j dd|jdk(r3|jd5}tj||d d d d y |jjd r3|jd5}tj||d d d d y |d}t|#1swYy xYw#1swYy xYw)aSave the agent. Args: file_path: Path to file to save the agent to. Raises: NotImplementedError: If agent does not support saving. ValueError: If `file_path` is not json or yaml. Example: ```python # If working with agent executor agent.agent.save(file_path="path/agent.yaml") ``` rkrxz does not support saving.Truryrzr{r|Nr~Frr)rnror rsr`rrrrrrrrrX)r7rrrr\rrs r8rzBaseMultiActionAgent.save>s"(2)S'ADOy YY[ * $4& 9:C%c* *"))TD9   w &$ 3 *a2 3 3    & &'8 9$ C *aEB C CK45CS/ ! 3 3 C CrciSrr5r6s r8rz,BaseMultiActionAgent.tool_run_logging_kwargsdrr;rrrrBrrCrrDr rlist[AgentAction] | AgentFinishrrrrr)rrrrrr9r>rrErIrOr]rhr'rsrrrrs@r8rrs(  $ 9    )   $ $ 9    )   $%%":   2""$"Lr;rc"eZdZdZeddZy)AgentOutputParserz=Base class for parsing agent output into agent action/finish.cy)z$Parse text into agent action/finish.Nr5r7texts r8parsezAgentOutputParser.parselrFr;N)rrorrrrrrrrr5r;r8rrisG33r;rc"eZdZdZeddZy)MultiActionAgentOutputParserzBase class for parsing agent output into agent actions/finish. This is used for agents that can return multiple actions. cy)zParse text into agent actions/finish. Args: text: Text to parse. Returns: List of agent actions or agent finish. Nr5rs r8rz"MultiActionAgentOutputParser.parseyrFr;N)rrorrrr5r;r8rrqs   r;rceZdZUdZded< gZded<gZded<dZded < ed Z e dd Z e dd Z d ddZ d ddZy ) RunnableAgentAgent powered by Runnables.z)Runnable[dict, AgentAction | AgentFinish]runnablerinput_keys_argreturn_keys_argTboolstream_runnablearbitrary_types_allowedc|jSr3rr6s r8r9zRunnableAgent.return_values###r;c|jSrNrr6s r8rOzRunnableAgent.input_keyss"""r;Nc i|d|i}d}|jr0|jj|d|iD] }||}||z }|S|jj|d|i}|SaPBased on past history and current inputs, decide what to do. Args: intermediate_steps: Steps the LLM has taken to date, along with the observations. callbacks: Callbacks to run. **kwargs: User inputs. Returns: Action specifying what tool to use. rBNrCconfigrrstreaminvoker7rBrCrDinputs final_outputchunks r8rEzRunnableAgent.plan"FFE02DE   --fk9=U-V *'#(L E)L  * ==// Y?W/XLr;cKi|d|i}d}|jr3|jj|d|i23d{}||}||z }|jj|d|id{}|S786|S7 waRAsync based on past history and current inputs, decide what to do. Args: intermediate_steps: Steps the LLM has taken to date, along with observations. callbacks: Callbacks to run. **kwargs: User inputs. Returns: Action specifying what tool to use. rBNrCrrrastreamainvokers r8rIzRunnableAgent.aplans"FFE02DE    $}}44#Y/ 5  * *e '#(L E)L!%!6!6#Y/"7"L *  26A<A7A5A7.A<.A:/A<5A77A<rrrrrrr__annotations__rrrr$ model_configrr9rOrErIr5r;r8rrs%77/ "NI"!#OY# OT  $L$$## $"9"" " # "N $'9'' ' # 'r;rceZdZUdZded< gZded<gZded<dZded < ed Z e dd Z e dd Z d ddZ d ddZy )RunnableMultiActionAgentrz/Runnable[dict, list[AgentAction] | AgentFinish]rrrrTrrrc|jSrrr6s r8r9z&RunnableMultiActionAgent.return_valuesrr;c|jS)QReturn the input keys. Returns: List of input keys. rr6s r8rOz#RunnableMultiActionAgent.input_keys s"""r;Nc i|d|i}d}|jr0|jj|d|iD] }||}||z }|S|jj|d|i}|Srrrs r8rEzRunnableMultiActionAgent.planrr;cKi|d|i}d}|jr3|jj|d|i23d{}||}||z }|jj|d|id{}|S786|S7 wrrrs r8rIzRunnableMultiActionAgent.aplan7s"FFE02DE    $}}44#Y/ 5  * *e '#(L E)L!%!6!6#Y/"7"L  *  rrrrrr5r;r8rrs%==0 "NI"!#OY# OT  $L$$## $"9"" " ) "N $(9(( ( ) (r;rz0.1.0z1.0)messageremovalceZdZUdZded< ded< ded< ed dZedfd Z d dd Z d dd Z dd Z xZ S)LLMSingleActionAgentz$Base class for single action agents.r- llm_chainr output_parserrstopcZtt|jjdhz S)rrBlistsetrrOr6s r8rOzLLMSingleActionAgent.input_keysqs'C1126J5KKLLr;c *t|}|d=|Srrrlrsrs r8rszLLMSingleActionAgent.dictz  / " r;c |jjd||j|d|}|jj |S)a4Given input, decided what to do. Args: intermediate_steps: Steps the LLM has taken to date, along with the observations. callbacks: Callbacks to run. **kwargs: User inputs. Returns: Action specifying what tool to use. rBrrCr5)rrunrrrr7rBrCrDr4s r8rEzLLMSingleActionAgent.plansO"$## 1    !!''//r;cK|jjd||j|d|d{}|jj |S7w)rHr Nr5)rarunrrrr s r8rIzLLMSingleActionAgent.aplans]"+t~~** 1     !!''//  s.AA Ac^dt|jdk(rddS|jddS)rrTr llm_prefixobservation_prefix)lenrr6s r8rz,LLMSingleActionAgent.tool_run_logging_kwargss8(+DII!(;"  AE1  r;rrrrr) rrrrrrrOr'rsrErIrrrs@r8rrbs /$$$) O% MM $0900 0 # 08 $0900 0 # 02 r;rceZdZUdZded< ded< dZded< edfd Zdd Ze dd Z e dd Z d d Z d! d"dZ d! d"dZ d#dZe ddZedd$dZe ed%dZe ed%dZeed&dZed'dZeed(dZe d) d*dZ d+dZd,dZxZS)-AgentzAgent that calls the language model and deciding the action. This is driven by a LLMChain. The prompt in the LLMChain MUST include a variable called "agent_scratchpad" where the agent can put its intermediary work. r-rrrNr allowed_toolsc *t|}|d=|Srrrs r8rsz Agent.dictrr;c|jS)r=)rr6s r8r>zAgent.get_allowed_toolss!!!r;cdgSr2r5r6s r8r9zAgent.return_valuesr:r;cvd|jjd|jjgS)N z )rrstripr6s r8_stopz Agent._stopsB((//12 34**1134 5  r;c~d}|D]5\}}||jz }|d|j|d|jz }7|S)zJConstruct the scratchpad that lets the agent continue its thought process.rTr)logrr)r7rBthoughtsaction observations r8_construct_scratchpadzAgent._construct_scratchpads[ #5 W FK  "H "T445k]"T__DUV VH Wr;c |j|fi|}|jjdd|i|}|jj |S)r@rCr5)get_full_inputsrpredictrrr7rBrCrD full_inputs full_outputs r8rEz Agent.plansQ"+d**+=HH ,dnn,,PyPKP !!'' 44r;cK|j|fi|}|jjdd|i|d{}|jj |d{S7'7w)rHrCNr5)r$rapredictraparser&s r8rIz Agent.aplanse"+d**+=HH 3DNN33WiW;WW ''..{;;;X;s!5A#A"A#A!A#!A#c N|j|}||jd}i||S)a!Create the full inputs for the LLMChain from intermediate steps. Args: intermediate_steps: Steps the LLM has taken to date, along with observations. **kwargs: User inputs. Returns: Full inputs for the LLMChain. agent_scratchpadr)r"r)r7rBrDr new_inputss r8r$zAgent.get_full_inputss3--.@A*2DJJG '&'J''r;cZtt|jjdhz S)rNr.rr6s r8rOzAgent.input_keys.s'C1126H5IIJJr;aftermodec||jj}d|jvrtj d|jj dt |tr|xjdz c_|St |tr|xjdz c_ |Sdt|}t||S)aCValidate that prompt matches format. Args: values: Values to validate. Returns: Validated values. Raises: ValueError: If `agent_scratchpad` is not in prompt.input_variables and prompt is not a FewShotPromptTemplate or a PromptTemplate. r.zl`agent_scratchpad` should be a variable in prompt.input_variables. Did not find it, so adding it at the end.z {agent_scratchpad}zGot unexpected prompt type ) rpromptinput_variablesloggerwarningappendrnrtemplaterrtyperX)r7r5r\s r8validate_promptzAgent.validate_prompt3s&& V%;%; ; NN=   " " ) )*< =&.1#99  F$9: !77  4DL>B o% r;cy)z&Prefix to append the observation with.Nr5r6s r8rzAgent.observation_prefixQrFr;cy)z#Prefix to append the LLM call with.Nr5r6s r8rzAgent.llm_prefixVrFr;cy)zCreate a prompt for this class. Args: tools: Tools to use. Returns: Prompt template. Nr5rarcs r8 create_promptzAgent.create_prompt[rFr;cy)zgValidate that appropriate tools are passed in. Args: tools: Tools to use. Nr5r@s r8_validate_toolszAgent._validate_toolsgrFr;c y)z)Get default output parser for this class.Nr5)rarDs r8_get_default_output_parserz Agent._get_default_output_parserorFr;c |j|t||j||}|Dcgc]}|j}}|xs|j } |d||| d|Scc}w)aHConstruct an agent from an LLM and tools. Args: llm: Language model to use. tools: Tools to use. callback_manager: Callback manager to use. output_parser: Output parser to use. kwargs: Additional arguments. Returns: Agent object. )rbr5rd)rrrr5)rCr-rAnamerE) rarbrcrdrrDrtool tool_names_output_parsers r8rezAgent.from_llm_and_toolsts* E"$$U+- -22Ddii2 2&J#*H*H*J $(    3sA(c |dk(rtddidS|dk(rd}|D]5\}}||jz }|d|j|d|jz }7|dz }||jd}i||}|j j d i|} |jj| } t| tr| Std| i| Sd |} t| ) aReturn response when agent has been stopped due to max iterations. Args: early_stopping_method: Method to use for early stopping. intermediate_steps: Steps the LLM has taken to date, along with observations. **kwargs: User inputs. Returns: Agent finish object. Raises: ValueError: If `early_stopping_method` is not in ['force', 'generate']. rRr4rSrTgeneraterzB I now need to return a final answer based on the previous steps:r-zBearly_stopping_method should be one of `force` or `generate`, got r5) rrrrrrr%rrrnrX) r7rZrBrDrr r!r/r'r( parsed_outputr\s r8r]zAgent.return_stopped_responses"( !G +PQ  !J .H'9 # FJJ&001+b@QR  V H/7 KJ2V2z2K0$..00?;?K ..44[AM-5$$+6 D D () + or;c4|j|jdS)rrrr6s r8rzAgent.tool_run_logging_kwargss//"&"9"9  r;rrr)rBrrzstr | list[BaseMessage]rr)rBrrDr rzbuiltins.dict[str, Any]rr&r)rcrrr)rcrrr)rDr rr)NN) rbrrcrrdrrzAgentOutputParser | NonerDr rr)rZrorBrrDr rrr)rrrrrrr'rsr>rr9rr"rErIr$rOr%r<rrrrrArCrErer]rrrs@r8rrs< $$$)&*M#*H  "  9  !  $5955 5 # 50 $<9<< < # <*(9(( ! (&KK'"#:5522    88 8<26 ! ! "! 5 ! 0 !  !  ! ! F5"5:5 5  5n r;rcreZdZUdZdZded< dZded< e d d dZe d d d Z y) ExceptionToolz!Tool that just returns the query. _ExceptionrorGzException tool descriptionNc|Srr5r7query run_managers r8_runzExceptionTool._runs  r;cK|Swrr5rUs r8_arunzExceptionTool._aruns  sr)rVrorWz CallbackManagerForToolRun | Nonerro)rVrorWz%AsyncCallbackManagerForToolRun | Nonerro) rrrrrGrrSr'rXrZr5r;r8rQrQs+D#'K'" 9=6 >B; r;rQceZdZUdZded< ded< dZded< d Zd ed < d Zd ed< dZded< dZ ded< dZ ded< e d5 d6dZ e dd7dZe de d8dZed9dZed:dZd:dZ d5ddd  d;d!Zedd%Z d5 d?d&Z d5 d@d'Z dAd(Z d5 dBd)Z d5 dCd*Z d5 dDd+Z d5 dEd,Z d5 dFd-Z! d5 dGd.Z" d5 dHd/Z# d5 dId0Z$ dJd1Z% dKd2Z&e d5 dLd3Z'e d5 dMd4Z(y )N AgentExecutorzAgent that is using tools.7BaseSingleActionAgent | BaseMultiActionAgent | RunnableagentrrcFrreturn_intermediate_stepsz int | Nonemax_iterationsNz float | Nonemax_execution_timerRrorZz3bool | str | Callable[[OutputParserException], str]handle_parsing_errorszNint | Callable[[list[tuple[AgentAction, str]]], list[tuple[AgentAction, str]]]trim_intermediate_stepsc |d|||d|S)zCreate from agent and tools. Args: agent: Agent to use. tools: Tools to use. callbacks: Callbacks to use. kwargs: Additional arguments. Returns: Agent executor object. )r^rcrCr5r5)rar^rcrCrDs r8from_agent_and_toolsz"AgentExecutor.from_agent_and_tools"s)&     r;r1r2c|j}|j}|j}|Rt||Dchc]}|jc}k7r,d|d|Dcgc]}|jc}d}t ||Scc}wcc}w)zValidate that tools are compatible with agent. Args: values: Values to validate. Returns: Validated values. Raises: ValueError: If allowed tools are different than provided tools. zAllowed tools (z!) different than provided tools ())r^rcr>rrGrX)r7r^rcrrHr\s r8validate_toolszAgentExecutor.validate_tools<s  //1  $]);"'@ DII@ * "-1#:?#@$DII#@"AD S/ ! @ $As A?B beforecd|jd}|rjt|trZ |j}|tt t zk(}|jdd}|rt|||d<|St|||d<|S#t$rd}YEt$rtjdd}YfwxYw)zConvert runnable to agent if passed in. Args: values: Values to validate. Returns: Validated values. r^Fz.Unexpected error getting OutputType from agentrT)rr)getrnr OutputTyperr r TypeError Exceptionr7 exceptionpoprr)ravaluesr^ output_type multi_actionrs r8validate_runnable_agentz%AgentExecutor.validate_runnable_agentVs 7# Zx0 N#..  +d;.?+.MM $jj):DAO":"$3#w #0"$3#w ' %$  %  !QR$  %s B B/ B/.B/czt|jtrtd|jS|jS)aCType cast self.agent. If the `agent` attribute is a Runnable, it will be converted one of RunnableAgentType in the validate_runnable_agent root_validator. To support instantiating with a Runnable, here we explicitly cast the type to reflect the changes made in the root_validator. RunnableAgentType)rnr^rr r6s r8 _action_agentzAgentExecutor._action_agentzs. djj( ++TZZ8 8zzr;cd}t|)zRaise error - saving not supported for Agent Executors. Args: file_path: Path to save to. Raises: ValueError: Saving not supported for agent executors. zpSaving not supported for agent executors. If you are trying to save the agent, please use the `.save_agent(...)`)rX)r7rr\s r8rzAgentExecutor.saves ! or;c8|jj|S)zZSave the underlying agent. Args: file_path: Path to save to. )ryr)r7rs r8 save_agentzAgentExecutor.save_agents !!&&y11r;)include_run_infoasync_c6t||||j|S)aPEnables iteration over steps taken to reach final output. Args: inputs: Inputs to the agent. callbacks: Callbacks to run. include_run_info: Whether to include run info. async_: Whether to run async. (Ignored) Returns: Agent executor iterator object. )tagsr})r)r)r7rrCr}r~s r8iterzAgentExecutor.iters$&%   -   r;c.|jjSr)ryrOr6s r8rOzAgentExecutor.input_keyss!!,,,r;cz|jrg|jjdS|jjS)zReturn the singular output key.rB)r_ryr9r6s r8 output_keyszAgentExecutor.output_keyss<  ) )LT''55L7KL L!!///r;c\|jDcic]}|j|c}|Scc}w)zwLookup tool by name. Args: name: Name of tool. Returns: Tool object. )rcrG)r7rGrHs r8 lookup_toolzAgentExecutor.lookup_tools)-1JJ7D 47==7s)cx|j||jk\ry|jduxs||jkS)NF)rarb)r7 iterations time_elapseds r8_should_continuezAgentExecutor._should_continues?    *zT=P=P/P&&$.X,AXAX2XXr;c|r|j|d|j|j}|jr||d<|SNgreen)colorverboserBon_agent_finishrr9r_r7r4rBrWrs r8_returnzAgentExecutor._returnsE   ' 'gt|| ' T++  ) )1CL- .r;cK|r&|j|d|jd{|j}|jr||d<|S7#wrrrs r8_areturnzAgentExecutor._areturnsa -- .   ++  ) )1CL- . s$A A $A ct|dtr t|dk7r d}t||dS|Dcgc]+}t|ts|j |j f-c}Scc}w)Nrdz>Expected a single AgentFinish output, but got multiple values.)rnrrrXrr r!)r7rsr\as r8_consume_next_stepz AgentExecutor._consume_next_stepsa fRj+ .6{aV o%": 39VaZ9=U1==)VVVs A+A+c \|jt|j|||||Sr)rr_iter_next_step)r7name_to_tool_map color_mappingrrBrWs r8_take_next_stepzAgentExecutor._take_next_steps<&& $$$!&    r;c#tK |j|}|jj|fd|r|jndi|}t |t,r|yt |tr|gn|} | D]}|| D]}|j/||||y#t$r}t |j tr|j }nd}|rd|} t| |t|} t |j tr:|jr+t|j} t|j} n_d} n\t |j tr |j } n5t|j r|j |} nd} t| |td| | }|r|j|d |jj!} t#j$|j&f|j(d|r|jndd | } t+|| Yd}~yd}~wwxYww) Take a single step in the thought-action-observation loop. Override this to take control of how the agent makes and acts on choices. rCNFAn output parsing error occurred. In order to pass this error back to the agent and have it try again, pass `handle_parsing_errors=True` to the AgentExecutor. This is the error: Invalid or incomplete response.Got unexpected type of `handle_parsing_errors`rRrrrrrCr r!)_prepare_intermediate_stepsryrE get_childrrnrcrrXro send_to_llmr! llm_outputcallabler on_agent_actionrrQr  tool_inputrrr_perform_agent_action)r7rrrrBrWr4e raise_errorr\rr!tool_run_kwargsactions agent_actions r8rzAgentExecutor._iter_next_steps=0 !%!A!ABT!U -T'',,"5@+//1dF^ fk *L )=6(6# L  # L,,    e%' $44d;"&"<"<< # *+,0 !o1,q6D$44d;=="%amm"4Kq||,D"BKD66<"88 $445"88; F o1, {DAF++F'+B"00HHJO--/--!! 5@+//1d  " K6{C C O' s1H8ABAH8 H5#FH0+H80H55H8cf|r|j|d|j|vr||j}|j}||j}|jj }|rd|d<|j |j f|j||r|jndd|} ny|jj }tj |jt|jdf|jd|r|jndd|} t|| S)NrrrTrrrequested_tool_nameavailable_tool_namesr) rrH return_directryrr rrrr+rkeysr r7rrrrWrHrrrr!s r8rz#AgentExecutor._perform_agent_actionds;   ' ' G ' D    0 0#L$5$56D ..M!,"3"34E"00HHJO02 -"$(('' 5@+//1d  " K#00HHJO++-+++7+<+<,01A1F1F1H,I   5@+//1d " K +FFr;c K|j|j|||||2cgc3d{}| 76c}Scc}wwr)r_aiter_next_step)r7rrrrBrWrs r8_atake_next_stepzAgentExecutor._atake_next_steps]&& $44$!&         s&#A=5 35 =5 = Ac K |j|}|jj|fd|r|jndi|d{}t |t*r|yt |tr|gn|} | D]}| t-j.| Dcgc]}|j1||||c}d{}|D]}| y7#t$r}t |j tr|j }nd}|rd|} t| |t|} t |j tr:|jr+t|j} t|j} n_d} n\t |j tr |j } n5t|j r|j |} nd} t| |td| | }|jj} t!j"|j$f|j&d|r|jndd| d{7} t)|| Yd}~yd}~wwxYwcc}w7w) rrCNFrrrrRrr)rryrIrrrnrcrrXrorr!rrr rrQr rrrrasynciogather_aperform_agent_action)r7rrrrBrWr4rrr\rr!rrrresultrs r8rzAgentExecutor._aiter_next_stepst. !%!A!ABT!U 44--33"5@+//1dFZ fk *L )=6(6# L  ~~%,! ++$!      EK I %% $44d;"&"<"<< # *+,0 !o1,q6D$44d;=="%amm"4Kq||,D"BKD66<"88 $445"88; F o1, {DAF"00HHJO 4  4 4!!! 5@+//1d ! " !K6{C C K% f  sgI*AC C  CA I*I"3I*:I';I* C IE'I?III*I I*cK|r&|j||jdd{|j|vr||j}|j}||j}|jj }|rd|d<|j |jf|j||r|jndd|d{} n|jj }tj |jt|jdf|jd|r|jndd|d{} t|| S7277w)Nr)rrrTrrrr) rrrHrryrr rrr+rrrrs r8rz$AgentExecutor._aperform_agent_actionst -- .      0 0#L$5$56D ..M!,"3"34E"00HHJO02 - ) ''! 5@+//1d ! " !K#00HHJO 2 2 2+7+<+<,01A1F1F1H,I !  5@+//1d !" ! K +FFC  s5$EEBEEBEEEEEc|jDcic]}|j|}}t|jDcgc]}|jc}ddg}g}d}d}tj} |j ||r|j |||||} t | tr|j| ||S|j| t| dk(r,| d} |j| } | |j| ||S|dz }tj| z }|j ||r|jj|j|fi|} |j| ||Scc}wcc}w)z(Run text through and get agent response.rredexcluded_colorsrrWr)rcrGr"timerrrnrrextendr_get_tool_returnryr]rZr7rrWrHrrrBrr start_timenext_step_outputnext_step_action tool_returnr4s r8_callzAgentExecutor._call"s9= CDIItOCC)#':: .4TYY .$e, =?  YY[ ##J =#33 "' 4  *K8||$& +$  % %&6 7#$)#3A#6 "334DE *<<#*$/( !OJ99;3L7##J =8<##;;  & &    ||F$6K|PP[D /s E2E7cK|jDcic]}|j|}}t|jDcgc]}|jc}dg}g}d}d}tj} t |j 4d{|j ||r|j|||||d{} t| tr-|j| ||d{cdddd{S|j| t| dk(rE| d} |j| } | -|j| ||d{cdddd{S|dz }tj| z }|j ||r|jj|j |fi|} |j| ||d{cdddd{Scc}wcc}w7d7777777/7!#1d{7swYyxYw#t"t$j"f$rH|jj|j |fi|} |j| ||d{7cYSwxYww)z.Async run text through and get agent response.rrrrNrr)rcrGr"rr.rbrrrnrrrrrryr]rZ TimeoutErrorrrs r8_acallzAgentExecutor._acallXs9= CDIItOCC)#':: .4TYY .$I =?  YY[ 4 &t'>'>?' ' ++J E-1-B-B(%*$/ .C.($""2K@%)]],.(3&3& ' ' '  '--.>?+,1+;A+>(&*&;&;4B7B>BBr) r^r]rcrrCrrDr rr\rO)rsrsrr )rz,BaseSingleActionAgent | BaseMultiActionAgentr) rr rCrr}rr~rrr)r)rGrorr!)rrrfloatrr)r4rrBrrW!CallbackManagerForChainRun | Nonerdict[str, Any])r4rrBrrW&AsyncCallbackManagerForChainRun | Nonerr)rsNextStepOutputr+AgentFinish | list[tuple[AgentAction, str]]) rdict[str, BaseTool]rdict[str, str]rrrBrrWrrr) rrrrrrrBrrWrrz/Iterator[AgentFinish | AgentAction | AgentStep]) rrrrrr rWrrr) rrrrrrrBrrWrrr) rrrrrrrBrrWrrz4AsyncIterator[AgentFinish | AgentAction | AgentStep]) rrrrrr rWrrr)rrrWrrr)rrrWrrr)rztuple[AgentAction, str]rzAgentFinish | None)rBrrr)rdict[str, Any] | AnyrRunnableConfig | NonerDr rzIterator[AddableDict])rrrrrDr rzAsyncIterator[AddableDict]))rrrrrr_rarbrZrcrerrgr%rjrvrryr'rr|rrOrrrrrrrrrrrrrrrrrrr5r;r8r\r\s$ BB3 -&+t+3!#NJ#9(, +")3(RWNV V  $  F "        2'"#2(# $ D   2 $ "'          6--00 >Y:>  ! 7   "?C !<   " W W 5 W":>  - &   :  7   5 4:> M-M&M M : M 7 M 9Mh:> %G-%G&%G" %G 7 %G  %GZ?C  - &   :  <   5 6?C U-U&U U : U < U >Ux?C )G-)G&)G" )G < )G  )G\:>4Q4Q74Q  4Qr?CGG<G  GR1 * "9 " ' ")-#&   <)-#&  $ r;r\)`r __future__rrbuiltinsrrloggingrabcrcollections.abcrrrrpathlibr typingr r rlangchain_core._apir langchain_core.agentsr rrlangchain_core.callbacksrrrrrrlangchain_core.exceptionsrlangchain_core.language_modelsrlangchain_core.messagesrlangchain_core.output_parsersrlangchain_core.promptsrlangchain_core.prompts.few_shotrlangchain_core.prompts.promptrlangchain_core.runnablesrrrlangchain_core.runnables.utilsr langchain_core.toolsr!langchain_core.utils.inputr"pydanticr#r$r%typing_extensionsr&r'"langchain_classic._api.deprecationr('langchain_classic.agents.agent_iteratorr)$langchain_classic.agents.agent_typesr*langchain_classic.agents.toolsr+langchain_classic.chains.baser,langchain_classic.chains.llmr-#langchain_classic.utilities.asyncior. getLoggerrr7r0rrrrrrrrrQrrxr\r5r;r8rsK"  GG  *EE<</:5A8LL6)8;;,HI:6/1?   8 $cIcLI9IX3({)BC3 T+&45 (i)iXn3nb  %  Q 0Q  Q h  %  T !T  T nH2kK/);<!$<<L EL r;