"jf"SSKJr SSKrSSKJrJrJrJr SSKJ r SSK J r J r J r JrJr SSKJrJr SSKJr SSKJr SS KJrJr S S KJrJr S S KJr S S KJ r S SK!J"r" S SK#J$r$J%r% Sr&\"S5r'\"SSS9r(\"S5r)\"S\*SS9r+\Sr,\Sr-\"S\.\(\S\\(\(-4-\(4S9r/\"S\\ \ \*/\\(\(-4\\*/\\(\(-4-\(4S9r0\ "SS9"S S!\ \+55r1\ "SS9"S"S#\ \+55r2\ "SS9"S$S%\ \+55r3\ "S&S'55r4\ "S(S)55r5\ "S*S+\ \+55r6S:S;S,jjr7\ "SS9"S-S.\ \(55r8SS4jj5r;\SSS2.S?S5jj5r;SSS2.S@S6jjr;\"S7\/\(\1\(-\2\(-\3\(-\6\(-\(4S9r<\"S8\<\(\S9-\(4S9r=g)A) annotationsN) AwaitableCallableMappingSequence) dataclass)AnyClassVarGenericLiteraloverload)GetCoreSchemaHandlerGetJsonSchemaHandler)JsonSchemaValue) core_schema) TypeAliasTypeTypeVar)_utils exceptions)InlineDefsJsonSchemaTransformer) RunContext) ToolCallPart)ObjectJsonSchemaToolDefinition) ToolOutput NativeOutputPromptedOutput TextOutputStructuredDictChoiceChoicesOutputObjectDefinition OutputContext OutputDataT OutputModeStructuredOutputMode OutputSpecOutputTypeOrFunctionTextOutputFuncTT_coT) covariant ChoiceValueTr%)defaultr-)texttoolnativeprompted tool_or_textimageauto)r1r2r3r).) type_paramsr*F)initc\rSrSr%SrS\S'S\S'S\S'S\S 'S \S 'S \S 'SSSSSS.SSjjrSrg)rRaKMarker class to use a tool for output and optionally customize the tool. Example: ```python {title="tool_output.py"} from pydantic import BaseModel from pydantic_ai import Agent, ToolOutput class Fruit(BaseModel): name: str color: str class Vehicle(BaseModel): name: str wheels: int agent = Agent( 'openai:gpt-5.2', output_type=[ ToolOutput(Fruit, name='return_fruit'), ToolOutput(Vehicle, name='return_vehicle'), ], ) result = agent.run_sync('What is a banana?') print(repr(result.output)) #> Fruit(name='banana', color='yellow') ``` !OutputTypeOrFunction[OutputDataT]output str | Nonename description int | None max_retries bool | Nonestrictbool sequentialNF)r>r?rArCrEcUbUS:a[R"SU35eXlX lX0lX@lXPlX`lg)Nrzmax_retries must be >= 0, got )r UserErrorr<r>r?rArCrE)selftype_r>r?rArCrEs O/home/mande/repo/quber/.venv/lib/python3.13/site-packages/pydantic_ai/output.py__init__ToolOutput.__init__sI  "{Q&&)G }'UV V  && $)r?rAr>r<rErC) rIr;r>r=r?r=rAr@rCrBrErD__name__ __module__ __qualname____firstlineno____doc____annotations__rK__static_attributes__rMrJrrRs@ .-% zU 2 "&"&" %0% %  %  %%%%rMrc\rSrSr%SrS\S'S\S'S\S'S\S 'S \S 'S S S S S .SSjjrSrg )ra[Marker class to use the model's native structured outputs functionality for outputs and optionally customize the name and description. Example: ```python {title="native_output.py" requires="tool_output.py"} from pydantic_ai import Agent, NativeOutput from tool_output import Fruit, Vehicle agent = Agent( 'openai:gpt-5.2', output_type=NativeOutput( [Fruit, Vehicle], name='Fruit or vehicle', description='Return a fruit or vehicle.' ), ) result = agent.run_sync('What is a Ford Explorer?') print(repr(result.output)) #> Vehicle(name='Ford Explorer', wheels=4) ``` OOutputTypeOrFunction[OutputDataT] | Sequence[OutputTypeOrFunction[OutputDataT]]outputsr=r>r?rBrCstr | Literal[False] | NonetemplateN)r>r?rCr\c@XlX lX0lX@lXPlgN)rZr>r?rCr\)rHrZr>r?rCr\s rJrKNativeOutput.__init__s  &  rM)r?r>rZrCr\) rZrYr>r=r?r=rCrBr\r[rNrVrMrJrrs,]\( vB N)) "&"04 !` ! !  !  !. ! !rMrcn\rSrSr%SrS\S'S\S'S\S'S\S 'S S S S .SS jjrS rg )raMarker class to use a prompt to tell the model what to output and optionally customize the prompt. Example: ```python {title="prompted_output.py" requires="tool_output.py"} from pydantic import BaseModel from pydantic_ai import Agent, PromptedOutput from tool_output import Vehicle class Device(BaseModel): name: str kind: str agent = Agent( 'openai:gpt-5.2', output_type=PromptedOutput( [Vehicle, Device], name='Vehicle or device', description='Return a vehicle or device.' ), ) result = agent.run_sync('What is a MacBook?') print(repr(result.output)) #> Device(name='MacBook', kind='laptop') agent = Agent( 'openai:gpt-5.2', output_type=PromptedOutput( [Vehicle, Device], template='Gimme some JSON: {schema}' ), ) result = agent.run_sync('What is a Ford Explorer?') print(repr(result.output)) #> Vehicle(name='Ford Explorer', wheels=4) ``` rYrZr=r>r?r[r\N)r>r?r\c4XlX lX0lX@lgr^)rZr>r?r\)rHrZr>r?r\s rJrKPromptedOutput.__init__ s  & rM)r?r>rZr\)rZrYr>r=r?r=r\r[rNrVrMrJrrsr'R]\( vi)) "&04 !` ! !  ! . ! !rMrcN\rSrSr%SrS\S'SrS\S'SrS\S'SrS \S 'S r g) r#izEDefinition of an output object used for structured output generation.r json_schemaNr=r>r?rBrCrV) rOrPrQrRrSrTr>r?rCrUrVrMrJr#r#s)O!!D*"K"FKrMr#c\rSrSr%SrS\S'S\S'S\S'S \S 'S rS \S 'S rS\S'S rS\S'Sr S \S'Sr S \S'Sr S \S'Sr g )r$i%zAContext about the output being processed, passed to output hooks.r&modeztype[Any] | None output_typezOutputObjectDefinition | None object_defrD has_functionNr= function_namezToolCallPart | None tool_callzToolDefinition | Nonetool_defF allows_text allows_imageallows_deferred_toolsrV) rOrPrQrRrSrTrkrlrmrnrorprUrVrMrJr$r$%sK  "!C--YI $M:$>%)I")D&*H#*tKlL$2"'4'FrMr$c$\rSrSr%SrS\S'Srg)riEaMarker class to use text output for an output function taking a string argument. Example: ```python from pydantic_ai import Agent, TextOutput def split_into_words(text: str) -> list[str]: return text.split() agent = Agent( 'openai:gpt-5.2', output_type=TextOutput(split_into_words), ) result = agent.run_sync('Who was Albert Einstein?') print(result.output) #> ['Albert', 'Einstein', 'was', 'a', 'German-born', 'theoretical', 'physicist.'] ``` !!! note When streaming, [`stream_text()`][pydantic_ai.result.StreamedRunResult.stream_text] does not apply the wrapped function. Use [`stream_output()`][pydantic_ai.result.StreamedRunResult.stream_output] to stream the value it produces. zTextOutputFunc[OutputDataT]output_functionrVN)rOrPrQrRrSrTrUrVrMrJrrEs410FrMrc^[R"T5mST;a5[T5R5mST;a[R "S5eU(aUTS'U(aUTS'"U4SjS[ 5nU$)aReturns a `dict[str, Any]` subclass with a JSON schema attached that will be used for structured output. Args: json_schema: A JSON schema of type `object` defining the structure of the dictionary content. name: Optional name of the structured output. If not provided, the `title` field of the JSON schema will be used if it's present. description: Optional description of the structured output. If not provided, the `description` field of the JSON schema will be used if it's present. Example: ```python {title="structured_dict.py"} from pydantic_ai import Agent, StructuredDict schema = { 'type': 'object', 'properties': { 'name': {'type': 'string'}, 'age': {'type': 'integer'} }, 'required': ['name', 'age'] } agent = Agent('openai:gpt-5.2', output_type=StructuredDict(schema)) result = agent.run_sync('Create a person') print(result.output) #> {'name': 'John Doe', 'age': 30} ``` z$defsz`StructuredDict` does not currently support recursive `$ref`s and `$defs`. See https://github.com/pydantic/pydantic/issues/12145 for more information.titler?c`>\rSrSrSr\SSj5r\SU4Sjj5rSrg) 'StructuredDict.._StructuredDictiTcx[R"[R"5[R"5S9$)N) keys_schema values_schema)r dict_schema str_schema any_schemacls source_typehandlers rJ__get_pydantic_core_schema__DStructuredDict.._StructuredDict.__get_pydantic_core_schema__s/**'224)446 rMc>T$r^rVr~rrres rJ__get_pydantic_json_schema__DStructuredDict.._StructuredDict.__get_pydantic_json_schema__s  rMrVNrr rrreturncore_schema.CoreSchemarrrrrr) rOrPrQrR__is_model_like__ classmethodrrrU)resrJ_StructuredDictrvs_   ! ,@  #     4 ?S     rMr)rcheck_object_json_schemarwalkrrGr)rer>r?rs` rJr r es:11+>K+5kBGGI k !&&i  # G%0 M"/$ rMc\rSrSr%SrS\S'S\S'\SSSjj5r\SSS jj5r\SSS jj5r\SSS jj5rS\RS .SS jjjrSr g)r!iaqOne choice in a [`Choices`][pydantic_ai.output.Choices] set: what it means, and what it stands for. Building a set from descriptions alone yields the key the model picked, so `Choice` is only reached for when a choice stands for something other than its key: ```python {title="choice.py"} from pydantic_ai import Agent, Choice, Choices Card = Choices( { 'visa': Choice('Any card starting with a 4.', value=4), 'amex': Choice('Any card starting with a 3.', value=3), } ) agent = Agent('openai:gpt-5.2', output_type=Card) result = agent.run_sync('4111 1111 1111 1111') print(result.output) #> 4 ``` r=r?r,valueNcgr^rV)rHr?s rJrKChoice.__init__sMPrMcgr^rVrHr?rs rJrKrrMcgr^rVrs rJrKrrrMcgr^rVrs rJrKrsnqrM)rcr[U5(a[R"SU<S35eXlX lg)NzSA callable `Choice` value is called with no arguments when the model picks it, but z@ requires some. Bind them with `functools.partial` or a closure.)_requires_argumentsrrGr?rrs rJrKrs@ u % %&&i_a ' rM)r?rr^)rHz Choice[str]r?r=rNone)rHChoice[ChoiceValueT]r?r=rz%Callable[[], Awaitable[ChoiceValueT]]rr)rHrr?r=rzCallable[[], ChoiceValueT]rr)rHrr?r=rr.rr)r?r=rr rr) rOrPrQrRrSrTr rKrUNSETrUrVrMrJr!r!s,N KPP >B"1;Ns >B"1;Nh qqv||rMr!c[U5(dg[R"U5RR 5n[SU55$![ [ 4a gf=f)zLWhether `value` is a callable that would fail when called with no arguments.Fc3# UHTnURURL=(a0 URURURUR 4;v MV g7fr^)r/emptykindPOSITIONAL_ONLYPOSITIONAL_OR_KEYWORD KEYWORD_ONLY).0 parameters rJ &_requires_arguments..sZ$I Y__, s NNy88):Y:Y[d[q[qr r s#sAA)callableinspect signature parametersvalues TypeError ValueErrorany)rrs rJrrsi E??&&u-88??A  $  z "s.AA&%A&cF\rSrSr%SrS\S'S\S'\S Sj5rSrg ) _ChoicesActionsiaiBase class of the type [`Choices`][pydantic_ai.output.Choices] returns when a choice value is callable. Such a set is only meaningful as an agent's `output_type`, where `_output.py` validates the pick against `keys_type` and resolves and calls the value itself. Anywhere else there is nothing to call the action, so the core schema refuses to be built at all. zClassVar[type[Any]] keys_typezClassVar[Mapping[str, Any]]rc\[U[5(a[U[5(aU$g)zN`output` if it is a `Choices` set whose values include callables, else `None`.N) isinstancetype issubclassr)r<s rJof_ChoicesActions.ofs% fd # # 6?(K(KMrMrVN)r<r rztype[_ChoicesActions] | None) rOrPrQrRrSrT staticmethodrrUrVrMrJrrs-#"a ''DrMr)r>r?cgr^rVchoicesr>r?s rJr"r" srMcgr^rVrs rJr"r"srMcgr^rVrs rJr"r"srMc ^ ^ ^ ^^[U[5(a[R"S5e[U[5(aFUR 5VVs0sH(up4U[U[5(a [ U5OU_M* snnOUVs0sH o3[ 5_M snnU(d[R"S5eUR 5VVs0sH9up4U[R"UR5(a UROU_M; snnm [ST R 555m[ST R555m UR 5VVs0sH$up4UR(dMX4R_M& nnnU(aB/nT H4nSU0nURU5=n (aXS'URU5 M6 SUS.mOS[T 5S .mU(aUTS'"U U UU4S jS 5m U=(d S =T lT lT (dT $"U U 4S jS[$5n U=(d S =U lU lU $s snnfs snfs snnfs snnf)aIReturns a type the model can only fill with one of `choices`, each described where it is built. Use it when the set is only known once the run is under way -- the actions available on the screen in front of an agent, the records a search returned. For a set you know when you write the code, use a `Literal` or an `Enum` (with [`UseEnumMemberDocstrings`][pydantic_ai.UseEnumMemberDocstrings] to describe its members), which give you exhaustiveness checking that a run-time set cannot. Like [`StructuredDict`][pydantic_ai.output.StructuredDict] it returns a type, so it works as an `output_type`, as a field of a Pydantic model, and as a tool parameter. Args: choices: The choices, as a sequence of keys, a mapping from key to its description, or a mapping from key to a [`Choice`][pydantic_ai.output.Choice] carrying both a description and what the key stands for. name: Name of the output tool or structured output. Defaults to `'Choices'`. description: What the model is being asked to pick, e.g. `'Which action to take next.'`. Example: ```python {title="choices.py"} from pydantic_ai import Agent, Choices Intent = Choices( { 'refund': 'The customer wants their money back.', 'replace': 'The customer wants a working unit instead.', 'escalate': 'Nobody on this tier can resolve it.', }, name='customer_intent', description='What the customer is asking for.', ) agent = Agent('openai:gpt-5.2', output_type=Intent) result = agent.run_sync('The blender arrived smashed. Just send me another one.') print(result.output) #> replace ``` zF`Choices` takes a sequence or mapping of choices, not a single string.z'`Choices` requires at least one choice.c3,# UH upX!Lv M g7fr^rV)rkeyrs rJrChoices..XsS=Rzs%*=Rsc38# UHn[U5v M g7fr^)r)rrs rJrrYsJ3I%huoo3Isconstr?string)ranyOf)renumcf>\rSrSr\SUUU4Sjj5r\SU4Sjj5rSrg)Choices.._Choicesilc>[R"[T55nT(a(T(d![R"TRU5nU$r^)rliteral_schemalist no_info_after_validator_function __getitem__)r~rrschema choice_values has_actionsresolves_valuess rJr6Choices.._Choices.__get_pydantic_core_schema__ms?!//]0CDF{$EEmF_F_aghMrMc>[T5$r^)dictrs rJr6Choices.._Choices.__get_pydantic_json_schema__xs $ $rMrVNrr)rOrPrQrRrrrrU)rrrersrJ_Choicesrls^  ! ,@  #      %4 %?S %  %  %rMrr"cB>\rSrSrYrYr\SSj5rSrg)$Choices.._ChoicesWithActionsic.[R"S5e)NaA `Choices` set with a callable `Choice` value can only be used as an agent's `output_type`, as that is the only place Pydantic AI can call the action the model picked. To pick one as a tool parameter or a model field, give the choices plain values and call the action yourself.)rrGr}s rJrAChoices.._ChoicesWithActions.__get_pydantic_core_schema__s&&# rMrVNr) rOrPrQrRrrrrrU)rrsrJ_ChoicesWithActionsrs7   ! ,@  #   rMr)rstrrrGritemsr!ris_setrrrr?getappendrrOrQr)rr>r?rchoiceresolved descriptionsoptionsoptionchoice_descriptionrrrrrers @@@@@rJr"r"s.V'3""#kll gw ' 'X_WdWdWfgWf  63 7 7fVnV CWfg'. /w68mw / ""#LMMV^UcUcUe%UekcV]]6<<88V\\cAUe%MS]=P=P=RSSOJ=3G3G3IJJK?G~~?Of?O SYSeSe+C+++?OLf*, C'.nF%1%5%5c%::!:(:}% NN6 " ! 197'K 'm1DE %0 M"%%$150A AH-   o GKFWiW #6#C K h / %gs/II!AI&.I, I,_OutputSpecItemr(zOutputSpec[T_co])NN)rerr>r=r?r=rztype[JsonSchemaValue])rr rrD)rz!Sequence[str] | Mapping[str, str]r>r=r?r=rz type[str])rzMapping[str, Choice[T_co]]r>r=r?r=rz type[T_co])rz Mapping[str, str | Choice[T_co]]r>r=r?r=rztype[str | T_co])rz/Sequence[str] | Mapping[str, str | Choice[Any]]r>r=r?r=rz type[Any])> __future__rrcollections.abcrrrr dataclassesrtypingr r r r r pydanticrrpydantic.json_schemar pydantic_corertyping_extensionsrrrr _json_schemar _run_contextrmessagesrtoolsrr__all__r+r,r.rr%r&r'rr)r*rrrr#r$rr r!rrr"rr(rVrMrJrs"BB!<<?0%4 9$"3 ( CLv&~& omSDA @Z [  ;<n%DJ#y7M2M)NN]a\c jos #Yt_t%; ; F%F F@VZ@ @(2@HR@@F :WT]::z  , FJfj .9CYc   ?C_c '2<R\   EIei -8BXb " t <t t t  tn D!11L4FFX\I]]`jko`pp DH%788 rM