`validate()`

Parse a source into a target type and then validate it against one or more constraints. You can choose to raise on violations or receive a structured result.

API Reference

validate()

Synchronously parse a source into a target type, then validate the result against constraints.

Parameters:

Name	Type	Description	Default
`source`	`SourceParam`	The value to parse and validate.	required
`target`	`TargetParam[Output]`	The type or schema to parse into. Defaults to str.	`str`
`context`	`ContextType \| List[ContextType] \| None`	Optional context or conversation history. Defaults to None.	`None`
`constraints`	`List[str] \| None`	List of constraint strings to validate against. Defaults to None.	`None`
`raise_on_error`	`bool`	If True (default), raise AssertionError if validation fails. If False, return a ValidationResult with any violations. Defaults to True.	`True`
`confidence`	`bool`	If True, includes confidence scoring. Defaults to False.	`False`
`model`	`ModelParam`	The model to use for parsing/validation. Defaults to "openai:gpt-4o-mini".	`'openai:gpt-4o-mini'`
`model_settings`	`PydanticAIModelSettings \| None`	Model settings to use for the operation. Defaults to None.	`None`
`instructions`	`PydanticAIInstructions \| None`	Additional instructions for the model. Defaults to None.	`None`
`attachments`	`AttachmentType \| List[AttachmentType] \| None`	Attachments to provide to the model. Defaults to None.	`None`
`tools`	`ToolType \| List[ToolType] \| None`	Tools available to the model. Defaults to None.	`None`
`deps`	`Deps \| None`	Optional RunContext dependencies. Defaults to None.	`None`
`usage_limits`	`PydanticAIUsageLimits \| None`	Usage limits for the model call. Defaults to None.	`None`
`observe`	`bool \| Observer \| None`	If True or provided, enables CLI observation output.	`None`

Returns:

Type	Description
`Result[Output] \| ValidationResult[Output]`	Result[Output] \| ValidationResult[Output]: Result[Output] if raise_on_error is True and validation passes. ValidationResult[Output] if raise_on_error is False (always includes violations).

avalidate()

Asynchronously parse a source into a target type, then validate the result against constraints.

Parameters:

Name	Type	Description	Default
`source`	`SourceParam`	The value to parse and validate.	required
`target`	`TargetParam[Output]`	The type or schema to parse into. Defaults to str.	`str`
`context`	`ContextType \| List[ContextType] \| None`	Optional context or conversation history. Defaults to None.	`None`
`constraints`	`List[str] \| None`	List of constraint strings to validate against. Defaults to None.	`None`
`raise_on_error`	`bool`	If True (default), raise AssertionError if validation fails. If False, return a ValidationResult with any violations. Defaults to True.	`True`
`confidence`	`bool`	If True, includes confidence scoring. Defaults to False.	`False`
`model`	`ModelParam`	The model to use for parsing/validation. Defaults to "openai:gpt-4o-mini".	`'openai:gpt-4o-mini'`
`model_settings`	`PydanticAIModelSettings \| None`	Model settings to use for the operation. Defaults to None.	`None`
`instructions`	`PydanticAIInstructions \| None`	Additional instructions for the model. Defaults to None.	`None`
`attachments`	`AttachmentType \| List[AttachmentType] \| None`	Attachments to provide to the model. Defaults to None.	`None`
`tools`	`ToolType \| List[ToolType] \| None`	Tools available to the model. Defaults to None.	`None`
`deps`	`Deps \| None`	Optional RunContext dependencies. Defaults to None.	`None`
`usage_limits`	`PydanticAIUsageLimits \| None`	Usage limits for the model call. Defaults to None.	`None`
`observe`	`bool \| Observer \| None`	If True or provided, enables CLI observation output.	`None`

Returns:

Type	Description
`Result[Output] \| ValidationResult[Output]`	Result[Output] \| ValidationResult[Output]: Result[Output] if raise_on_error is True and validation passes. ValidationResult[Output] if raise_on_error is False (always includes violations).

Overview

The validate() semantic operation parses a source into a target type and then validates it against one or more constraints. You can choose to raise on violations or receive a structured result.

SyncAsync

Parsing + Constraint Validation

from zyx import validate
from pydantic import BaseModel


class User(BaseModel):
    name: str
    age: int


result = validate(
    source="Sam is 15 years old.",
    target=User,
    constraints=[
        "age must be at least 18",
    ],
    raise_on_error=False,
)


print(result.output)
print(result.violations.violations)
"""
User(name="Sam", age=15)
[ConstraintViolation(constraint="age must be at least 18", reason="age is 15")]
"""

Parsing + Constraint Validation

from zyx import avalidate
from pydantic import BaseModel


class User(BaseModel):
    name: str
    age: int


async def main():
    result = await avalidate(
        source="Sam is 15 years old.",
        target=User,
        constraints=[
            "age must be at least 18",
        ],
        raise_on_error=False,
    )

Raise on Error

Set raise_on_error=True to raise an AssertionError when any constraint is violated. Set it to False to receive a ValidationResult with violations instead.