Help / Question - ModelFactory Hardcode

[AsgerJ95]

When using ModelFactory for testing purposes model_validators return an error due to the random values.

Is it possible to hard code certain BaseModels similar to what is described in 1.4.1 (Old) polyfactory doc., but without the assert from model_validators?

It seems like the it generates random values and then overwrites it with the hard coded input. Which will still return the assert.

Old documentation: [https://pypi.org/project/pydantic-factories/1.4.1/]

Example:
When using polyfactory - ModelFactory for generating random values for testing, the PeriodBase class the model_validator,

In this case a bunch of schedules should be hardcoded, but the model_validator will still return the assert.

class PeriodBase(BaseModel):
    start_time: int = Field(ge=0, le=24, description="Starting time of period in decimal hour [h]")
    end_time: int = Field(ge=0, le=24, description="End time of period in decimal hour [h]")

    @model_validator(mode='after')
    def check_times(self) -> "PeriodBase":
        assert self.start_time < self.end_time, "Start time should come before end time"
        return self

class PeriodFraction(PeriodBase):
    value: float = Field(ge=0., le=1., description="Fractional value [-]")

[guacs]

Hey @AsgerJ95, you can set hard coded inputs by setting a field with the same name on the factory class. The relevant docs are here. For this specific use case, I would recommend using PostGenerated fields which are documented here. Basically, PostGenerated fields allow for specifying the value of a field based on the generated value of another field.

Let me know if that works for you!