How do I define a nested Pydantic model with a Tuple containing Optional models?

Question

I am currently working with following nested model (FinalModel):

class Model1(BaseModel):
    count: int

class Model2(BaseModel):
    method: str
    selector: dict
    count: int

class FinalModel(BaseModel):
    slow: bool
    models: Tuple[
        Optional[Model1],
        Optional[Model2],
    ]]

I am expecting the Tuple to have a variable length since I'd like to pass in either Model1 or Model2 or both based on the requirements. However, the tuple seems to be expecting a fixed length of 2.

Please note that Union is not an option for me since I expect number of models to increase in future and the possible combinations will increase as well.

alex_noname · Accepted Answer · 2021-02-09 13:51:50Z

2

To specify a variable-length tuple of multiple types you can use ellipsis literal ... and Union, like below:

class FinalModel(BaseModel):
    slow: bool
    models: Tuple[Union[Model2, Model1], ...]

When using Union one also need to keep in mind that:

However, as can be seen above, pydantic will attempt to 'match' any of the types defined under Union and will use the first one that matches.[...]

As such, it is recommended that, when defining Union annotations, the most specific type is included first and followed by less specific types.

Therefore, for your example, Model2 must precede Model1 in Union, since the objects of Model2 when parsing will match with Model1, and the other members will be discarded. Or use extra = forbid.

class Model1(BaseModel):
    count: int

    class Config:
        extra = "forbid"

edited Feb 9, 2021 at 13:51

answered Feb 9, 2021 at 13:31

alex_noname

33.2k6 gold badges95 silver badges110 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Samuel Colvin · Accepted Answer · 2021-02-09 16:25:26Z

1

Are you sure you want to use a tuple? A tuple is fixed size in its nature, using it in you Pydantic model will require both elements to be present.

Instead, you could use a List[Union[Model1, Model2]]

edited Feb 9, 2021 at 16:25

Samuel Colvin

13.6k7 gold badges64 silver badges78 bronze badges

answered Feb 9, 2021 at 8:31

JarroVGIT

5,6042 gold badges25 silver badges38 bronze badges

1 Comment

shubhendu madhukar Over a year ago

That works as well. I ended up using Tuple itself. As suggested by Eric Jolibois on github, Tuple[Union[Model1, Model2],...]

Collectives™ on Stack Overflow

How do I define a nested Pydantic model with a Tuple containing Optional models?

2 Answers 2

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related