How to generate a strict json schema with pydantic?

Question

I recently started to use pydandic to generate JSON schemas for validating data but I found the by default the generated schema does not complain about unknown keys inside my BaseModel.

Example:

class Query(BaseModel):
    id: str
    name: Optional[str]

The generated schema would pass validation even if the object has other attributes than the two one mentioned here.

How can I assure validation will fail if someone adds a "foo: bar" property?

score 10 · Accepted Answer · answered Jan 15 '21 at 14:37

10

You need to use a configuration on your model:

from pydantic import BaseModel, Extra
class Query(BaseModel):
    id: str
    name: Optional[str]
    class Config:
        extra = Extra.forbid

It defaults to Extra.ignore, the other option is Extra.allow which adds any extra fields to the resulting object.

You can also just use the strings "ignore", "allow", or "forbid"

Here are all the model config options you can use:

https://pydantic-docs.helpmanual.io/usage/model_config/

answered Jan 15 '21 at 14:37

juanpa.arrivillaga

88,713
10
131
172

1

That was quick and awesome! Thanks. I tried searching for strict in docs before posting it, but no luck. – sorin Jan 15 '21 at 14:55
Maybe we also know how to deal with https://github.com/samuelcolvin/pydantic/issues/2266 ? – sorin Jan 15 '21 at 14:57
@sorin you can use `a_valid_identifier: Annotation = Field(alias="not a valid-identifier")`. – juanpa.arrivillaga Jan 15 '21 at 14:59

How to generate a strict json schema with pydantic?

1 Answers1