Field Constraints in Pydantic V2

Pydantic V2 now issues a warning when unsupported keywords are used in Field(). This article explains how to set range constraints for integers.

Background

In Pydantic V1, no warning was issued even if minimum and maximum were specified as keyword arguments.

class Score(BaseModel):
  score: int = Field(minimum=0, maximum=20)

Validation does not function; it simply does not issue a warning.

However, running this code in Pydantic V2 displays the following warning:

PydanticDeprecatedSince20: Using extra keyword arguments on `Field` is deprecated
and will be removed. Use `json_schema_extra` instead. (Extra keys: 'minimum', 'maximum').

While the warning suggests using json_schema_extra, you should actually use the official constraint parameters of Pydantic V2.

ge/le Constraint Parameters

In Pydantic V2, specifying ge (greater than or equal) and le (less than or equal) will result in them being converted to minimum and maximum in the JSON Schema.

https://docs.pydantic.dev/2.12/api/standard_library_types/

Constraint	Description	JSON Schema
le	The value must be less than or equal to this number	maximum keyword
ge	The value must be greater than or equal to this number	minimum keyword

import json
from pydantic import BaseModel, Field

class Score(BaseModel):
  score: int = Field(ge=0, le=20)

print(json.dumps(Score.model_json_schema(), indent=2))

Execution Result

{
  "properties": {
    "score": {
      "maximum": 20,
      "minimum": 0,
      "title": "Score",
      "type": "integer"
    }
  },
  "required": [
    "score"
  ],
  "title": "Score",
  "type": "object"
}

In this way, specifying ge=0, le=20 automatically sets "minimum": 0 and "maximum": 20 in the JSON schema.

• ge=0: score >= 0 (0 or more)
• le=20: score <= 20 (20 or less)

Verification of Validation Behavior

# Valid values
valid = Score(score=15)  # OK

# Values out of range are rejected
invalid_low = Score(score=-1)   # ValidationError
invalid_high = Score(score=21)  # ValidationError

# Boundary values are allowed
edge_low = Score(score=0)   # OK
edge_high = Score(score=20)  # OK

json_schema_extra (Deprecated)

class Score(BaseModel):
  score: int = Field(json_schema_extra={"minimum": 0, "maximum": 20})

While the result is the same if you are only outputting the JSON schema, validation (numeric range checking) will not function because these keywords are not officially recognized by Pydantic.

Summary

• Use ge and le for integer range constraints.
• Use json_schema_extra only when adding custom metadata that cannot be represented by standard constraint parameters.

References

https://docs.pydantic.dev/latest/concepts/fields/

https://docs.pydantic.dev/latest/migration/