validation_error_definition schema is incorrect when detecting an error in an array

First Check

• I added a very descriptive title to this issue.
• I used the GitHub search to find a similar issue and didn’t find it.
• I searched the FastAPI documentation, with the integrated search.
• I already searched in Google “How to X in FastAPI” and didn’t find any information.
• I already read and followed all the tutorial in the docs and didn’t find an answer.
• I already checked if it is not related to FastAPI but to Pydantic.
• I already checked if it is not related to FastAPI but to Swagger UI.
• I already checked if it is not related to FastAPI but to ReDoc.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from typing import Optional, List

from fastapi import FastAPI
from pydantic import BaseModel

class SubItem(BaseModel):
    price: float

class Item(BaseModel):
    price_list: Optional[List[SubItem]] = None


app = FastAPI()


@app.post("/items/")
async def create_item(item: Item):
    return item

Description

Run the sample program and call the /items/ api with this body:

{
  "price_list": [
    {
      "price": "zero"
    }
  ]
}

You will get a 422 like this

{
  "detail": [
    {
      "loc": [
        "body",
        "price_list",
        0,
        "price"
      ],
      "msg": "value is not a valid float",
      "type": "type_error.float"
    }
  ]
}

But this 422 response does not follow the validation_error_definition schema :

validation_error_definition = {
    "title": "ValidationError",
    "type": "object",
    "properties": {
        "loc": {"title": "Location", "type": "array", "items": {"type": "string"}},
        "msg": {"title": "Message", "type": "string"},
        "type": {"title": "Error Type", "type": "string"},
    },
    "required": ["loc", "msg", "type"],
}

So the 422 response from fastapi does not respect the schema of the response…

The issue comes from the error in the array, pydantic returns 0 as index of the error, but it is not a string, it is an integer. And it makes sense to return an integer, otherwise we wouldn’t know if it is a key in a dict or an index.

As a developer I can workaround this issue by redefining validation_error_definition in my own code

from fastapi.openapi.utils import validation_error_definition

validation_error_definition["properties"] = {
    "loc": {
        "title": "Location", "type": "array", "items": {
            "oneOf": [
                {"type": "string"},
                {"type": "integer"}
            ]
        }
    },
    "msg": {"title": "Message", "type": "string"},
    "type": {"title": "Error Type", "type": "string"},
}

But it would be better to do the change directly in the code. What is your opinion ?

Operating System

Windows

Operating System Details

No response

FastAPI Version

0.68.1

Python Version

3.7.9

Additional Context

No response

[dconatha]

I just stumbled across this error by using https://schemathesis.readthedocs.io/en/stable/ and I agree this should be fixed.

The schema should match the schema of Pydantic’s “loc” found here:

https://github.com/samuelcolvin/pydantic/blob/5ccbdcb5904f35834300b01432a665c75dc02296/pydantic/error_wrappers.py#L12

[AbhijithGanesh]

Thanks for the clarification