[Python] Parsing Pydantic dict params

I have an endpoint that takes a Pydantic model, Foo, as a query parameter.

from typing import Annotated

import uvicorn
from fastapi import FastAPI, Query
from pydantic import BaseModel

app = FastAPI()


class Foo(BaseModel):
bar: str
baz: dict[str, str]


@app.get("/")
def root(foo: Annotated[Foo, Query()]):
return foo


if __name__ == "__main__":
uvicorn.run("test:app")


I'm defining my query params using Swagger, so the encoding should be correct. I know the baz param syntax looks redundant because I've nested a dictionary, but parsing fails even without nesting.



But when I call the endpoint...

curl -X 'GET' \
'http://127.0.0.1:8000/?bar=sda&baz=%7B%22abc%22%3A%22def%22%7D' \
-H 'accept: application/json'


FastAPI does not seem to read in Foo.baz correctly, returning

{
"detail": [
{
"type": "dict_type",
"loc": [
"query",
"baz"
],
"msg": "Input should be a valid dictionary",
"input": "{\"abc\":\"def\"}"
}
]
}


I've read similar questions and I know I can ingest the dictionary by accessing dict(request.query_params), but this bypasses FastAPI's validation and I'd prefer to keep the endpoint simple and consistent with the rest of my codebase by keeping the param as a Pydantic model.

How can I get FastAPI to parse Foo as a query param?

Continue reading...