Pydantic refuse to validate my schema

[goldyfruit]

Description

Hi,

I'm pretty new with Pydantic validation but it makes me think a lot to Golang structures.
I'm trying to POST a pretty simple JSON payload but Pydantic keeps refusing the validation when using curl.

{
  "target": "https://google.fr",
  "link": "BHDH",
  "extras": {
    "code": "ABCDE"
  }
}

Here is my schemas.py.

from typing import Optional, Any, Dict
from pydantic import BaseModel, HttpUrl, Json

class Link(BaseModel):
    link: str
    target: httpUrl
    extras: Json[Dict[str, Any]]

print(Link(link='BHDH', target='https://google.com, extras='{"code": "ABCDE"}'))

Here is the curl command used to POST the JSON payload.

$ curl -X POST http://localhost:8000 -d '{"target": "https://google.fr", "link": "BHDH", "extras": {"code": "ABCDE"}}'

And this is the error I got:

{"detail":[{"loc":["body","extras"],"msg":"JSON object must be str, bytes or bytearray","type":"type_error.json"}]}

Thanks for your help :)

Environment

OS: Linux
FastAPI Version: 0.63.0
Python-version: 3.9.2

[Kludex]

This will work for you:

curl -X POST "http://127.0.0.1:8000/" -H  "accept: application/json" -H  "Content-Type: application/json" -d "{\"link\":\"BHDH\",\"target\":\"https://google.fr\",\"extras\":\"{\\\"code\\\": \\\"ABCDE\\\"}\"}"

Code to test:

from fastapi import FastAPI
from typing import Optional, Any, Dict
from pydantic import BaseModel, HttpUrl, Json

class Link(BaseModel):
    link: str
    target: HttpUrl
    extras: Json[Dict[str, Any]]

app = FastAPI()

@app.post("/")
async def home(link: Link):
    return link

Swagger input:

{
  "link": "BHDH",
  "target": "https://google.fr",
  "extras": "{\"code\": \"ABCDE\"}"
}

But... I do believe that you don't want to do this 😅
That whole body is a JSON, you probably want to use typing.Dict instead of pydantic.Json.

[goldyfruit]

@Kludex ahah indeed, I prefer to avoid the \ !!
typing.Dict did the trick, thanks.

[tiangolo]

Thanks for the help here @Kludex ! 👏 🙇

Thanks for reporting back and closing the issue @goldyfruit 👍

Sorry for the long delay! 🙈 I wanted to personally address each issue/PR and they piled up through time, but now I'm checking each one in order.