Pydantic 2 Regression - TypeError: unhashable type on json_schema_extra examples

[angely-dev]

First Check

• I added a very descriptive title here.
• I used the GitHub search to find a similar question 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 fastapi import FastAPI
from pydantic import BaseModel
from typing import List


class Toy(BaseModel):
    name: str


class Cat(BaseModel):
    toys: List[Toy]

    model_config = {
        "json_schema_extra": {
            "examples": [
                {
                    "toys": [Toy(name="a"), Toy(name="b")]
                }
            ]
        }
    }


app = FastAPI()


@app.post("/cats")
def create_cat(cat: Cat):
    print("cat was created")

Description

I am aware of the discussion #10097. I did comment on it but since it is closed, I guess nobody gets notified. It seems the issue (#10321) has been fixed for Field examples but not for json_schema_extra examples.

Reproduce with Pydantic v2

$ vim api.py  # put Example Code in this file

$ python3 -m venv venv
$ source venv/bin/activate

(venv) $ pip install fastapi uvicorn
(venv) $ uvicorn api:app --reload
INFO:     Will watch for changes in these directories: ['/tmp/sample']
INFO:     Uvicorn running on http://127.0.0.1:8000 (Press CTRL+C to quit)
INFO:     Started reloader process [23192] using StatReload
INFO:     Started server process [23194]
INFO:     Waiting for application startup.
INFO:     Application startup complete.

Then go to: http://localhost:8000/docs

INFO:     127.0.0.1:60640 - "GET /docs HTTP/1.1" 200 OK
INFO:     127.0.0.1:60640 - "GET /openapi.json HTTP/1.1" 500 Internal Server Error
ERROR:    Exception in ASGI application
Traceback (most recent call last):
  File "/tmp/sample/venv/lib/python3.10/site-packages/uvicorn/protocols/http/h11_impl.py", line 407, in run_asgi
    result = await app(  # type: ignore[func-returns-value]
  File "/tmp/sample/venv/lib/python3.10/site-packages/uvicorn/middleware/proxy_headers.py", line 69, in __call__
    return await self.app(scope, receive, send)
  File "/tmp/sample/venv/lib/python3.10/site-packages/fastapi/applications.py", line 1054, in __call__
    await super().__call__(scope, receive, send)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/applications.py", line 123, in __call__
    await self.middleware_stack(scope, receive, send)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/middleware/errors.py", line 186, in __call__
    raise exc
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/middleware/errors.py", line 164, in __call__
    await self.app(scope, receive, _send)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/middleware/exceptions.py", line 65, in __call__
    await wrap_app_handling_exceptions(self.app, conn)(scope, receive, send)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/_exception_handler.py", line 64, in wrapped_app
    raise exc
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    await app(scope, receive, sender)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/routing.py", line 756, in __call__
    await self.middleware_stack(scope, receive, send)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/routing.py", line 776, in app
    await route.handle(scope, receive, send)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/routing.py", line 297, in handle
    await self.app(scope, receive, send)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/routing.py", line 77, in app
    await wrap_app_handling_exceptions(app, request)(scope, receive, send)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/_exception_handler.py", line 64, in wrapped_app
    raise exc
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/_exception_handler.py", line 53, in wrapped_app
    await app(scope, receive, sender)
  File "/tmp/sample/venv/lib/python3.10/site-packages/starlette/routing.py", line 72, in app
    response = await func(request)
  File "/tmp/sample/venv/lib/python3.10/site-packages/fastapi/applications.py", line 1009, in openapi
    return JSONResponse(self.openapi())
  File "/tmp/sample/venv/lib/python3.10/site-packages/fastapi/applications.py", line 981, in openapi
    self.openapi_schema = get_openapi(
  File "/tmp/sample/venv/lib/python3.10/site-packages/fastapi/openapi/utils.py", line 475, in get_openapi
    field_mapping, definitions = get_definitions(
  File "/tmp/sample/venv/lib/python3.10/site-packages/fastapi/_compat.py", line 229, in get_definitions
    field_mapping, definitions = schema_generator.generate_definitions(
  File "/tmp/sample/venv/lib/python3.10/site-packages/pydantic/json_schema.py", line 378, in generate_definitions
    definitions_remapping = self._build_definitions_remapping()
  File "/tmp/sample/venv/lib/python3.10/site-packages/pydantic/json_schema.py", line 2188, in _build_definitions_remapping
    return _DefinitionsRemapping.from_prioritized_choices(
  File "/tmp/sample/venv/lib/python3.10/site-packages/pydantic/json_schema.py", line 181, in from_prioritized_choices
    schemas_for_alternatives[defs_ref] = _deduplicate_schemas(schemas_for_alternatives[defs_ref])
  File "/tmp/sample/venv/lib/python3.10/site-packages/pydantic/json_schema.py", line 2299, in _deduplicate_schemas
    return list({_make_json_hashable(schema): schema for schema in schemas}.values())
  File "/tmp/sample/venv/lib/python3.10/site-packages/pydantic/json_schema.py", line 2299, in <dictcomp>
    return list({_make_json_hashable(schema): schema for schema in schemas}.values())
TypeError: unhashable type: 'Toy'

Same working example with Pydantic v1

# api.py

from fastapi import FastAPI
from pydantic import BaseModel
from typing import List


class Toy(BaseModel):
    name: str


class Cat(BaseModel):
    toys: List[Toy]

    class Config:
        schema_extra = {
            "examples": [
                {
                    "toys": [Toy(name="a"), Toy(name="b")]
                }
            ]
        }


app = FastAPI()


@app.post("/cats")
def create_cat(cat: Cat):
    print("cat was created")

Downgrade to Pydantic v1:

(venv) $ pip uninstall pydantic
(venv) $ pip install pydantic==v1.10.15
(venv) $ python -c "import pydantic; print(pydantic.version.VERSION)"
1.10.15

(venv) $ uvicorn api:app --reload
INFO:     Will watch for changes in these directories: ['/tmp/sample']
INFO:     Uvicorn running on http://127.0.0.1:8000 (Press CTRL+C to quit)
INFO:     Started reloader process [23707] using StatReload
INFO:     Started server process [23709]
INFO:     Waiting for application startup.
INFO:     Application startup complete.

Then go to: http://localhost:8000/docs

INFO:     127.0.0.1:48500 - "GET /docs HTTP/1.1" 200 OK
INFO:     127.0.0.1:48500 - "GET /openapi.json HTTP/1.1" 200 OK

Workaround with Pydantic v2

I have to use model_dump on each model instances:

class Cat(BaseModel):
    toys: List[Toy]

    model_config = {
        "json_schema_extra": {
            "examples": [
                {
                    "toys": [Toy(name="a").model_dump(), Toy(name="b").model_dump()]  # here
                }
            ]
        }
    }

Operating System

Linux

Operating System Details

No response

FastAPI Version

0.110.2

Pydantic Version

2.7.1

Python Version

3.10.12

Additional Context

Edit: after more research, it appears to be an issue of Pydantic as per pydantic/pydantic#7625. @dmontagu did mention the fix doesn't affect json_extra_schema examples:

Note this doesn't affect the result of a mutating json_schema_extra. To make that work would be a bit more complex (…)
pydantic/pydantic#7625 (comment)

[angely-dev]

Any thoughts on this regression?

[YuriiMotov]

As you already mentioned this comes from Pydantic. Looking at their discussions I see that they are not going to solve this.
So, the only solution is to use model_dump.