Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

bug: `predict()` sends request body as raw string, breaks with `pydantic>=2`

Open Beginner friendly
  • Is something wrong?
#241 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
2/5
Estimated time
1-3 hours
Newbie friendliness
62/100
Issue type
Bug
Clarity
Clearly specified
Activity status
Stale
Tech stack
fastapi, python
Domain
api

Research direction

Start in vetiver/server.py at predict() and inspect how the request body is sent to the /predict endpoint. Reproduce the failure with the provided FastAPI TestClient example under pydantic 2, then verify that predict() returns a DataFrame of predictions without a 422 response.

Written by the indexing model from the issue text.

Description

Describe the bug
The predict() function in vetiver/server.py sends DataFrame data using requests.post(endpoint, data=...), which transmits the JSON as a raw byte string without setting Content-Type: application/json.

On the server side, the /predict endpoint declares its input as List[Model], relying on FastAPI + Pydantic to parse the body. Pydantic v1 was lenient enough to coerce the raw bytes into the expected type, but pydantic v2 strictly rejects it with a 422:

"Input should be a valid list".

data= sends the body as raw bytes. It should use json= (or set the content type to application/json) so that FastAPI correctly parses the body before handing it to pydantic for validation.

pydantic: 2.12.5
vetiver: 0.2.6

To Reproduce

import vetiver
from fastapi.testclient import TestClient

X, y = vetiver.mock.get_mock_data()
model = vetiver.mock.get_mock_model().fit(X, y)
v = vetiver.VetiverModel(model=model, model_name="model", prototype_data=X)
app = vetiver.VetiverAPI(v).app

client = TestClient(app)
response = vetiver.predict(
    "http://testserver/predict", X, test_client=client
)

Traceback (most recent call last):
  File "vetiver/server.py", line 379, in predict
    response.raise_for_status()
  File "httpx/_models.py", line 829, in raise_for_status
    raise HTTPStatusError(message, request=request, response=self)
httpx.HTTPStatusError: Client error '422 Unprocessable Entity' for url 'http://testserver/predict'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "demo.py", line 11, in <module>
    response = vetiver.predict(
        "http://testserver/predict", X, test_client=client
    )
  File "vetiver/server.py", line 382, in predict
    raise TypeError(re.sub(r"\n", ": ", response.text))
TypeError: 1 validation error::   {'type': 'list_type', 'loc': ('body',), 'msg': 'Input should be a valid list', 'input':
b'[{"B":54,"C":48,"D":68},...]'}: :   File "vetiver/server.py", line 238, in custom_endpoint:     POST /predict

Expected behavior
vetiver.predict() should return a DataFrame of predictions. With pydantic<2 installed, it does

Additional context
Add any other context about the problem here.

Dominant language
Python
Stars
71
Forks
20
No merged PRs in 30d

Getting set up

This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from rstudio/vetiver-python

All issues in rstudio/vetiver-python

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.