angely-dev / freeradius-api Goto Github PK

i changed the API url parameters to my host Ip and the code runs okay, but cant access on browser

Looking at the code (https://github.com/angely-dev/freeradius-api/blob/master/src/api.py#L9), the lack of support for updating entities is intentionally, and it's recommended to remove and re-create entities when an update is needed.

However, this is problematic for the use case where you want to update a group that already has users attached. I would like to update this group to add some more replies, but removing this isn't possible without removing the group (and links to the users). Also, since this isn't possible in one atomic operation, the users will be unlinked to a group for a brief moment.

I'm thinking on adding this functionality, but will be this be accepted in this project, or is this beyond the scope of this project?

Pydantic recently released the v2 major version which breaks or deprecates some features of the v1. A migration guide has been published:

• @root_validator is deprecated and should be replaced with @model_validator
• The following config settings have been renamed: schema_extra → json_schema_extra

Some endpoints currently fetch and return all results without pagination. This could lead to performance issues.

In the era of infinite scrolling, the keyset pagination mechanism is generally preferred over the offset pagination mechanism. Work has started in 460de53 in that sense.

• The REST API must now call the new find_usernames(), find_groupnames() and find_nasnames() methods in place of find_all_usernames(), find_all_groupnames() and find_all_nasnames() methods still there for legacy purposes.
• The REST API must also tell the client how to navigate through pages. It can be implemented with either JSON metadata or the HTTP Link header as per RFC 5988 (this is what GitHub does actually).

Some endpoints trigger HTTP errors (e.g., 404). Except for the 422 (internal RequestValidationError), HTTP errors are not yet specified per endpoint. FastAPI allows it through the additional-responses feature:

class Message(BaseModel):
    message: str

@app.get("/items/{item_id}", response_model=Item, responses={404: {"model": Message}})
async def read_item(item_id: str):
    if item_id == "foo":
        return {"id": "foo", "value": "there goes my hero"}
    return JSONResponse(status_code=404, content={"message": "Item not found"})

FastAPI allows for endpoints example data using Pydantic schema customization. This is handy for testing the API. Prepare example data for:

• User base model
• Group base model
• Nas base model

First of all, I don't want to force this project into a direction you don't want to. Feel free to reject my comments and remarks :)

This project has a good functionality, but it's lacking options to configure it without having to edit the files. My goal is to have an application (probably packaged in a docker image) that can be configured and started with all database drivers included.

I'm looking at https://fastapi.tiangolo.com/advanced/settings/ to parse environment variables (and probably also command line variables now that pydantic/pydantic-settings#214 has been merged, a few days ago!), so the hardcoded configurations like the db_connection parameters, or the RadTables overrides can be removed and filled in from configuration.

I've started working on this, but if you don't like this direction, feel free to say so, and I'll setup a fork of this project with these changes.