How to use UploadFile's file attribute directly?

[marcmk6]

Doc mentions that file of a UploadFile is a file-like that we "can pass directly to other functions or libraries that expect a "file-like" object".
I'm wondering, how can I use it without using the async read method to explicitly awaiting for the file?
Directly using the file attribute gives me error (presumably the file is not ready yet). I have to call the async read.

The doc indicates this file attribute is a SpooledTemporaryFile which saves RAM. This is important if the file if very large.
The critical issue of explicitly using the async read is it requires the entire content of the file to reside in RAM.

[11kkw]

UploadFile.file is actually a tempfile.SpooledTemporaryFile instance. This means it can be used directly as a standard file-like object, as the documentation states. You do not need to call await file.read() in order to pass it to other libraries.

One detail to be aware of is the file pointer position: if the file has already been read once, the pointer will be at the end of the file. In that case you need to call file.file.seek(0) before passing it to another library.

from starlette.applications import Starlette
from starlette.responses import JSONResponse
from starlette.routing import Route
from starlette.datastructures import UploadFile
import pandas as pd


async def upload_endpoint(request):
    form = await request.form()
    file: UploadFile = form["file"]

    file.file.seek(0)
    try:
        df = pd.read_csv(file.file)
        rows, cols = df.shape
        head_preview = df.head().to_dict(orient="records")
    except Exception as e:
        rows, cols, head_preview = 0, 0, str(e)

    return JSONResponse({
        "rows": rows,
        "cols": cols,
        "head": head_preview,
        "file_type": str(type(file.file)),
    })


app = Starlette(
    debug=True,
    routes=[
        Route("/upload", upload_endpoint, methods=["POST"]),
    ],
)

Does this look right?