read_csv with s3 glob pattern fails with TypeError: Object does not have a .read() method.

[laurentS]

Checks

• I have checked that this issue has not already been reported.
• I have confirmed this bug exists on the latest version of Polars.

Reproducible example

storage_options = {...}

df = pl.read_csv(
    [f"s3://{prefix}/my_data/**"],
    storage_options=storage_options,
)

This fails with TypeError: Object does not have a .read() method.

Log output

Traceback (most recent call last):
  File "./main.py", line 51, in <module>
    load_files()
  File "./main.py", line 30, in load_files
    df = pl.read_csv(
         ^^^^^^^^^^^^
  File ".venv/lib/python3.12/site-packages/polars/_utils/deprecation.py", line 92, in wrapper
    return function(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File ".venv/lib/python3.12/site-packages/polars/_utils/deprecation.py", line 92, in wrapper
    return function(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File ".venv/lib/python3.12/site-packages/polars/_utils/deprecation.py", line 92, in wrapper
    return function(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File ".venv/lib/python3.12/site-packages/polars/io/csv/functions.py", line 534, in read_csv
    df = _read_csv_impl(
         ^^^^^^^^^^^^^^^
  File ".venv/lib/python3.12/site-packages/polars/io/csv/functions.py", line 682, in _read_csv_impl
    pydf = PyDataFrame.read_csv(

Issue description

I am aware that read_csv does not officially accept a list argument, but I gave it a shot as it is allowed by fsspec and it goes down to prepare_file_arg which returns an OpenFiles from fsspec. It then ends up in _read_csv_impl where it fails, as it cannot handle a list.

On the other hand, I can write pl.read_csv('/some/local/path/**') and it reads all the files in that folder.

Expected behavior

I would expect the remote s3 read to work the same way, especially as the underlying fsspec and s3fs allow it.
Whether it works through a glob like s3://some/path/** or a list like above doesn't really matter, though I think the latter is easier, as fsspec knows how to deal with it.

Installed versions

--------Version info---------
Polars:              1.19.0
Index type:          UInt32
Platform:            Linux-6.12.8-amd64-x86_64-with-glibc2.40
Python:              3.12.8 (main, Jan 11 2025, 09:42:09) [GCC 14.2.0]
LTS CPU:             False

----Optional dependencies----
adbc_driver_manager  <not installed>
altair               <not installed>
azure.identity       <not installed>
boto3                1.36.1
cloudpickle          <not installed>
connectorx           <not installed>
deltalake            <not installed>
fastexcel            <not installed>
fsspec               2024.12.0
gevent               <not installed>
google.auth          <not installed>
great_tables         <not installed>
matplotlib           <not installed>
nest_asyncio         <not installed>
numpy                <not installed>
openpyxl             <not installed>
pandas               <not installed>
pyarrow              <not installed>
pydantic             <not installed>
pyiceberg            <not installed>
sqlalchemy           2.0.37
torch                <not installed>
xlsx2csv             <not installed>
xlsxwriter           <not installed>