numpydoc does not render parameters as expected

[ritchie46]

This docstring is rendered as:

    """
    Read into a DataFrame from a parquet file.

    Parameters
    ----------
    source
        Path to a file, list of files, or a file like object. If the path is a directory, that directory will be used
        as partition aware scan.
        If ``fsspec`` is installed, it will be used to open remote files
    use_pyarrow
            Use pyarrow instead of the rust native parquet reader. The pyarrow reader is more stable.
    stop_after_n_rows
        After n rows are read from the parquet, it stops reading.
        Only valid when 'use_pyarrow==False'
    memory_map
        Memory map underlying file. This will likely increase performance.
        Only used when 'use_pyarrow==True'
    columns
        Columns to project/ select.
        Only valid when 'use_pyarrow==True'
    storage_options
        Extra options that make sense for ``fsspec.open()`` or a particular storage connection, e.g. host, port, username, password, etc.
    **kwargs
        kwargs for [pyarrow.parquet.read_table](https://arrow.apache.org/docs/python/generated/pyarrow.parquet.read_table.html)

    Returns
    -------
    DataFrame
    """

Any ideas on what that could be?

[rossbar]

I can't replicate this - if I add a dummy function to an existing project and cp/paste the above docstring it renders fine. Do you have a link to the source?

[ritchie46]

I can't replicate this - if I add a dummy function to an existing project and cp/paste the above docstring it renders fine. Do you have a link to the source?

Thanks for taking a look, here is the source: https://github.com/pola-rs/polars/tree/master/py-polars/docs

[ritchie46]

I already found what messes it up. It's the commonmark extension: https://github.com/pola-rs/polars/blob/8a926f9cb15005debd0b6be230e0dc901acbe7ee/py-polars/docs/source/conf.py#L92

Have you guys any tip for markdown rendering and numpy style docs?