Nullable and Checks for Polars List of Structs?

[fredlarochelle]

I am looking for a way to set nullable=True and add checks to values in a nested structure with Polars. Something similar to this:

schema = pa.DataFrameSchema(
    {
        "value_1": pa.Column(int, checks=pa.Check.greater_than_or_equal_to(0), nullable=True),
        "value_2" pas.Column(pl.List(pl.Struct({
            "value_3": pl.Utf8(),  # Possible to set to nullable=True and do checks?
            "value_4": pl.Float64(),  # Idem
        })), nullable=True),
    },
    strict=True,
    coerce=True,
)

Is it possible? Or is there a different approach to use?

Thanks.

[cosmicBboy]

hi @fredlarochelle I'm not a polars expert, but as far as I know I don't think this is possible. What happens in polars if that struct value contains nulls in value_3? If you can provide a small code snippet for polars code I can help iterate on a possible solution