Unexpected behavior in clip

[marsupialtail]

Polars version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of Polars.

Issue description

Clip doesn't seem to work across data types.

Reproducible example

a = polars.from_dict({"a":[1,1,2,2], "b":['{"my_field": "quack"}','{"my_field": "quack"}','{"my_field": "quack"}','{"my_field": "quack"}']})
a.with_columns([polars.col("a").clip(1.5, 1.6)])

This will return 1,1,1,1 for the a column.

Expected behavior

The a column should maybe be 1.5, 1.5, 1.6, 1.6?

Installed versions

0.16.15

[ritchie46]

Yeah.. should the argument type cast the whole column? What if the column is a float and users pass integers as clip args? I am not sure about this one.

[marsupialtail]

I think maybe you can raise an error or warning and say the types are incongruent when this happens

[stinodego]

Closing in favor of #11385