[Python] How to add numeric value from one column to other List column elements in Polars?

Suppose I have the following Polars DataFame:

import polars as pl

df = pl.DataFrame({
'lst': [[0, 1], [9, 8]],
'val': [3, 4]
})


And I want to add the number in the val column, to every element in the corresponding list in the lst column, to get the following result:

┌───────────┬─────┐
│ lst ┆ val │
│ --- ┆ --- │
│ list[i64] ┆ i64 │
╞═══════════╪═════╡
│ [3, 4] ┆ 3 │
│ [13, 12] ┆ 4 │
└───────────┴─────┘


I know how to add a constant value, e.g.:

new_df = df.with_columns(
pl.col('lst').list.eval(pl.element() + 2)
)


But when I try:

new_df = df.with_columns(
pl.col('lst').list.eval(pl.element() + pl.col('val'))
)


I get the following error:

polars.exceptions.ComputeError: named columns are not allowed in `list.eval`; consider using `element` or `col("")`


Is there any elegant way to achieve my goal (without map_elements)?

Thanks in advance.

Continue reading...