[Python] Row-wise updates in polars

Say I have this dataframe:

>>> pl.DataFrame([[1,2,3],[4,5,6],[7,8,9]],list('abc'))
shape: (3, 3)
┌─────┬─────┬─────┐
│ a ┆ b ┆ c │
│ --- ┆ --- ┆ --- │
│ i64 ┆ i64 ┆ i64 │
╞═════╪═════╪═════╡
│ 1 ┆ 4 ┆ 7 │
│ 2 ┆ 5 ┆ 8 │
│ 3 ┆ 6 ┆ 9 │
└─────┴─────┴─────┘


Note that below I'm specifically asking about the case where columns are "unimporant" - i.e. I'm looking for a solution that doesn't necessarily depend on columns being named a, b and c or that there is a certain number of columns.

I can use the following to update a certain column at a certain row:

>>> pl.DataFrame([[1,2,3],[4,5,6],[7,8,9]],list('abc')).with_row_index().with_columns(a=pl.when(pl.col('index') == 1).then(42).otherwise(pl.col('a'))).drop('index')
shape: (3, 3)
┌─────┬─────┬─────┐
│ a ┆ b ┆ c │
│ --- ┆ --- ┆ --- │
│ i64 ┆ i64 ┆ i64 │
╞═════╪═════╪═════╡
│ 1 ┆ 4 ┆ 7 │
│ 42 ┆ 5 ┆ 8 │
│ 3 ┆ 6 ┆ 9 │
└─────┴─────┴─────┘


but that's a bit hard if I don't know that I should be updating a.

How can I do the following:

1. Replace the whole row - e.g. I want to replace the 2nd row with 43,42,41:

┌─────┬─────┬─────┐
│ a ┆ b ┆ c │
│ --- ┆ --- ┆ --- │
│ i64 ┆ i64 ┆ i64 │
╞═════╪═════╪═════╡
│ 1 ┆ 4 ┆ 7 │
│ 43 ┆ 42 ┆ 41 │
│ 3 ┆ 6 ┆ 9 │
└─────┴─────┴─────┘

1. Replace any column in a certain row by condition - e.g. I want to negate any values > 4 in 2nd row:

┌─────┬─────┬─────┐
│ a ┆ b ┆ c │
│ --- ┆ --- ┆ --- │
│ i64 ┆ i64 ┆ i64 │
╞═════╪═════╪═════╡
│ 1 ┆ 4 ┆ 7 │
│ 2 ┆ -5 ┆ -8 │
│ 3 ┆ 6 ┆ 9 │
└─────┴─────┴─────┘

Continue reading...