How to replace empty lists in polars?

Question

Given a polars dataframe with empty lists in column values. How can I replace them with pl.Null or None in order to count them as missing values?

Input:

df = pl.DataFrame([
    pl.Series('contacts', [[], [], ['1081'], ['1313'], ['3657']], dtype=pl.List(pl.String)),
    pl.Series('line_items', [[], [], [], [], []], dtype=pl.List(pl.String)),
])

I have tried:

df.with_columns(
   pl.when(pl.col(pl.List(pl.Null)))
     .then(None)
     .otherwise(pl.col(pl.List))
     .name.keep()
)

But the [] are still present in the output.

Instead, I want this:

shape: (5, 2)
┌───────────┬────────────┐
│ contacts  ┆ line_items │
│ ---       ┆ ---        │
│ list[str] ┆ list[str]  │
╞═══════════╪════════════╡
│ null      ┆ null       │
│ null      ┆ null       │
│ ["1081"]  ┆ null       │
│ ["1313"]  ┆ null       │
│ ["3657"]  ┆ null       │
└───────────┴────────────┘

Answer 1

As @Dean MacGregor mentioned, you can do this with .list.len()==0

Here is the code for it:

# transformation for 1 column

df.with_columns(
    pl.when(pl.col('contacts').list.len() == 0)
      .then(None)
      .otherwise(pl.col('contacts'))
      .name.keep()
)

shape: (5, 2)
┌───────────┬────────────┐
│ contacts  ┆ line_items │
│ ---       ┆ ---        │
│ list[str] ┆ list[str]  │
╞═══════════╪════════════╡
│ null      ┆ []         │
│ null      ┆ []         │
│ ["1081"]  ┆ []         │
│ ["1313"]  ┆ []         │
│ ["3657"]  ┆ []         │
└───────────┴────────────┘

# transformation for all columns of datatype List(Str)

df.with_columns(
    pl.when(pl.col(pl.List(pl.String)).list.len() == 0)
      .then(None)
      .otherwise(pl.col(pl.List(pl.String)))
      .name.keep() 
)

shape: (5, 2)
┌───────────┬────────────┐
│ contacts  ┆ line_items │
│ ---       ┆ ---        │
│ list[str] ┆ list[str]  │
╞═══════════╪════════════╡
│ null      ┆ null       │
│ null      ┆ null       │
│ ["1081"]  ┆ null       │
│ ["1313"]  ┆ null       │
│ ["3657"]  ┆ null       │
└───────────┴────────────┘