.pivot should offer a column prefix option.

[bfeif]

Description

For readability of the results of pivot, users should be able to add a prefix to the columns that are produced. Some similar functionality appears to be included in the separator argument, but this is only used for repeating values, and doesn't allow for prefixing with something other than the columns argument itself.

Desired behavior:

import polars as pl

orders_df = pl.DataFrame({
    "customer": ["dan", "stan", "dan", "stan", "stan"],
    "basket_size": [1, 2, 3, 2, 1]
})

pivot = (
    orders_df
    .pivot(
        columns="basket_size",
        index="customer",
        values="customer",
        aggregate_function="len",
        column_prefix="num_orders_basket_size_eq_"
    )
)

print(pivot)

shape: (2, 4)
┌──────────┬───────────────────────────────┬────────────────────────────────┬────────────────────────────────┐
│ customer ┆ num_orders_basket_size_eq_1   ┆ num_orders_basket_size_eq_2    ┆ num_orders_basket_size_eq_3    │
│ ---      ┆ ---                           ┆ ---                            ┆ ---                            │
│ str      ┆ u32                           ┆ u32                            ┆ u32                            │
╞══════════╪═══════════════════════════════╪════════════════════════════════╪════════════════════════════════╡
│ dan      ┆ 1                             ┆ null                           ┆ 1                              │
│ stan     ┆ 1                             ┆ 2                              ┆ null                           │
└──────────┴───────────────────────────────┴────────────────────────────────┴────────────────────────────────┘

Possibly Related Issues

#14491
#15775