Sort this expression

Description

If used in a groupby context, values within each group are sorted.

Usage

<Expr>$sort(..., descending = FALSE, nulls_last = FALSE)

Arguments

…	These dots are for future extensions and must be empty.
descending	Sort in descending order.
nulls_last	Place null values last.

Value

A polars expression

Examples

library("polars")

df <- pl$DataFrame(a = c(6, 1, 0, NA, Inf, NaN))

df$with_columns(
  sorted = pl$col("a")$sort(),
  sorted_desc = pl$col("a")$sort(descending = TRUE),
  sorted_nulls_last = pl$col("a")$sort(nulls_last = TRUE)
)

#> shape: (6, 4)
#> ┌──────┬────────┬─────────────┬───────────────────┐
#> │ a    ┆ sorted ┆ sorted_desc ┆ sorted_nulls_last │
#> │ ---  ┆ ---    ┆ ---         ┆ ---               │
#> │ f64  ┆ f64    ┆ f64         ┆ f64               │
#> ╞══════╪════════╪═════════════╪═══════════════════╡
#> │ 6.0  ┆ null   ┆ null        ┆ 0.0               │
#> │ 1.0  ┆ 0.0    ┆ NaN         ┆ 1.0               │
#> │ 0.0  ┆ 1.0    ┆ inf         ┆ 6.0               │
#> │ null ┆ 6.0    ┆ 6.0         ┆ inf               │
#> │ inf  ┆ inf    ┆ 1.0         ┆ NaN               │
#> │ NaN  ┆ NaN    ┆ 0.0         ┆ null              │
#> └──────┴────────┴─────────────┴───────────────────┘

# When sorting in a group by context, values in each group are sorted.
df <- pl$DataFrame(
  group = c("one", "one", "one", "two", "two", "two"),
  value = c(1, 98, 2, 3, 99, 4)
)

df$group_by("group")$agg(pl$col("value")$sort())

#> shape: (2, 2)
#> ┌───────┬──────────────────┐
#> │ group ┆ value            │
#> │ ---   ┆ ---              │
#> │ str   ┆ list[f64]        │
#> ╞═══════╪══════════════════╡
#> │ two   ┆ [3.0, 4.0, 99.0] │
#> │ one   ┆ [1.0, 2.0, 98.0] │
#> └───────┴──────────────────┘