`group-by` for filters

Splits a list or table into groups, and returns a record containing those groups.

Signature

> group-by {flags} (grouper)

Flags

--to-table, -: Return a table with "groups" and "items" columns

Parameters

grouper: The path to the column to group on.

Input/output types:

input	output
list<any>	any

Examples

Group items by the "type" column's values

> ls | group-by type

Group items by the "foo" column's values, ignoring records without a "foo" column

> open cool.json | group-by foo?

Group using a block which is evaluated against each input value

> [foo.txt bar.csv baz.txt] | group-by { path parse | get extension }
╭─────┬─────────────────╮
│     │ ╭───┬─────────╮ │
│ txt │ │ 0 │ foo.txt │ │
│     │ │ 1 │ baz.txt │ │
│     │ ╰───┴─────────╯ │
│     │ ╭───┬─────────╮ │
│ csv │ │ 0 │ bar.csv │ │
│     │ ╰───┴─────────╯ │
╰─────┴─────────────────╯

You can also group by raw values by leaving out the argument

> ['1' '3' '1' '3' '2' '1' '1'] | group-by
╭───┬───────────╮
│   │ ╭───┬───╮ │
│ 1 │ │ 0 │ 1 │ │
│   │ │ 1 │ 1 │ │
│   │ │ 2 │ 1 │ │
│   │ │ 3 │ 1 │ │
│   │ ╰───┴───╯ │
│   │ ╭───┬───╮ │
│ 3 │ │ 0 │ 3 │ │
│   │ │ 1 │ 3 │ │
│   │ ╰───┴───╯ │
│   │ ╭───┬───╮ │
│ 2 │ │ 0 │ 2 │ │
│   │ ╰───┴───╯ │
╰───┴───────────╯

You can also output a table instead of a record