pivot

DataFrame pivot(DataFrame df,  string[]  index,  string on,  string[]  values,
       PivotAgg agg =  PivotAgg.NONE)

Pivots the frame: the distinct values of on become output columns (first-appearance order), the distinct index tuples become output rows (first-appearance order), and the cells hold the values aggregated per (index tuple, on value) (ADR-2026-10-24-20).

A single value column names each output column by the rendered on-value (the CSV cell renderers: decimal integers, shortest- round-trip floats, true/false, strict temporals, null); several value columns name {value}_{on} and run value-outer, on-inner. Duplicate cells error under the default NONE (naming the key and the count); missing cells take the aggregate's empty-set value (SUM/COUNT 0, FIRST/LAST/MIN/MAX/MEAN null). Null keys group (the grouping semantics). The output shape - index tuples x value columns x on values - is discovered before any cell is written; O(I.V.D) growth is the caller's to judge.

Parameters

`df`	the frame to reshape
`index`	the row-key columns (empty: one aggregate row)
`on`	the column whose distinct values become columns
`values`	the cell columns (one or more)
`agg`	the duplicate-cell policy

Returns

the pivoted frame: the index columns, then one column per

value column and on value.

Throws

DataFrameException for absent columns, no value columns,

duplicate cells under NONE, or a generated name colliding with an index column (or another generated name).

Example:

auto df = DataFrame([
   Series("idx", ["a", "a", "b"]),
   Series("on", ["x", "y", "x"]),
   Series("v", [1L, 2, 3]),
]);
auto wide = pivot(df, ["idx"], "on", ["v"]);
assert(wide.columns() == ["idx", "x", "y"]);
assert(wide["x"].toArray!long() == [1, 3]);
assert(wide["y"].toNullableArray!long()[1].isNull);