#include "tensorstore/index_space/dim_expression.h"
template <typename TargetDimensions>
auto tensorstore::DimExpression<Op...>::Transpose(
    const TargetDimensions& target_dimensions) const;

Transposes the input dimensions such that the selected dimensions have the specified indices. Dimensions not in the selection retain their relative order and fill in the dimension indices not in target_dimensions.

The new dimension selection is equal to target_dimensions after normalization.

For example, Dims(2, 0).Transpose({1, 2}) has the following effects:

Before

After

Dimension selection

{2, 0}

{1, 2}

Input domain

[1*, 3], [2, 5*], [3, 4]

[2, 5*], [3, 4], [1*, 3]

Labels

{"x", "y", "z"}

{"y", "z", "x"}

Equivalent input indices

{2, 3, 4}

{3, 4, 2}

Equivalent input indices

{x, y, z}

{y, z, x}

where x is any index in [1, 3], y is any index in [2, 5], and z is any index in [3, 4].

Requires:

TargetDimensions is span-compatible with a span::value_type of DimensionIndex and a static extent compatible with the static rank of the dimension selection.

Parameters:
const TargetDimensions &target_dimensions

The new dimension indices corresponding to each selected dimension. May be a braced list, e.g. Transpose({1, 2}). A negative value of -n is equivalent to input_rank - n, where input_rank is the input rank of the transform to which this operation is applied.

Error absl::StatusCode::kInvalidArgument:

if the rank of the dimension selection is not equal to the length of target_dimensions, or if the indices in target_dimensions are not unique or outside the valid range.