↑
Take r←x↑y
x
must be a scalar or vector of integers. r
is the result of taking from each axis of y
as many elements as specified by the corresponding element of x
. If x
is positive, the first x
elements are taken; if x
is negative, the last x
elements are taken; if x
is zero, no element is taken. Supports tbe backward
extra argument, which negates x
.
Overtake
If the fill
extra argument is provided, axes where the corresponding element of x
is (in absolute value) larger than the axis length are filled with the fill provided.
Outfix
If (an element of) x
is nested, it must be a two-element vector, with the first item a nonnegative integer and the second item a nonpositive integer. In that case, the items taken on the corresponding axis are the first ⊃x
and the last -⊇x
, except that if they overlap they're taken only once.