restrict_array: Restrict (or permute) an array.
In LAJordanger/leanRcoding: Leaner R Coding

Description Usage Arguments Details Value

This function will restrict an array based upon a specified subset of its dimension-names, without any need for the user to worry about how these dimension-names are positioned within the array. However, if the user desires that the resulting array should have the specified order on the dimension-names, then the argument .permute can be set to TRUE in order to achieve that result.

1 2	restrict_array(.arr, .restrict, .drop = FALSE, .never_drop = NULL, .keep_attributes = TRUE, .permute = FALSE)

`.arr`	The array we want to restrict. The code in its present incarnation requires an array with a full set of unique dimension-names, although I suppose it could be possible to extend to a case where the dimension-names are missing too.
`.restrict`	A named `list` that collects the arguments to be used in the restriction, i.e. the names should correspond to names of the dimension-names, and the values should specify a subset (or a permutation, if you are so inclined) of the dimension-names.
`.drop`	A logical argument, default value `FALSE`, that decides whether or not dimensions are allowed to drop.
`.never_drop`	A character vector to be used if `.drop` is `TRUE`, but some dimensions should be protected. The default value `NULL` does nothing, whereas the argument will be ignored if `.drop` is `FALSE`.
`.keep_attributes`	A logical argument, default value `TRUE`, that decides whether or not attributes should be kept after the restriction. Note that relevant `class` attributes are kept by default, and that `dim` and `dimnames` must be as specified by the restriction.
`.permute`	A logical argument, default value `FALSE`, that will decide whether or not the dimensions of the final array should be adjusted to one used by `.restrict`.

Arrays are awesome, but it can alas be agonising to keep track of the dimension-names. This function frees the user from such frustrations.

The result will depend on how .sub_arr and .arr is related. If the dimension-names of .sub_arr is contained in those of .arr, but its dimension is in fact smaller, then the result will be that .arr will be filled with copies of .sub_arr along the unspecified dimensions. If the dimensions are equal, then only those components in .arr that directly match the dimension-names of .sub_arr will be affected.