dropDims: Drop singleton dimensions of an array
In tdeenes/eegR: Analyze EEG Signals

dropDims drops singleton dimensions (whose lengths is 1) of a multidimensional array. Compared to drop and adrop this function gives more control over which singleton dimensions to drop.

dropDims(
  x,
  drop = TRUE,
  keep = NULL,
  return_array = TRUE,
  named_vector = TRUE,
  stop_if_missing = TRUE
)

`x`	an array (or a matrix, or a list with `dim` attribute)
`drop`	1) either a single logical whether all singleton dimensions should be dropped (default: TRUE); or 2) a logical, integer or character vector indicating which dimensions to drop (if character, `x` must have named dimensions). If `drop` is a vector (case 2), the referred dimensions must be present in x.
`keep`	an integer or character vector indicating those dimensions which must remain in the returned value even if they are singletons (if character, `x` must have named dimensions). Note that `keep` has a higher priority than `drop`. Also note that to-be-kept dimensions which are actually not present in `x` are simply ignored (instead of resulting in error).
`return_array`	logical value whether `dropDims` should return a one-dimensional array instead of a vector even if all or all but one dimension of `x` is dropped (default: TRUE)
`named_vector`	logical value whether a vector result should be named (TRUE, the default). Ignored if `return_array` is TRUE.
`stop_if_missing`	logical value whether dropping or keeping non-existent dimensions should result in error (TRUE, default), or should be ignored (FALSE)

# create example data
x <- array(1:4, c(2, 1, 2, 1),
           dimnames = list(dimA = letters[1:2],
                           dimB = "a",
                           dimC = LETTERS[1:2],
                           dimD = "z"))

# drop all singleton dimensions
x0 <- dropDims(x)
stopifnot(identical(dim(x0), c(2L, 2L)))

# drop all singleton dimensions but always keep dimD
x1 <- dropDims(x, keep = "dimD")
stopifnot(identical(dim(x1), c(2L, 2L, 1L)))

# create a new example; a list with dim attribute
( x <- array(list(1:2, letters[1:2]), c(2, 1, 1),
             dimnames = list(type = c("numeric", "character"),
                             single1 = "a",
                             single2 = "b")) )

# drop the single1 dimension
x0 <- dropDims(x, "single1")
stopifnot(identical(dim(x0), c(2L, 1L)))

# wrong dimension in drop: by default, it results in an error
# (note that x has only 3 dimensions, not 4)
x1 <- try(dropDims(x, drop = 1:4), silent = TRUE)
stopifnot(inherits(x1, "try-error"))

# however, you can also ask for skipping those missing dimensions in 'drop'
# and 'keep'
x2 <- dropDims(x, drop = 1:4, stop_if_missing = FALSE)
stopifnot(identical(dim(x2), 2L))

# by default, dropDims returns an array, even if it has only one dimension
( x <- array(1:3, c(3, 1), list(dimA = letters[1:3], dimB = "A")) )
( x0 <- dropDims(x) )
stopifnot(is.array(x0))

# you can change this behaviour
( x1 <- dropDims(x, return_array = FALSE) )
stopifnot(!is.array(x1))

# vector results are named by default...
stopifnot(identical(names(x1), letters[1:3]))

# ...but not necessarily
( x2 <- dropDims(x, return_array = FALSE, named_vector = FALSE) )
stopifnot(is.null(names(x2)))
stopifnot(identical(unname(x1), x2))