Binary Flip (Not) Operator

The binary flip (‘not’, R's !) operator, bitFlip(a, w), “flips every bit” of a up to the w-th bit.

bitFlip(a, bitWidth = 32)

Arguments

a

numeric vector.

bitWidth

scalar integer between 0 and 32.

Note

bitFlip(a, w) is an “involution”, i.e. it is its own inverse – when a is in \(\{0, 1, .., 2^{32}-1\}\). Notably, negative values a are equivalent to their values in the above range, see also bitUnique() in the ‘Examples’.

Value

(“binary”) numeric vector of the same length as a masked with (2^bitWidth)-1. NA is returned for any value of a that is not finite or whose magnitude is greater or equal to \(2^{32}\).

Author

Steve Dutky

See also

bitShiftL, bitXor, etc.

Examples

bitFlip(0:5)
#> [1] 4294967295 4294967294 4294967293 4294967292 4294967291 4294967290
##
bitUnique <- function(x) bitFlip(bitFlip(x)) # "identity" when x in  0:(2^32-1)
bitUnique(  0:16 ) # identical (well, double precision)
#>  [1]  0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16
bitUnique(-(1:16)) # 4294967295 ...
#>  [1] 4294967295 4294967294 4294967293 4294967292 4294967291 4294967290
#>  [7] 4294967289 4294967288 4294967287 4294967286 4294967285 4294967284
#> [13] 4294967283 4294967282 4294967281 4294967280
stopifnot(
  identical(bitUnique(-(1:16)), 2^32 -(1:16)),
  bitFlip(-1) == 0,
  bitFlip(0 ) == 2^32 - 1,
  bitFlip(0, bitWidth=8) == 255
)