; docformat = 'rst'
;+
; Truncate argument towards 0.0, i.e., takes the `FLOOR` of positive values
; and the `CEIL` of negative values.
;
; :Examples:
; Try the main-level program at the end of this file. It does::
;
; IDL> print, mg_trunc([1.2, -1.2, 0.0])
; 1 -1 0
; IDL> print, floor([1.2, -1.2, 0.0])
; 1 -2 0
; IDL> print, ceil([1.2, -1.2, 0.0])
; 2 -1 0
;
; :Returns:
; array of same type as argument
;
; :Params:
; x : in, required, type=float/double
; array containing values to truncate
;-
function mg_trunc, x
compile_opt strictarr
result = ceil(x)
posInd = where(x gt 0, nposInd)
if (nposInd gt 0L) then begin
result[posInd] = floor(x[posInd])
endif
return, result
end
; main-level example program
print, mg_trunc([1.2, -1.2, 0.0])
end