R/is_positive_integer.R

In chinese.misc: Miscellaneous Tools for Chinese Text Mining and More

#' A Convenient Version of is.integer
#'
#' This function checks if all elements of an object can be taken to be valid integers.
#'
#' The reasons to use \code{is_positive_integer} are: 
#'
#' \itemize{
#'   \item (1) We often check if an object is a vector of positive integer. 
#' But \code{is.numeric} cannot 
#' do this because it also returns \code{TRUE} for a numeric matrix. 
#'   \item (2) Sometimes \code{is.integer} 
#' returns a too strict result. For example, \code{is.integer(3.0)} returns \code{FALSE},  
#' but the number 3.0 is 
#' valid in codes such as \code{rep(10, 3.0)}, that is to say, as long as a number can be taken 
#' to be a valid integer, we take it to be a integer, even when \code{is.integer} 
#' returns \code{FALSE}. 
#'   \item (3) \code{is_positive_integer} returns \code{FALSE} for 
#' length = 0 object, even when it is \code{integer(0)}. To let the function return this result is 
#' because integer of length 0 is a invalid input for many functions. 
#'   \item (4) \code{is_positive_integer} 
#' returns \code{FALSE} for any object that contains \code{NA}, so that object that gets a 
#' \code{TRUE} from this function is more likely to be a valid value to be passed to 
#' other functions.
#' }
#' 
#' @param x an object to be checked
#' @param len numeric vector specifying the allowed length of the \code{x}. If the length 
#' of the checked object is not in \code{len}, the function will return \code{FALSE}, even 
#' when it is a positive integer vector. 
#' The default is \code{NULL}, which means any length is OK.
#'
#' @return \code{TRUE} or \code{FALSE}
#'
#' @export
#' @examples
#' is_positive_integer(NULL)
#' is_positive_integer(as.integer(NA))
#' is_positive_integer(integer(0))
#' is_positive_integer(3.0)
#' is_positive_integer(3.3)
#' is_positive_integer(1:5)
#' is_positive_integer(1:5, len = c(2, 10))
#' is_positive_integer(1:5, len = c(2:10))
is_positive_integer <-
function(x, len = NULL) {
  if (!class(x) %in% c("numeric", "integer")) {
    y <- FALSE
  }
  else if (length(x)==0){
	y <- FALSE
  } 
  else {
    as_int <- as.integer(x)
    check_diff <- x - as_int
    if (all(check_diff %in% c(0))) {
      y <- ifelse(all(as_int > 0), TRUE, FALSE)
    }
    else {
      y <- FALSE
    }
  }
  if (y == FALSE) {
    return(FALSE)
  }
  else {
    if (is.null(len)) {
      return(TRUE)
    }
    else {
      y <- ifelse(length(x) %in% len, TRUE, FALSE)
      return(y)
    }
  }
}