R/rm.newline.chars.R

In readmoRe: Utilities for Importing and Manipulating Biomedical Data Files

# TODO: handle newline characters in data frames: remove at end of line or replace by space within line
# 
# Author: vidal
###############################################################################


#' Remove 'newline' Characters From Imported Excel Sheets
#' @description \command{rm.newline.chars} removes \sQuote{newline} characters (\code{\\n}) from any column of a data frame.
#' @param x (\code{data.frame}). A data frame resulting from an imported Excel sheet by means of \command{read.xls}.
#' @param verbose (\code{logical}). Should verbose output be printed, defaults to \code{TRUE}.
#' @details \sQuote{Newline} characters in data frame rows are read verbatim and will cause rows in output
#'     text files to be distributed across two ore more lines. Such characters, entered accidentally or
#'     deliberately in the source Excel file, should be avoided. This function removes all \sQuote{newline}
#'     characters found at the end of a line or replaces them when found within the line text.
#' @return A data frame.
#' @keywords utilities
#' @export
rm.newline.chars <-
                function(x, verbose = TRUE)
{
        
        nx <- names(x)
        nle <- try(apply(x, 2, function(z) length(grep("\n[[:blank:]]{0,}$", z)) > 0), silent = TRUE)
        if (is(nle, "try-error")) return(x)
        if (any(nle)) {
                if (verbose) cat(paste("Found newline characters at end of line in column(s) _", names(x)[nle], "_.", sep=""), "Removing from:\n")
                x <- lapply(1 : ncol(x), function(z) {
                                        if (verbose) {
                                                if (names(x)[z] %in% names(x)[nle]) cat("  ", names(x)[z], "\n")
                                        }
                                        if (is.character(x[[z]])) x[[z]] <- sub("\n[[:blank:]]{0,}$", "", x[[z]])
                                        x[[z]]
                                })
                x <- as.data.frame(x, stringsAsFactors=FALSE)
                names(x) <- nx
                if (verbose) cat("...done\n\n")
        }
        nl <- try(apply(x, 2, function(z) length(grep("\n", z)) > 0), silent = TRUE)
        if (is(nl, "try-error")) return(x)
        if (any(nl)) {
                if (verbose) cat(paste("Found newline characters within line in column(s) _", names(x)[nle], "_.", sep=""), "Removing from:\n")
                x <- lapply(1 : ncol(x), function(z) {
                                        if (verbose) {
                                                if (names(x)[z] %in% names(x)[nl]) cat("  ", names(x)[z], "\n")
                                        }
                                        if (is.character(x[[z]])) x[[z]] <- sub("\n", " ", x[[z]])
                                        x[[z]]
                                })
                x <- as.data.frame(x, stringsAsFactors=FALSE)
                names(x) <- nx
                if (verbose) cat("...done\n\n")
        }
        return(x)
        
}