A function to audit disk usage by subdirectory

du.R

#' A function to audit disk usage by subdirectory
#'
#' @param path the path to audit 
#' @param extra character extra arguments for linux `du` command
#' @param app character the `du` executable path
#' @return data frame or NULL
#'  \itemize{
#'      \item{size numeric in megabytes}
#'      \item{name character}
#'  }
#' @examples
#'  \dontrun{
#'   du_subdirectories(path = "/opt/data")
#' #     size                    name
#' # 1    926         /opt/data/bathy
#' # 2    275        /opt/data/cpcugb
#' # 3    332     /opt/data/jellycast
#' # 4      1          /opt/data/misc
#' # 5  18516          /opt/data/narr
#' # 6  37825        /opt/data/natmap
#' # 7   1132 /opt/data/natural_earth
#' # 8      1          /opt/data/ncep
#' # 9      1      /opt/data/salpcast
#' # 10  4193           /opt/data/sat
#' # 11  4526      /opt/data/tickcast
#' # 12   144            /opt/data/wx
#' }
du_subdirectories <- function(path = "/opt/data",
    extra = "-ms",
    app = Sys.which("du") ){
  
    # btupper@ecocast ~ $ du -hs /opt/data/jellycast
    # 332M	/opt/data/jellycast
    
    dd <- list.dirs(path[1], 
        recursive = FALSE, full.names = TRUE)
    
    if (length(dd) <= 0) {
        x <- data.frame(size = numeric(), name = character(),
            stringsAsFactors = FALSE)
        return(x)
    }
    
    CMD <- sprintf("%s %s %s", app, extra, dd)
    
    xx <- strsplit(sapply(CMD, system, intern = TRUE), '\t')
    x <- do.call(rbind, xx)
    
    x <- data.frame(size = as.numeric(x[,1]), 
        name = x[,2], stringsAsFactors = FALSE)
    rownames(x) <- NULL
    x
}