FTP tree mapper: save an FTP site's directory stucture as a list

map_ftp.r

#' FTP tree mapper
#' Save an FTP site's directory stucture as a list.
#' @author Aaron Wolen
#' 
#' @example
#' url <- 'ftp://ftp.genboree.org/EpigenomeAtlas/Current-Release/experiment-sample' 
#' roadmap <- map_ftp(url = url, dirs = "Histone_H2BK120ac", recursive = TRUE)

map_ftp <- function(url, dirs, recursive = FALSE) {
  require(RCurl, quietly = TRUE)
  
  # Add trailing forward slash
  add_fslash <- function(x) sub("/*$", "/", x)
  
  # Verify required curl options
  list.opts <- c("dirlistonly", "ftplistonly")
  curl.opts <- listCurlOptions()
  if ( any(list.opts %in% curl.opts) ) {
    opts <- list()
    opts[[list.opts[list.opts %in% curl.opts][1]]] <- TRUE
  } else {
    stop("Your version of curl lacks the listonly option.", call. = FALSE)
  }
  
  # Print file list in FTP directory
  list_files <- function(url) {
    url.con <- textConnection(getURL(url, ftp.use.epsv = FALSE, .opts = opts))
    url.out <- readLines(url.con)
    close(url.con)
    url.out <- url.out[url.out != ""]
    return(url.out)
  }
  
  url <- add_fslash(url)
  if (missing(dirs)) dirs <- list_files(url)
  
  tree <- list()
  for(d in dirs) {
    d.url <- add_fslash(paste(url, d, sep = ""))
    
    if (!url.exists(d.url)) next
    
    d.out <- list_files(d.url)
 
    if(recursive) {
      
      # Test if each string represents a file or a new directory
      d.out.urls <- paste(d.url, d.out, sep = "")
      is.file <- sapply(d.out.urls, url.exists)
      is.dir <- sapply(add_fslash(d.out.urls), url.exists)
      
      d.files <- d.out[is.file & !is.dir]
      d.dirs <- setdiff(d.out[is.dir], d.files)
      
      tree[[d]] <- d.files
      if (length(d.dirs) > 0) {
        tree[[d]] <- c(tree[[d]], ftp_tree(d.url, dirs = d.dirs, recursive))
      }
        
    } else {
      tree[[d]] <- d.out
    }    
  }
  return(tree)
}