R/PageProcessing.R

In ropensci/reol: R Interface to the Encyclopedia of Life

#' Process XML Data into tree format
#'
#' This function will read in the XML data and parse it into a tree structure for R to read.
#'
#' @export
#'
#' @param MyEOL A filename or R object for downloaded EOL pages
#' @param MyFiles A vector of filenames or R objects for EOL  or Hier pages
#' @param ... Parameters passed on to \code{\link[XML]{xmlRoot}}
#'
#' @return \code{PageProcessing} returns XML tree as an R object. \code{RemoveNAFiles} is an
#' internal function that will take a vector of filenames and check to make sure they all have
#' data associated (sometimes requested pages will be empty and contain no information).  If they
#' are empty, they are cleaved from the analysis.
#'
#' @examples \dontrun{
#' # Reads in a file
#' # PageProcessing("eol1.xml")
#'
#' # or reads in an R object
#' data(MyEOLs)
#' PageProcessing(MyEOLs[1])
#' }

PageProcessing <- function(MyEOL, ...) {
  if(class(MyEOL) == "character" || class(MyEOL) == "vector")
    res <- xmlToList(xmlRoot(xmlParse(MyEOL, getDTD=FALSE), ...), simplify=FALSE)
  if(class(MyEOL) == "list")
    res <- xmlToList(xmlRoot(xmlParse(MyEOL[[1]], getDTD=FALSE), ...), simplify=FALSE)
  if(is.na(GetHierID(MyEOL)))
    return(paste("Filenames contain NAs"))
  if(!is.null(res$error)){
    paste("Bad file", MyEOL, "has an error:", res$error)
    return(NULL)
  }
  if(is.null(res$error))
    return(res)
}

#' @rdname PageProcessing
#' @export
RemoveNAFiles <- function(MyFiles){
  if(any(is.na(GetHierID(MyFiles)))) {
    whichNAs <- which(is.na(GetHierID(MyFiles)))
    MyFiles <- MyFiles[-whichNAs]
  }
  return(MyFiles)
}