R/utils.R

In dsidavis/Rgoodreads: Access GoodReads data via API and scraping

getSimpleFields =
function(x, dataframe = TRUE, ...)
{
  i = xmlSApply(x, function(x)  {
                     kids = xmlChildren(x)
                     (length(kids) == 1 &&  inherits(kids[[1]], c("XMLInternalTextNode", "XMLInternalCDataNode"))) || (length(kids) == 0)
                })
  ans = structure(lapply(x[i], getFieldValue), names = sapply(x[i], xmlName))
  if(dataframe)
      as.data.frame(ans, ...)
  else
      ans
}


getFieldValue =
function(node)
{
  type = xmlGetAttr(node, "type", "")

  if(xmlGetAttr(node, "nil", FALSE) == "true")
      return(NA)

  if(xmlSize(node) == 0)
      return(NA)
  
  val = xmlValue(node)
  
  switch(type,
          integer = as.integer(val),
          boolean = as.logical(toupper(val)),
          val
         )
}


addElement =
function(x, field, order = character()) {
    i = !(field %in% names(x))
    if(any(i))
        x[field[i]] = NA

    if(length(order))
       x = x[order]
    x
}    



rbindDFs =
    #
    #  This takes a list of data frames and combines them into a single data frame.
    #  This adds missing field/columns as NAs for any data frame that doesn't have all of the
    #  variables.
    #
function(x)
{
   cts = table(unlist(sapply(x, names)))
   if(!all(cts == length(x))) {

        i = which(cts < length(x))
        var = names(cts)[i]
        x = lapply(x, addElement, var, names(cts))
   }

   do.call(rbind, x)
}
    

trim = 
function (x) 
gsub("(^[[:space:]]+|[[:space:]]+$)", "", x)