#' Word count distribution from directory of literature
#'
#' This function accepts a folder of pdf journal articles and returns the
#' frequency of scientifically relevant terms
#'
#' @param literatureFileDirectory directory of pdf files to count words from.
#' @param omitStop - if TRUE omit stop words
#' @param omitScientific - if TRUE omit words common to science not unique
#' @param drawCloud - if TRUE draw word cloud
#' @return TODO: figure this out
#' @examples wordCountDistFromLitFolder(literatureFileDirectory)
#'
#'

wordCountDistFromLitFolder = function (literatureFileDirectory = "",
                                       omitStop = TRUE,
                                       omitUselessTerms = TRUE,
                                       firstWordRankInPlot = 1,
                                       lastWordRankInPlot = 30,
                                       freqPlot = TRUE,
        uselessTerms = c("result", "fig", "tabl", "use", "research", "total",
                        "can")
                                       ){
    library(pdftools)
    library(NLP)
    library(tm)
    library(tidytext)
    library(reshape2)
    library(ggplot2)

    if (!dir.exists(literatureFileDirectory)){
      stop("The specified directory does not exist")
    }

    setwd(literatureFileDirectory)

    literatureCorpus <-
      Corpus(URISource(list.files(pattern = "pdf$")),
             readerControl = list(reader = readPDF))

    #remove potentially problematic symbols
    toSpace <- content_transformer(function(x, pattern) {
      return (gsub(pattern, ' ', x))
      })
    literatureCorpus <- tm_map(literatureCorpus, toSpace, '-')
    literatureCorpus <- tm_map(literatureCorpus, toSpace, '’')
    literatureCorpus <- tm_map(literatureCorpus, toSpace, '‘')
    literatureCorpus <- tm_map(literatureCorpus, toSpace, '•')
    literatureCorpus <- tm_map(literatureCorpus, toSpace, '”')
    literatureCorpus <- tm_map(literatureCorpus, toSpace, '“')

     tdm <- TermDocumentMatrix(
              literatureCorpus,
              control = list(removePunctuation = TRUE,
                             stopwords = TRUE,
                             tolower = TRUE,
                             stemming = TRUE,
                             removeNumbers = TRUE)
              )

    frequentTerms <- findFreqTerms(tdm, lowfreq = 1, highfreq = Inf)

    if (omitUselessTerms){
       frequentTerms <- frequentTerms[!(frequentTerms %in% uselessTerms)]
    }

    frequentTerms.tdm <- as.matrix(tdm[frequentTerms,])

    wordcount <- rowSums(frequentTerms.tdm)

    if (freqPlot){

      topWords <- sort(wordcount, decreasing=TRUE)[
        c(firstWordRankInPlot:lastWordRankInPlot)]

      dfplot <- as.data.frame(melt(topWords))
      dfplot$word <- dimnames(dfplot)[[1]]
      dfplot$word <-
        factor(dfplot$word,
               levels = dfplot$word[order(dfplot$value, decreasing=TRUE)])

      fig <- ggplot(dfplot, aes(x=word, y=value)) + geom_bar(stat="identity")
      fig <- fig + xlab("Word in pdf files")
      fig <- fig + theme(axis.text.x = element_text(angle = 90))
      fig <- fig + ylab("Count")
      print(fig)
    }

    return(data.frame(word = frequentTerms, wordcount = wordcount))
}