Parse Google Ngram Viewer

ngram_parse.R

ngram_parse <- function(html){
  if (any(grepl("No valid ngrams to plot!",
                html))) stop("No valid ngrams.") 
    
  cols <- lapply(strsplit(grep("addColumn", html,
                               value=TRUE), ","),
                getElement, 2)
  
  cols <- gsub(".*'(.*)'.*", "\\1", cols)

  html <- paste(html[-(1:grep("data.addRows\\(", html))],
                collapse='')
  html <- gsub("\\).*", "", html)
  
  data <- as.data.frame(t(sapply(fromJSON(html),
                                 unlist)))
  colnames(data) <- cols
  return(data)
}