(PhantomJS) Parse Bleach episodes in Wikipedia

bleach_parse.coffee

page = require('webpage').create()
fs = require 'fs'
system = require 'system'

page.onConsoleMessage = (msg) -> console.log(msg)

page.open "http://en.wikipedia.org/wiki/Bleach_episodes", (status) ->
  csv = page.evaluate ->
    formatDate = (date) ->
      date.getFullYear() + "-" + ((if date.getMonth() < 9 then "0" else "")) + (date.getMonth() + 1) + "-" + ((if date.getDate() < 10 then "0" else "")) + date.getDate()
    
    ret = ""
    i = 1
    totalEps = 366
    while i <= totalEps
      text = $("#ep" + i).parent().children("td.summary").text()
      try
        [_, englishTitle, romajiTitle, kanjiTitle] = text.match(/"(.*)"\s"(.*)"\s\((.*)\)/)
        date = $("#ep#{i}").parent().children("td.summary").next().text().replace(/\[.*\]/gi,'')
        date = formatDate(new Date(date))
      catch e
        console.log "Episode: ", i
        console.log "Error:", e
        return      
      ret += "\"#{i}\", \"#{englishTitle}\", \"#{romajiTitle}\", \"#{kanjiTitle}\", \"#{date}\"\n"
      i++
    ret

  filename = system.args[1] or "bleach_episodes.csv"
  fs.remove filename
  fs.write filename, csv, "w"
  phantom.exit()

I've also used XPath. This was in my notes: //table[@class="wikitable" and position()<17]//td[position()<4]