東進の過去問DLサービスの問題PDFと解答PDFを自動生成するスクリプト

toshin_fuck.rb

# encoding: utf-8
require 'nokogiri'
require 'open-uri'
require 'net/http'
require 'kconv'
require "pdf/toolkit"
#AUTH INFO
$username = "(ENTER YOUR USERNAME)"
$password = "(ENTER YOUR PASSWORD)"
#URL
mainurl = ARGV[0]

#HTTP GET
def ToshinAuthHTTPGET(url)
  uri = URI.parse(url)
  req = Net::HTTP::Post.new(uri.path)
  req.basic_auth($username, $password)
  res = Net::HTTP.new(uri.host, uri.port).start {|http| http.request(req) }
  html = res.body.toutf8
end
def ToshinAuthSaveBinary(url,savepath)
  uri = URI.parse(url)
  req = Net::HTTP::Post.new(uri.path)
  req.basic_auth($username, $password)
  res = Net::HTTP.new(uri.host, uri.port).start {|http| http.request(req) }
  open(savepath, 'wb'){|f|
    f.write(res.body)
  }
end

#main logic
if mainurl.nil?
  p "URL empty."
  exit
end
#kakomon INFO
mainpageRAW = ToshinAuthHTTPGET(mainurl)
#print mainpageRAW
urls = []
doc = Nokogiri::HTML(mainpageRAW)
#大学名取得
univName = doc.search("title")[0].content
univName.gsub!(/nyushimondai.com - /,"")
#使えない文字を置き換える
univName.gsub!(/[?"\\\/\<>*|]/,"_")
#Get university name, year and subject name
match = /(?<univ>.*) - (?<faculty>.*) - (?<subject>.*)/.match(univName)
p match
univName = "#{match[:univ]} #{match[:subject]}"
print univName + "\n"
#一問目も追加
monpdf = nil
urls.push(mainurl)
doc.search("a").each do |atag|
  if atag.content =~ /第\d*問/
    #相対パスから絶対パスに変換する
    urls.push(URI::join(mainurl,atag["href"]).to_s)
    print "Found " + atag.content + "\n"
  elsif atag.content =~ /問題PDF/
    monpdf = URI::join(mainurl,atag["href"]).to_s
    print "Found " + atag.content + "\n"
  end
end
#p urls
#urls内のページから解答へのリンクを探しだす
answer_URL = []
urls.each do |murl|
  print "fetching " + murl + "\n"
  rawhtml = ToshinAuthHTTPGET(murl)
  html = Nokogiri::HTML(rawhtml)
  html.search("a").each do |tag|
    if tag.content =~ /解答/
      answer_URL.push(URI::join(mainurl,tag["href"]).to_s);
      break
    end
  end
end
p answer_URL
#answer_URLからGIFを探しだす
gif_url = []
answer_URL.each do |ans|
  print "fetching " + ans + "\n"
  rawhtml = ToshinAuthHTTPGET(ans)
  html = Nokogiri::HTML(rawhtml)
  gif = html.search("img")[0]["src"]
  gif_url.push(URI::join(mainurl,gif).to_s)
end
p gif_url
#gif_urlからGIFを適当に落とす
Dir.mkdir(univName) unless Dir.exist?(univName)
gif_url.each do |answer|
  print "Saving " + answer + "\n"
  fname = File.basename(answer)
  ToshinAuthSaveBinary(answer,univName + "/" + fname)
end
#trim images with ImageMagick
print "Trimming images....\n"
system("mogrify -trim +repage \"#{univName}/*.gif\"")
#convert it to PDF
print "Converting to PDF....\n"
system("convert \"#{univName}/*.gif\" \"#{univName}/#{univName} 解答.pdf\"")
#Fix Mojibake(Please somebody fix multibyte bugs on ImageMagick!!!!)
print "Fixing mojibake....\n"
my_pdf = PDF::Toolkit.open("#{univName}/#{univName} 解答.pdf")
my_pdf["Title"] = "#{univName} 解答"
my_pdf.save!
#Save PDF
print "Downloading kakomon files....\n"
ToshinAuthSaveBinary(monpdf,univName + "/" + "#{univName}.pdf")
#Fix title for some PDFs
print "Fixing title....\n"
k_pdf = PDF::Toolkit.open("#{univName}/#{univName}.pdf")
k_pdf["Title"] = "#{univName}"
k_pdf.save!
#Complete
print "ALL OK!!\n"

Update 3

間違えて最初のやつのgist消してしまいました。ごめんなさい。
PDFの文字化けを直したりとか何やらかんやら処理加えてます。
かなり切羽詰っているので処理も大分適当になってます。