#!/usr/bin/ruby -Ke

=begin

製作開始：2005/03/01
最終更新：2005/03/03

=end



require "kconv"

$stdout.sync = 1


$cookiefile = "cookies.txt" # mozzila形式
$useragent = "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.0)"
$retry_wait = 180
$each_wait = 30

$url_filename = "org_url.txt"

$proxy = "-Y "

$板名s = ["", "アニキャプ", "萌え絵", "その他", "まったり", "お絵かき"]


# ======================================================================
class Webpage
  def initialize(page)
    @p = 0
    @prev_link = ""
    @lines = []
    for line in page
      line.strip!
      line.gsub!(/</, "\n<")
      line.gsub!(/>/, ">\n")
      for elm in line.to_a
	elm.chomp!
	elm.sub!(/^\s$/, '')
	if /^</ =~ elm
	  case elm
	  when /html>/i, /head>/i, /title>/i, /body>/i, /table>/i, /<table/i, /tbody>/i, /<tr/i, /tr>/i, /<td/i, /td>/i, /<a/i, /a>/i, /<!/i, /<input/i, /<form/i, /form>/i, /<select/i, /select>/i, /<option/i, /option>/i, /<img/i
	    @lines.push(Kconv::toeuc(elm))
	  end
	elsif elm != ""
	  @lines.push(Kconv::toeuc(elm))
	end
      end
    end
  end
  attr_accessor :p, :lines
end		



# ======================================================================
def wait_s(s)
  print "WAIT #{s}s : "
  s.times do |i|
    print "."
    sleep(1)
  end
  print "\n"
end


# ======================================================================
def get_moejp(url)
  /res\d+c(\d)/ =~ url
  w = $1.to_i
  板名 = $板名s[w]

  wget = open("|wget -O - -U \"#{$useragent}\" --cookies on --load-cookies #{$cookiefile} \"#{url}\"", "r")
  page = []
  while line = wget.gets
    page.push(line)
  end
  wget.close

  moe_html = Webpage.new(page)

  タイトル = moe_html.lines[moe_html.lines.rindex("<title>")+1]
  タイトル.gsub!(/\'/, "’")
  タイトル = 板名 + "/" + タイトル

  print "#{タイトル}\n"

  grp_uris = []
  for i in 0..moe_html.lines.size-1 do
    line = moe_html.lines[i]
    if line == "画像："
      /href=\"(http.+?)\"/ =~ moe_html.lines[i+1]
#      /href=\"(http.*(?:(?:jpg)|(?:png)|(?:gif)|(?:mpg)|(?:avi)|(?:wmv)))/ =~ moe_html.lines[i+1]
      grp_uris << $1
      i = i+1
    end
  end

  grp_uris.uniq!


#  wget = open("|wget -i - -U \"#{$useragent}\" --cookies on --load-cookies #{$cookiefile} --referer='#{url}' -P '#{タイトル}' -nc --wait 10 --random-wait --waitretry=10 --read-timeout=30 --limit-rate=4k --tries 5", "w")

  count = 1
  wait = $retry_wait
  grp_uris.each do |uri|
    print "\n#{count} / #{grp_uris.size}\n"
    filename = uri.split("/").last
    fullpath = "#{タイトル}/#{filename}"
    if FileTest.exist?(fullpath)
      print "#{fullpath} は既にあります。\nスキップします。\n"
      count = count+1
      next
    else
      ret = system("wget -U \"#{$useragent}\" --cookies on --load-cookies #{$cookiefile} --referer='#{url}' -P '#{タイトル}' -nc --wait 10 --random-wait --waitretry=10 --connect-timeout=10 --read-timeout=30 --tries 3 #{uri}")
      #    system("wget -U \"#{$useragent}\" --cookies on --load-cookies #{$cookiefile} --referer='#{url}' -P '#{タイトル}' -nc --wait 10 --random-wait --waitretry=10 --read-timeout=30 --limit-rate=2k --tries 5 #{uri}")
      if ret != true
        File.unlink(fullpath) if FileTest.exist?(fullpath)
        print "sleep #{wait}\n"
        wait_s(wait)
        wait = wait *2
        redo
      else
        wait = $retry_wait
        count = count+1
        wait_s($each_wait)
      end
    end
  end
  
  url_file = open(タイトル + "/#{$url_filename}", "w")
  url_file.print "#{url}\n"
  url_file.close
end


# **********************************************************************
# main
ARGV.each do |url|
  get_moejp(url)
end