#!/usr/bin/env ruby
# rcom_hkft.rb  ver.0.1 2002.4.10-2003.1.20

require 'net/http'
require 'cgi'
#require 'simpleuri'
#require 'webagent'

#今のところ早川文庫FT書評リンクのみ

class CacheBib_rcom < CacheBib
  attr_accessor :isbn
  attr_accessor :flag_review
  attr_accessor :url_site
  attr_accessor :title, :author, :publisher, :series, :pubdate, :translator
  attr_accessor :image, :author_id, :book_id, :comment
  attr_accessor :e_title, :e_author, :e_pubdate
  attr_accessor :url_ameq, :url_seraeno, :url_buy
  attr_accessor :reviews
  attr_accessor :insystemKcode, :outputKcode, :messageKcode

  def initialize ()
    super
    @reviews = Array.new()
    @flag_review = true
    @url_site = 'http://ippo.s5.xrea.com/dk/' 
    @insystemKcode = 3 #sjis #1-jis,2-euc,3-sjis
    @outputKcode   = 2 #euc
    @messagetKcode = 3 #sjis
  end

#===== ===== parts ===== =====

#===== ===== url ===== =====
  def url_list
    url = @url_site + 'list.html'
    return url
  end

  def url_bibinfo ()
    url = @url_site + "isbn#{@isbn.delete('-').upcase}.html"
    return url
  end

#===== ===== parse ===== =====
  def parse_list (str)
    @list = []
    str.each{ |line|
      if line =~ %r|(.+)<a href="([^"]+).*>(.+?)</a>(.+)\((\d+)\)<br>|i then
        @hcode, @url_file, @title, @author, @num_review = $1, $2, $3, $4, $5
        if @title =~ /(.+)《(.+)》/s
          @title, @series = $1, $2
        end
        @isbn = isbn_check('4-15-020' + @hcode[2..4].to_s)
        record = {}
        record["isbn"]       = @hcode
        record["url_file"]   = @url_file
        record["h-code"]     = @hcode
        record["title"]      = @title
        record["series"]     = @series
        record["author"]     = @author
        record["num_review"] = @num_review
        @list << record
      end
    }
  end

  def parse_bibinfo (str)
    #----- set key and data ------
    #----- read and sepalate by key(mode_tbl) ------
    str.collect{|line| line.chomp.chomp("\r").kconv(@insystemKcode)}
    block = str.split(/<hr>/i)
    block.shift
    return if block.size == 0

    #----- parse -----
    line = block.shift.delete("\n").split(/<br>/i)
    @publisher  = line[0].strip
    @hcode      = line[1].strip
    @title      = line[2].strip
    @author     = line[3].strip
    @translater = line[4].strip.chomp("訳")
    @pubdate    = line[5].strip
    @isbn       = line[6].strip.gsub(/^ISBN/,'')
    @comment    = line[7].strip if !line[7].nil?
    if @title =~ /(.+)(《(.+)》)?.*　\((.+)\)/s
      @title, @series, book_id = $1, $2, $4
    end
    if @author =~ /(.+)　\((.+)\)/s
      @author, @author_id = $1, $2
      if @author_id =~ %r|<a href="([^"]+).+>(.+)</a>|
        @url_author, @author_id = $1, $2
      end
    end

    line = block.shift.delete("\n").split(/<br>/i)
    line.each{ |rline|
      if rline =~ %r|<a href="([^"]+?)">(.+?)</a> さん　<a href="([^"]+?)">(.+?)</a>|
        @url_reviewer, @reviewer, @url_review, @title_review = $1, $2, $3, $4
      end
      review = {}
      review["url_reviewer"] = @url_reviewer
      review["reviewer"]     = @reviewer
      review["url_reviewe"]  = @url_review
      review["title_review"] = @title_review
      @reviews << review
    }

    if block[0] =~ %r|<img|
      line = block.shift
      line.scan(%r|<img src="([^"]+?)|){ |url|
        @image << "\t" if !@image.nil?
        @image =  ''   if @image.nil?
        @image << url
      }
    end

#    if line =~ %r|<a href="([^"]+?)">瀬良ヱ野図書館</a>|
#      @url_seraeno = $1
#    end

    line = block.shift.delete("\n").split(/<br>/i)
    @e_title      = line[0].strip
    @e_author     = line[1].strip
    @e_pubdate    = line[2].strip

    line = block.shift
    if line =~ %r|<a href="([^"]+?)">AMEQ LAND</a>|
      @url_ameq = $1
    end
    if line =~ %r|<a href="([^"]+?)">瀬良ヱ野図書館</a>|
      @url_seraeno = $1
    end
    if line =~ %r|<a href="([^"]+?)">↓</a>|
      @url_buy = $1
    end

    line = block.shift
    if line =~ %r|Last Updated (.+)<|
      @lastupdate = $1
    else
      message = 'WAR; maybe parse error; ' + @isbn
      @stderr.puts message.kconv(@messageKcode)
    end
  end

#===== ===== main ===== =====
  def set_info #===== set local-value from @info (from cache) =====
    #----- set book-info (from cache) -----
    @flag_overprint  = true if @info["overprint"]
    @isbn            = @info["isbn"] unless @info["isbn"].nil?
  end

  def get_info #===== get book-info(@info) from url =====
    message = 'CHK; access; ' + @url_site + ' -> ' + @isbn
    @stderr.puts message.kconv(@messageKcode)
    #----- get detail-code -----
    #----- get book-info -----
    agent = WebAgent.new()
    agent.uri = url_bibinfo
    agent.get()
    body = agent.body.kconv(@insystemKcode)
    parse_bibinfo(body)

    if !@flag_overprint
      @info = {}
    end
    @info["isbn"]       = @isbn
    @info["title"]      = @title
    @info["sereis"]     = @series
    @info["author"]     = @author
    @info["translator"] = @translator
    @info["publisher"]  = @publisher
    @info["pubdate"]    = @pubdate
    @info["image"]      = @image

    @info["author_id"]    = @author_id
    @info["book_id"]      = @book_id
    @info["e_title"]      = @e_title
    @info["e_author"]     = @e_author
    @info["e_pubdate"]    = @e_pubdate
    @info["comment"]      = @comment
    @info["url_ameq"]     = @url_ameq
    @info["url_seraeno"]  = @url_seraeno
    @info["url_buy"]      = @url_buy
    @info["lastupdate"]   = @lastupdate

    @info["overprint"]  = true if @flag_overprint

    if @flag_review
      @info["reviews"] = @reviews
    end
  end
end