#!/usr/bin/env ruby
# webcat.rb(books)
# ISBN-search -> detail-parse
# original; isbn2bib.rb for bk1 by takahashi
# ver.0.1 2003.1.27-2003.1.27
# ver.0.2 2004.01.06-2004.01.06 新キャッシュに対応

if __FILE__ == $0
  require 'cookie'
  require 'webagent'
  require 'cache'
  require 'cachebib'
end

class CacheBib_webcatplus < CacheBib
  attr_accessor :access_id
  attr_accessor :title2, :title3, :author2
  attr_accessor :pubtype, :type, :note, :note2, :code, :case
  
  def initialize ()
    super
    @shop_id = 'wcp'
    @url_site = 'http://webcatplus.nii.ac.jp/' 
  end
  
  #===== ===== parts ===== =====
  def word_delivery ()
    return '', false
  end
  
  #===== ===== url ===== =====
  def url_isbnsearch ( isbn = @isbn )
    url = @url_site + "assoc.cgi?mode=equal0&isbn=#{isbn}"
    return url
  end
  
  def url_bibinfo ( id = @id )
    url = @url_site + "tosho.cgi?mode=tosho&#{@access_id}=#{id}"
    return url
  end
  
  #===== ===== parse ===== =====
  def parse_isbnsearch ( str )
    count = []
    str.scan(%r|<a href="tosho.cgi\?mode=tosho&(..ID)\=([\w\d]+)"[^>]+?><b>(.*?)</b></a>|){
      count << [$1, $2, $3] unless $1.nil?
    }
    if count.uniq.size > 1
      message("WAR; ISBN-search hit multi-column. ISBN; #{@isbn}")
    end
    @access_id, @id, @title = count.first
    return count
  end
  
  def parse_bibinfo ( str )
    #----- set key and data ------
    mode_tbl = {
#      :data1     => '<div style="margin-top:11px;">',
      :desc      => '<img src="/webcatplus/widgets/images/etc/b01_naiyou.gif" border="0" width="46" height="13" alt="内容" title="内容" />',
      :index     => '<img src="/webcatplus/widgets/images/etc/b01_mokuji.gif" border="0" width="46" height="14" alt="目次" title="目次" />',
#      :data2     => '<img src="/webcatplus/widgets/images/etc/b01_jyouho.gif" border="0" width="54" height="13" alt="書誌情報" title="書誌情報" />',
      :title     => '<font size="2" class="fs100">タイトル</font>',
      :author    => '<font size="2" class="fs100">責任表示</font>',
      :type      => '<font size="2" class="fs100">資料種別</font>',
      :pubtype   => '<font size="2" class="fs100">版表示</font>',
      :publisher => '<font size="2" class="fs100">出版事項</font>',
      :pubsize   => '<font size="2" class="fs100">形態事項</font>',
      :isbn      => '<font size="2" class="fs100">ISBN</font>', #dummy
      :note2     => '<font size="2" class="fs100">内容著作注記</font>',
      :note      => '<font size="2" class="fs100">注記</font>',
      :title2    => '<font size="2" class="fs100">別タイトル</font>',
      :title3    => '<font size="2" class="fs100">統一タイトル</font>',
      :author2   => '<font size="2" class="fs100">著者標目</font>',
      :code      => '<font size="2" class="fs100">分類</font>',
      :case      => '<font size="2" class="fs100">件名</font>'
    }
    mode = :none
    data = Hash.new()
    
    #----- read and sepalate by key(mode_tbl) ------
    str.each{|line|
      l =  line.chomp.chomp("\r").kconv(@insystemKcode)
      mode_tbl.each{ |key,keydata|
        mode = key if /#{keydata}/s =~ l
      }
      if mode != :none
        if data[mode]
          data[mode] << l
        else
          data[mode] = l
        end
      end
    }
    
    #----- parse -----
    keys = [:desc, :index]
    stocks = [nil, nil]
    keys.each_index{ |i|
      key = keys[i]
      if !data[key].nil?
        if data[key] =~ /#{mode_tbl[key]}.+?<td>(.+?)<\/td>/smi
          stock = $1.gsub(/\s+/, ' ').gsub(/<br>/i, "\t").gsub(/<[^>]+>/,'').strip
          stocks[i] = stock if stock != ''
        end
      end
    }
    @desc, @index = stocks
    
    keys = [:title, :author, :type, :pubtype, :publisher, :pubsize, :note2, :note, :title2, :title3, :author2, :code, :case]
    stocks = [nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil, nil]
    keys.each_index{ |i|
      key = keys[i]
      if !data[key].nil?
        if data[key] =~ /#{mode_tbl[key]}.+?<td>(.+?)<\/td>/smi
          stock = $1.gsub(/\s+/, ' ').gsub(/<br>/i, "\t").gsub(/<[^>]+>/,'').strip
          stocks[i] = stock if stock != ''
        end
      end
    }
    @title, @author, @type, @pubtype, @publisher, @pubsize, @note2, @note, @title2, @title3, @author2, @code, @case = stocks
    if @title =~ /\((.+?)\)$/
      @title, @label = $`.strip, $1.strip
    end
    @author2.gsub!(/&nbsp;/, '') unless @author2.nil?
    if @pubsize =~ /;/
      str1, str2 = $`.strip , $'.strip
      @page, @pubsize = str1, str2 if str1 =~ /p/
      @page, @pubsize = str2, str1 if str2 =~ /p/
    end
    @note2.gsub!(/&nbsp;/, '') unless @note2.nil?
  end
  
  def set_info #===== set book-info(@info) from local-value =====
    super
    #----- set info add ----- #webcat plus独自の情報類
    @info["access_id"]   = @access_id
    @info["title2"]     = @title2
    @info["title3"]     = @title3
    @info["author2"]    = @author2
    @info["pubtype"]    = @pubtype
    @info["type"]       = @type
    @info["note"]       = @note
    @info["note2"]      = @note2
    @info["code"]       = @code
    @info["case"]       = @case
  end
end

if __FILE__ == $0
  p CacheBib_webcatplus.new.get_data('4-15-010739-4')
end