aozoragen 0.0.1 → 0.0.2

data/bin/aozoragen

@@ -15,6 +15,9 @@ ARGV.each do |u|
 	when 'sai-zen-sen.jp'
 		require 'aozoragen/sai-zen-sen'
 		book = SaiZenSen::new( uri )
+	when 'renzaburo.jp'
+		require 'aozoragen/renzaburo'
+		book = Renzaburo::new( uri )
 	else
 		p "Error: unknown URI: #{uri}"
 	end

data/lib/aozoragen/renzaburo.rb

@@ -0,0 +1,52 @@
+# -*- coding: utf-8; -*-
+#
+# scraping renzaburo.jp
+#
+require 'nokogiri'
+require 'open-uri'
+require 'pathname'
+
+class Renzaburo
+	def initialize( index_uri )
+		@index_uri = URI( index_uri.to_s.sub( /index\.html$/, '' ) )
+		@index_html = Nokogiri( open( @index_uri, 'r:CP932', &:read ) )
+	end
+
+	def metainfo
+		info = {:id => Pathname( @index_uri.path ).basename.to_s}
+		(@index_html / 'title').each do |t|
+			info[:title] = t.text.sub( /｜.*/, '' )
+		end
+		(@index_html / 'div.textBlock strong' ).each do |st|
+			info[:author] = [st.text]
+		end
+		info
+	end
+
+	def each_chapter
+		book_title = metainfo[:title]
+		(@index_html / 'ul.btnList li.withDate a' ).each do |a|
+			uri = @index_uri + a.attr( :href )
+			text = get_content( uri, book_title )
+			yield( {id: Pathname( uri.path ).dirname.basename.to_s, uri: uri, text: text} )
+		end
+	end
+
+	def get_content( uri, book_title = '' )
+		text = ''
+		html = open( uri, 'r:CP932', &:read ).encode( 'UTF-8' )
+		html = html.gsub( /\—/, "\u2500" ).gsub( /\"/, "\u201D" )
+		(Nokogiri( html ) / 'div#mainContent' ).each do |content|
+			(content / 'h3').each do |t|
+				title = t.text.sub( /^『#{book_title}』　/, '' )
+				text << "\n　　　　　#{title}\n\n"
+			end
+			(content / 'div.textBlock p' ).each do |para|
+				next if /＜次回につづく＞/ =~ para.text
+				text << '　' * 10 if (para.attr('class') || '').index( 'txtAlignC' )
+				text << para.text.gsub( /<br>/, "\n" ) << "\n\n"
+			end
+		end
+		text << "［＃改ページ］\n"
+	end
+end

data/lib/aozoragen/version.rb

@@ -1,3 +1,3 @@
 module Aozoragen
-  VERSION = "0.0.1"
+  VERSION = "0.0.2"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: aozoragen
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.0.2
   prerelease: 
 platform: ruby
 authors:
@@ -13,7 +13,7 @@ date: 2012-02-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
-  requirement: &71959020 !ruby/object:Gem::Requirement
+  requirement: &78022280 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ! '>='
@@ -21,7 +21,7 @@ dependencies:
         version: '0'
   type: :runtime
   prerelease: false
-  version_requirements: *71959020
+  version_requirements: *78022280
 description: Scraping some Ebook web site and generating AOZORA format text files.
 email:
 - t@tdtds.jp
@@ -38,6 +38,7 @@ files:
 - bin/aozora2pdf
 - bin/aozoragen
 - lib/aozoragen.rb
+- lib/aozoragen/renzaburo.rb
 - lib/aozoragen/sai-zen-sen.rb
 - lib/aozoragen/version.rb
 homepage: ''