aozoragen 0.0.1 → 0.0.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- data/bin/aozoragen +3 -0
- data/lib/aozoragen/renzaburo.rb +52 -0
- data/lib/aozoragen/version.rb +1 -1
- metadata +4 -3
data/bin/aozoragen
CHANGED
@@ -0,0 +1,52 @@
|
|
1
|
+
# -*- coding: utf-8; -*-
|
2
|
+
#
|
3
|
+
# scraping renzaburo.jp
|
4
|
+
#
|
5
|
+
require 'nokogiri'
|
6
|
+
require 'open-uri'
|
7
|
+
require 'pathname'
|
8
|
+
|
9
|
+
class Renzaburo
|
10
|
+
def initialize( index_uri )
|
11
|
+
@index_uri = URI( index_uri.to_s.sub( /index\.html$/, '' ) )
|
12
|
+
@index_html = Nokogiri( open( @index_uri, 'r:CP932', &:read ) )
|
13
|
+
end
|
14
|
+
|
15
|
+
def metainfo
|
16
|
+
info = {:id => Pathname( @index_uri.path ).basename.to_s}
|
17
|
+
(@index_html / 'title').each do |t|
|
18
|
+
info[:title] = t.text.sub( /|.*/, '' )
|
19
|
+
end
|
20
|
+
(@index_html / 'div.textBlock strong' ).each do |st|
|
21
|
+
info[:author] = [st.text]
|
22
|
+
end
|
23
|
+
info
|
24
|
+
end
|
25
|
+
|
26
|
+
def each_chapter
|
27
|
+
book_title = metainfo[:title]
|
28
|
+
(@index_html / 'ul.btnList li.withDate a' ).each do |a|
|
29
|
+
uri = @index_uri + a.attr( :href )
|
30
|
+
text = get_content( uri, book_title )
|
31
|
+
yield( {id: Pathname( uri.path ).dirname.basename.to_s, uri: uri, text: text} )
|
32
|
+
end
|
33
|
+
end
|
34
|
+
|
35
|
+
def get_content( uri, book_title = '' )
|
36
|
+
text = ''
|
37
|
+
html = open( uri, 'r:CP932', &:read ).encode( 'UTF-8' )
|
38
|
+
html = html.gsub( /\—/, "\u2500" ).gsub( /\"/, "\u201D" )
|
39
|
+
(Nokogiri( html ) / 'div#mainContent' ).each do |content|
|
40
|
+
(content / 'h3').each do |t|
|
41
|
+
title = t.text.sub( /^『#{book_title}』 /, '' )
|
42
|
+
text << "\n #{title}\n\n"
|
43
|
+
end
|
44
|
+
(content / 'div.textBlock p' ).each do |para|
|
45
|
+
next if /<次回につづく>/ =~ para.text
|
46
|
+
text << ' ' * 10 if (para.attr('class') || '').index( 'txtAlignC' )
|
47
|
+
text << para.text.gsub( /<br>/, "\n" ) << "\n\n"
|
48
|
+
end
|
49
|
+
end
|
50
|
+
text << "[#改ページ]\n"
|
51
|
+
end
|
52
|
+
end
|
data/lib/aozoragen/version.rb
CHANGED
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: aozoragen
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.0.
|
4
|
+
version: 0.0.2
|
5
5
|
prerelease:
|
6
6
|
platform: ruby
|
7
7
|
authors:
|
@@ -13,7 +13,7 @@ date: 2012-02-23 00:00:00.000000000 Z
|
|
13
13
|
dependencies:
|
14
14
|
- !ruby/object:Gem::Dependency
|
15
15
|
name: nokogiri
|
16
|
-
requirement: &
|
16
|
+
requirement: &78022280 !ruby/object:Gem::Requirement
|
17
17
|
none: false
|
18
18
|
requirements:
|
19
19
|
- - ! '>='
|
@@ -21,7 +21,7 @@ dependencies:
|
|
21
21
|
version: '0'
|
22
22
|
type: :runtime
|
23
23
|
prerelease: false
|
24
|
-
version_requirements: *
|
24
|
+
version_requirements: *78022280
|
25
25
|
description: Scraping some Ebook web site and generating AOZORA format text files.
|
26
26
|
email:
|
27
27
|
- t@tdtds.jp
|
@@ -38,6 +38,7 @@ files:
|
|
38
38
|
- bin/aozora2pdf
|
39
39
|
- bin/aozoragen
|
40
40
|
- lib/aozoragen.rb
|
41
|
+
- lib/aozoragen/renzaburo.rb
|
41
42
|
- lib/aozoragen/sai-zen-sen.rb
|
42
43
|
- lib/aozoragen/version.rb
|
43
44
|
homepage: ''
|