History

Migrate c scraper from filescraper to urlscraper

Enoc committed 5 years ago

Commit fdfcf3d1 parent e17bc8410 files · +24 −204Browse files

Unified · Split

10 files
  1. assets/stylesheets/application.css.scss +1 −1
  2. assets/stylesheets/pages/_cppref.scss +0 −0
  3. lib/docs/filters/c/clean_html.rb +0 −116
  4. lib/docs/filters/c/entries.rb +3 −0
  5. lib/docs/filters/c/fix_code.rb +0 −21
  6. lib/docs/filters/c/fix_urls.rb +0 −11
  7. lib/docs/scrapers/c.rb +0 −42
  8. lib/docs/scrapers/cppref/c.rb +12 −0
  9. lib/docs/scrapers/cppref/cpp.rb +0 −9
  10. lib/docs/scrapers/cppref/cppref.rb +8 −4
assets/stylesheets/application.css.scss+1−1
@@ -39,7 +39,7 @@ · 38 unmodified lines
39 'pages/async',39 'pages/async',
40 'pages/bash',40 'pages/bash',
41 'pages/bootstrap',41 'pages/bootstrap',
42 'pages/c',42 'pages/cppref',
43 'pages/cakephp',43 'pages/cakephp',
44 'pages/clojure',44 'pages/clojure',
45 'pages/codeception',45 'pages/codeception',
assets/stylesheets/pages/_cppref.scssrenamed from assets/stylesheets/pages/_c.scss+0−0

File renamed without changes.

lib/docs/filters/c/clean_html.rb+0−116
@@ -1,116 +0,0 @@
1module Docs
2 class C
3 class CleanHtmlFilter < Filter
4 def call
5 css('h1').remove if root_page?
6
7 css('.t-dcl-rev-aux td[rowspan]').each do |node|
8 rowspan = node['rowspan'].to_i
9 node['rowspan'] = node.ancestors('tbody').css('tr').length if rowspan > 3
10 end
11
12 css('#siteSub', '#contentSub', '.printfooter', '.t-navbar', '.editsection', '#toc',
13 '.t-dsc-sep', '.t-dcl-sep', '#catlinks', '.ambox-notice', '.mw-cite-backlink',
14 '.t-sdsc-sep:first-child:last-child', '.t-example-live-link',
15 '.t-dcl-rev-num > .t-dcl-rev-aux ~ tr:not(.t-dcl-rev-aux) > td:nth-child(2)').remove
16
17 css('#bodyContent', '.mw-content-ltr', 'span[style]', 'div[class^="t-ref"]', '.t-image',
18 'th > div', 'td > div', '.t-dsc-see', '.mainpagediv', 'code > b', 'tbody').each do |node|
19 node.before(node.children).remove
20 end
21
22 css('div > ul').each do |node|
23 node.parent.before(node.parent.children).remove
24 end
25
26 css('dl > dd:first-child:last-child > ul:first-child:last-child').each do |node|
27 dl = node.parent.parent
28 if dl.previous_element && dl.previous_element.name == 'ul'
29 dl.previous_element << node
30 dl.remove
31 else
32 dl.before(node).remove
33 end
34 end
35
36 css('dl > dd:first-child:last-child').each do |node|
37 node.parent.before(node.children).remove
38 end
39
40 css('ul').each do |node|
41 while node.next_element && node.next_element.name == 'ul'
42 node << node.next_element.children
43 node.next_element.remove
44 end
45 end
46
47 css('h2 > span[id]', 'h3 > span[id]', 'h4 > span[id]', 'h5 > span[id]', 'h6 > span[id]').each do |node|
48 node.parent['id'] = node['id']
49 node.before(node.children).remove
50 end
51
52 css('table[style]', 'th[style]', 'td[style]').remove_attr('style')
53 css('table[cellpadding]').remove_attr('cellpadding')
54
55 css('.t-dsc-hitem > td', '.t-dsc-header > td').each do |node|
56 node.name = 'th'
57 node.content = ' ' if node.content.empty?
58 end
59
60 css('tt', 'span > span.source-cpp', 'span.t-c', 'span.t-lc', 'span.t-dsc-see-tt').each do |node|
61 node.name = 'code'
62 node.remove_attribute('class')
63 node.content = node.content unless node.at_css('a')
64 end
65
66 css('div > span.source-cpp').each do |node|
67 node.name = 'pre'
68 node.inner_html = node.inner_html.gsub('<br>', "\n")
69 node.content = node.content
70 end
71
72 css('div > a > img[alt="About this image"]').each do |node|
73 node.parent.parent.remove
74 end
75
76 css('area[href]').each do |node|
77 node['href'] = node['href'].remove('.html')
78 end
79
80 css('p').each do |node|
81 while node.next && (node.next.text? || node.next.name == 'a' || node.next.name == 'code')
82 node << node.next
83 end
84 node.inner_html = node.inner_html.strip
85 node << '.' if node.content =~ /[a-zA-Z0-9\)]\z/
86 node.remove if node.content.blank? && !node.at_css('img')
87 end
88
89 css('pre').each do |node|
90 node['data-language'] = if node['class'].try(:include?, 'cpp') || node.parent['class'].try(:include?, 'cpp')
91 'cpp'
92 else
93 'c'
94 end
95 node.remove_attribute('class')
96 node.content = node.content.gsub("\t", ' ' * 8)
97 end
98
99 css('code code', '.mw-geshi').each do |node|
100 node.before(node.children).remove
101 end
102
103 css('h1 ~ .fmbox').each do |node|
104 node.name = 'div'
105 node.content = node.content
106 end
107
108 css('img').each do |node|
109 node['src'] = node['src'].sub! %r{http://en.cppreference.com/common/([^"']+?)\.svg}, 'http://upload.cppreference.com/mwiki/\1.svg'
110 end
111
112 doc
113 end
114 end
115 end
116end
lib/docs/filters/c/entries.rb+3−0
@@ -22,6 +22,9 @@ module Docs · 21 unmodified lines
22 end22 end
23 23
24 def get_type24 def get_type
25
26 return "C keywords" if slug =~ /keyword/
27
25 type = at_css('.t-navbar > div:nth-child(4) > :first-child').try(:content)28 type = at_css('.t-navbar > div:nth-child(4) > :first-child').try(:content)
26 type.strip!29 type.strip!
27 type.remove! ' library'30 type.remove! ' library'
lib/docs/filters/c/fix_code.rb+0−21
@@ -1,21 +0,0 @@
1module Docs
2 class C
3 class FixCodeFilter < Filter
4 def call
5 css('div > span.source-c', 'div > span.source-cpp').each do |node|
6 node.inner_html = node.inner_html.gsub(/<br>\n?/, "\n").gsub("\n</p>\n", "</p>\n")
7 node.parent.name = 'pre'
8 node.parent['class'] = node['class']
9 node.parent.content = node.content
10 end
11
12 nbsp = Nokogiri::HTML('&nbsp;').text
13 css('pre').each do |node|
14 node.content = node.content.gsub(nbsp, ' ')
15 end
16
17 doc
18 end
19 end
20 end
21end
lib/docs/filters/c/fix_urls.rb+0−11
@@ -1,11 +0,0 @@
1module Docs
2 class C
3 class FixUrlsFilter < Filter
4 def call
5 html.gsub! File.join(C.base_url, C.root_path), C.base_url[0..-2]
6 html.gsub! %r{#{C.base_url}([^"']+?)\.html}, "#{C.base_url}\\1"
7 html
8 end
9 end
10 end
11end
lib/docs/scrapers/c.rb+0−42
@@ -1,42 +0,0 @@
1module Docs
2 class C < FileScraper
3 self.type = 'c'
4 self.base_url = 'http://en.cppreference.com/w/c/'
5 self.root_path = 'header.html'
6
7 html_filters.insert_before 'clean_html', 'c/fix_code'
8 html_filters.push 'c/entries', 'c/clean_html', 'title'
9 text_filters.push 'c/fix_urls'
10
11 options[:decode_and_clean_paths] = true
12 options[:container] = '#content'
13 options[:title] = false
14 options[:root_title] = 'C Programming Language'
15 options[:skip] = %w(language/history.html)
16 options[:skip_patterns] = [/experimental/]
17
18 options[:fix_urls] = ->(url) do
19 url.sub! %r{\A.+/http%3A/}, 'http://'
20 url.sub! 'http://en.cppreference.com/upload.cppreference.com', 'http://upload.cppreference.com'
21 url
22 end
23
24 options[:attribution] = <<-HTML
25 &copy; cppreference.com<br>
26 Licensed under the Creative Commons Attribution-ShareAlike Unported License v3.0.
27 HTML
28
29 def get_latest_version(opts)
30 doc = fetch_doc('https://en.cppreference.com/w/Cppreference:Archives', opts)
31 link = doc.at_css('a[title^="File:"]')
32 date = link.content.scan(/(\d+)\./)[0][0]
33 DateTime.strptime(date, '%Y%m%d').to_time.to_i
34 end
35
36 private
37
38 def file_path_for(*)
39 URI.unescape(super)
40 end
41 end
42end
lib/docs/scrapers/cppref/c.rb+12−0
@@ -0,0 +1,12 @@
1module Docs
2 class C < Cppref
3 self.name = 'c'
4 self.slug = 'c'
5 self.base_url = 'https://en.cppreference.com/w/c/'
6
7 html_filters.insert_before 'cppref/clean_html', 'c/entries'
8
9 options[:root_title] = 'C Programming Language'
10
11 end
12end
lib/docs/scrapers/cppref/cpp.rb+0−9
@@ -2,7 +2,6 @@ module Docs · 1 unmodified line
2 class Cpp < Cppref2 class Cpp < Cppref
3 self.name = 'C++'3 self.name = 'C++'
4 self.slug = 'cpp'4 self.slug = 'cpp'
5 self.type = 'c'
6 self.base_url = 'https://en.cppreference.com/w/cpp/'5 self.base_url = 'https://en.cppreference.com/w/cpp/'
7 6
8 html_filters.insert_before 'cppref/clean_html', 'cpp/entries'7 html_filters.insert_before 'cppref/clean_html', 'cpp/entries'
@@ -16,13 +15,5 @@ module Docs · 7 unmodified lines
16 regex/regex_token_iterator/operator_cmp.html15 regex/regex_token_iterator/operator_cmp.html
17 )16 )
18 17
19 # Same as get_latest_version in lib/docs/scrapers/c.rb
20 def get_latest_version(opts)
21 doc = fetch_doc('https://en.cppreference.com/w/Cppreference:Archives', opts)
22 link = doc.at_css('a[title^="File:"]')
23 date = link.content.scan(/(\d+)\./)[0][0]
24 DateTime.strptime(date, '%Y%m%d').to_time.to_i
25 end
26
27 end18 end
28end19end
lib/docs/scrapers/cppref/cppref.rb+8−4
@@ -6,7 +6,7 @@ module Docs · 5 unmodified lines
6 6
7 html_filters.insert_before 'clean_html', 'cppref/fix_code'7 html_filters.insert_before 'clean_html', 'cppref/fix_code'
8 html_filters.push 'cppref/clean_html', 'title'8 html_filters.push 'cppref/clean_html', 'title'
9 # 'cpp20/entries',9
10 options[:decode_and_clean_paths] = true10 options[:decode_and_clean_paths] = true
11 options[:container] = '#content'11 options[:container] = '#content'
12 options[:title] = false12 options[:title] = false
@@ -21,9 +21,13 @@ module Docs · 8 unmodified lines
21 Licensed under the Creative Commons Attribution-ShareAlike Unported License v3.0.21 Licensed under the Creative Commons Attribution-ShareAlike Unported License v3.0.
22 HTML22 HTML
23 23
24 # def get_latest_version24 # Check if the 'headers' page has changed
25 25 def get_latest_version(opts)
26 # end26 doc = fetch_doc(self.base_url + self.root_path, opts)
27 date = doc.at_css('#footer-info-lastmod').content
28 date = date.match(/[[:digit:]]{1,2} .* [[:digit:]]{4}/).to_s
29 date = DateTime.strptime(date, '%e %B %Y').to_time.to_i
30 end
27 31
28 end32 end
29end33end