Finish webpack scraper

Thibaut committed 11 years ago

Commit f0cb4b61 parent ba165ae15 files · +25 −89Browse files

Showing the changes to lib/docs (3 of 15 files). Show all

Unified · Split

3 files

 module Docs
   class Webpack
     class CleanHtmlFilter < Filter
       def call
         root_page? ? root : other
         # in devdocs.
         hr_index = doc.children.find_index { |node| node.name == "hr" }
         doc.children[0..hr_index].each(&:remove)
-        # Add a page header :)
-        page_header_node = Nokogiri::XML::Node.new "h1", @doc
-        page_header_node.content = "Webpack"
-        @doc.children.first.add_previous_sibling page_header_node
       end
       def other
         # Re-create the header element
-        page_header = at_css("#wikititle").content.upcase
+        at_css("#wiki").child.before("<h1>#{at_css("#wikititle").content}</h1>")
-        page_header_node = Nokogiri::XML::Node.new "h1", @doc
-        page_header_node.content = page_header
-        # Change the scope of the page
         @doc = at_css("#wiki")
-        # Add the page header
+        css('.contents', 'a.anchor', 'hr').remove
-        @doc.children.first.add_previous_sibling page_header_node
-        # Remove the sidebar links in each page
-        css(".contents").remove
-        # Remove all anchors
-        css("a.anchor").remove
+        css('pre').each do |node|
+          node.content = node.content
+        end
       end
     end
   end

   class Webpack
     class EntriesFilter < Docs::EntriesFilter
       def get_name
-        entry_link.content.strip.titleize
+        entry_link.content
       end
       def get_type
         type_links_list = link_li.parent
         current_type = type_links_list.parent
         # current type is a
         # <li>
         #   TYPE
         #   <li> <ul> .. links .. </ul> </li>
         current_type.children.first.content.strip.titleize
       end
-      def additional_entries
-        return [] unless root_page?
-        css('.sidebar a').each_with_object [] do |link, entries|
-          name = link.content.strip
-          entries << [name, link['href']]
-        end
-      end
       private
       def entry_link

 module Docs
   class Webpack < UrlScraper
-    self.name = 'Webpack'
+    self.name = 'webpack'
     self.type = 'webpack'
-    self.version = '1.9'
+    self.version = '1.10'
     self.base_url = 'https://webpack.github.io/docs/'
-    self.root_path = 'index.html'
+    self.links = {
+      home: 'https://webpack.github.io/',
+      code: 'https://github.com/webpack/webpack'
+    }
-    html_filters.push 'webpack/entries', 'webpack/clean_html'
+    html_filters.push 'webpack/entries', 'webpack/clean_html', 'title'
+    options[:title] = false
+    options[:root_title] = 'webpack'
+    options[:skip] = %w(list-of-tutorials.html examples.html changelog.html ideas.html roadmap.html)
     options[:attribution] = <<-HTML
       &copy; 2012&ndash;2015 Tobias Koppers<br>