Finish LOVE scraper

assets/javascripts/news.json

@@ -1,10 +1,13 @@
 [
   [
+    "2016-06-19",
+    "New documentation: <a href=\"/love/\">L&Ouml;VE</a>"
+  ], [
     "2016-06-12",
     "New documentation: <a href=\"/angular/\">Angular 2</a>"
   ], [
     "2016-06-05",
-    "New documentation: <a href=\"/kotlin/\">Kotlin</a> and <a href=\"/padrino/\">Padrino</a>"
+    "New documentations: <a href=\"/kotlin/\">Kotlin</a> and <a href=\"/padrino/\">Padrino</a>"
   ], [
     "2016-04-24",
     "New documentations: <a href=\"/numpy/\">NumPy</a> and <a href=\"/apache_pig/\">Apache Pig</a>"

assets/javascripts/templates/pages/about_tmpl.coffee

@@ -275,6 +275,11 @@ credits = [
     '1994–2015 Lua.org, PUC-Rio',
     'MIT',
     'http://www.lua.org/license.html'
+  ], [
+    'LÖVE',
+    '2006-2016 LÖVE Development Team',
+    'GFDL',
+    'http://www.gnu.org/copyleft/fdl.html'
   ], [
     'Marionette.js',
     '2016 Muted Solutions, LLC',

assets/stylesheets/global/_icons.scss

@@ -142,3 +142,4 @@
 ._icon-kotlin:before        { background-position: -6rem -11rem; }
 ._icon-padrino:before       { background-position: -7rem -11rem; }
 ._icon-angular:before       { background-position: -8rem -11rem; }
+._icon-love:before          { background-position: -9rem -11rem; }

assets/stylesheets/pages/_love.scss

@@ -1,67 +1,17 @@
 ._love {
-  padding-left: 1rem;
+  @extend %simple;
 
-  h1, h2 { margin-left: -1rem; }
-  h2 { @extend %block-heading; }
-  h3 { margin-left: -0.5rem; @extend %block-label; }
-
-  ._mobile & {
-    padding-left: 0;
-
-    h1, h2, h3 { margin-left: 0; }
-  }
-
-  p > code, li > code { @extend %label; }
-  blockquote { @extend %note; }
-
-  .box { @extend %box; }
   .note { @extend %note; }
-  .label { @extend %label; }
+  .note-green { @extend %note-green; }
+  .note-red { @extend %note-red; }
 
-  .box-heading {
-    @extend %heading-box;
-    padding: .5em .75em;
-    margin-top: 1.5rem;
-    margin-bottom: 0px;
-    border-bottom: none;
-    border-bottom-left-radius: 0px;
-    border-bottom-right-radius: 0px;
-  }
-  .box-with-heading {
-    @extend %box;
-    padding: .5em .75em;
-    margin-top: 0px;
-    margin-bottom: 1.5rem;
-    border-top-left-radius: 0px;
-    border-top-right-radius: 0px;
-  }
+  .label, dt > code { @extend %label; }
+  .label-green { @extend %label-green; }
+  .label-red { @extend %label-red; }
+
+  .smwtable { width: 100%; }
+  .smwtable td:nth-last-child(2), .smwtable td:last-child { width: 2.5em; }
 
-  .note-green, .label-green { @extend %note-green; }
-  .note-red, .label-red { @extend %note-red; }
-  .note-orange, .label-orange { @extend %note-orange; }
   .cell-green { background: $noteGreenBackground; }
   .cell-red { background: $noteRedBackground; }
-
-  .smwtable {
-    width: 100%;
-    tr {
-      td {
-        word-wrap: break-word;
-      }
-
-      td:first-child, td:nth-last-child(2), td:last-child {
-        vertical-align: middle;
-        white-space: nowrap;
-        width: 1em;
-        overflow: hidden;
-      }
-    }
-  }
-
-  hr {
-    border: none;
-    height: 1px;
-    background-color: $textColorLighter;
-    margin: 1.5em 0 1em;
-  }
 }

lib/docs/filters/core/internal_urls.rb

@@ -39,6 +39,7 @@ module Docs
     def to_internal_url(str)
       return unless (url = parse_url(str)) && (subpath = subpath_to(url))
       normalize_subpath(subpath)
+      subpath = URI.unescape(subpath) if context[:decode_and_clean_paths]
       return if skip_subpath?(subpath)
       normalize_url(url, subpath)
       url

lib/docs/filters/love/clean_html.rb

@@ -2,100 +2,67 @@ module Docs
   class Love
     class CleanHtmlFilter < Filter
       def call
-        # Fix syntax highlighting
+        @doc = at_css('#mw-content-text')
+
         css('.mw-code').each do |node|
-          node.content = node.at_css("div > pre").content
+          node.content = node.at_css('div > pre').content
           node['data-language'] = 'lua'
           node.name = 'pre'
         end
 
-        # Move header tags up
-        css('h2', 'h3').each do |node|
-          headline = node.at_css('.mw-headline')
-          node['id'] = headline['id']
-          node.content = headline.inner_text
+        css('span[id]').each do |node|
+          node.parent['id'] = node['id']
+          node.before(node.children).remove
         end
 
-        # Move dt tags up
-        css('dt > span').each do |node|
-          node.parent.content = node.inner_text
+        css('table.notice').each do |node|
+          content = node.at_css('td:nth-child(2)').inner_html
+          node.replace %(<p class="note">#{content}</p>)
         end
 
-        # Style notices and new/removed sections
-        css('.notice', '.new-section', '.removed-section', '.removed-new-section').each do |node|
-          case node['class']
-          when 'notice'
-            node['class'] = 'note note-warning'
-            node.inner_html = node.at_css('td:nth-child(2)').inner_html
-            node.next.remove unless node.next.nil? or node.next.name != 'br'
-          when 'new-section', 'removed-section', 'removed-new-section'
-            node['class'] = node['class'] == 'new-section' ? 'note note-green' : 'note note-red'
-            node.inner_html = node.at_css('tr > td > i').inner_html \
-              + '<br>' \
-              + node.at_css('tr > td > small').inner_html
-          end
-
-          node.name = 'p'
-          node.remove_attribute('bgcolor')
-          node.remove_attribute('style')
-          node.remove_attribute('align')
+        css('table.new-section', 'table.removed-section', 'table.removed-new-section').each do |node|
+          klass = node['class'] == 'new-section' ? 'note-green' : 'note-red'
+          content = node.css('td').map(&:inner_html).join('<br>')
+          node.replace %(<p class="note #{klass}">#{content}</p>)
         end
 
-        # Style new/removed features
         css('.new-feature', '.removed-feature', '.removed-new-feature').each do |node|
-          node.name = 'div'
-          node['class'] = node['class'] == 'new-feature' ? 'box-heading label-green' : 'box-heading label-red'
-          node.remove_attribute('style')
+          klass = node['class'] == 'new-feature' ? 'label-green' : 'label-red'
+          content = node.content.sub(' LÖVE', '')
+          label = %( <span class="label #{klass}">#{content}</span>)
 
-          container = node.next_element
-          container.name = 'div'
-          container['class'] = 'box-with-heading'
-          container.remove_attribute('style')
+          node.next_element.css('dt').each { |n| n << label }
+          node.remove
         end
 
-        # Style tables
-        css('table.smwtable').each do |table|
-          table.remove_attribute('style')
-          table.css('td').each do |cell|
-            cell.remove_attribute('style')
-          end
-          table.css('td:last-child', 'td:nth-last-child(2)').each do |cell|
-            img = cell.at_css('img')
-            if img then
-              if img['alt'] == 'Added since' then
-                cell['class'] = 'cell-green'
-              elsif img['alt'] == 'Removed in'
-                cell['class'] = 'cell-red'
-              end
-              img.remove
-            end
-          end
+        css('img[src$="Add.png"]').each do |node|
+          node.parent['class'] = 'cell-green'
+          node.remove
         end
 
-        # Remove Other Languages
-        css('#Other_Languages').remove
-        css('.i18n').remove
+        css('img[src$="Remove.png"]').each do |node|
+          node.parent['class'] = 'cell-red'
+          node.remove
+        end
 
-        # Remove changelog
-        node = at_css('h2#Changelog')
-        if !node.nil? then
-          begin
-            nxt = node.next
-            node.remove
-            node = nxt
-          end while !node.nil? and node.name != 'h2'
+        css('table, tr, td, th').each do |node|
+          %w(style cellpadding cellspacing width height valign).each do |attribute|
+            node.remove_attribute(attribute)
+          end
         end
 
-        # Remove empty paragraphs
-        css('p').each do |node|
-          node.remove if node.inner_text.strip == ''
+        css('.note i', '.note small', 'div:not([class])', '.smwtable td:nth-last-child(2) > a', '.smwtable td:last-child > a').each do |node|
+          node.before(node.children).remove
         end
 
-        # Remove linebreaks that are the first or last child of a paragraph
-        css('p > br:first-child', 'p > br:last-child').each do |node|
-          node.remove
+        css('p > br').each do |node|
+          node.parent.remove if node.parent.content.empty?
         end
 
+        css('div > br', '> br', 'hr').remove
+        css('#Editing_the_wiki + p', '#Editing_the_wiki').remove
+        css('#Other_Languages', '.i18n').remove
+
         doc
       end
     end

lib/docs/filters/love/entries.rb

@@ -1,21 +1,55 @@
 module Docs
   class Love
     class EntriesFilter < Docs::EntriesFilter
+      TYPES = {
+        'require'                 => 'Lua',
+        'light_userdata'          => 'Lua',
+        'value'                   => 'Lua',
+        'variable'                => 'Lua',
+
+        'Audio_Formats'           => 'love.sound',
+        'ImageFontFormat'         => 'love.font',
+        'BlendMode_Formulas'      => 'love.graphics',
+        'Shader_Variables'        => 'love.graphics',
+
+        'AreaSpreadDistribution'  => 'love.graphics',
+        'BodyType'                => 'love.physics',
+        'BufferMode'              => 'love.filesystem',
+        'CompressedFormat'        => 'love.image',
+        'JoystickConstant'        => 'love.joystick',
+        'ParticleInsertMode'      => 'love.graphics',
+        'String'                  => 'love',
+        'TextureMode'             => 'love.graphics'
+      }
+
+      def call
+        if context[:initial_paths].include?(slug)
+          css('table.smwtable td:first-child > a').each do |node|
+            TYPES[node.content.strip] = slug
+          end
+        end
+
+        super
+      end
+
       def get_type
-        if TYPE_OVERRIDE.key?(slug) then
-          return TYPE_OVERRIDE[slug]
-        elsif m = slug.match(/\A(love\.\w+)\z/) then
-          # modules and funcions
-          return LOVE_MODULES.include?(m[1]) ? m[1] : 'love'
-        elsif m = slug.match(/\A(love\.\w+)\.(\w+)/) then
-          # functions in modules
-          return m[1]
-        elsif context[:list_classes] and (m = slug.match(/\A\(?([A-Z]\w+)\)?(\:\w+)?/)) then
-          # classes, their members and enums
-          return m[1] unless m[1].include?('_')
+        if slug == 'love'
+          'love'
+        elsif slug.start_with?('enet')
+          'enet'
+        elsif slug.include?('Joint') || slug.include?('Shape')
+          'love.physics'
+        elsif TYPES.key?(slug)
+          TYPES[slug]
+        elsif match = slug.match(/\A(love\.\w+)(\.\w+)?\z/)
+          match[2] || context[:initial_paths].include?(match[1]) ? match[1] : 'love'
+        elsif at_css('#catlinks a[title="Category:Lua"]')
+          'Lua'
+        elsif
+          type = slug.split(':').first
+          type.remove! %r{[\(\)]}
+          TYPES[type]
         end
-        # usually this shouldn't happen
-        "Other"
       end
     end
   end

lib/docs/scrapers/love.rb

@@ -1,98 +1,42 @@
 module Docs
   class Love < UrlScraper
-    LOVE_MODULES = %w(
-      love
-      love.audio
-      love.event
-      love.filesystem
-      love.font
-      love.graphics
-      love.image
-      love.joystick
-      love.keyboard
-      love.math
-      love.mouse
-      love.physics
-      love.sound
-      love.system
-      love.thread
-      love.timer
-      love.touch
-      love.video
-      love.window
-    )
-    TYPE_OVERRIDE = {
-      "Audio_Formats" => "love.sound",
-      "ImageFontFormat" => "love.font",
-      "BlendMode_Formulas" => "BlendMode",
-      "Shader_Variables" => "Shader"
-    }
-
     self.name = 'LÖVE'
     self.slug = 'love'
     self.type = 'love'
+    self.release = '0.10.1'
     self.base_url = 'https://love2d.org/wiki/'
-    self.root_path = 'love'
-    self.initial_paths = LOVE_MODULES
+    self.root_path = 'Main_Page'
+    self.initial_paths = %w(love love.audio love.event love.filesystem love.font love.graphics
+      love.image love.joystick love.keyboard love.math love.mouse love.physics love.sound
+      love.system love.thread love.timer love.touch love.video love.window enet socket utf8)
     self.links = {
       home: 'https://love2d.org/',
       code: 'https://bitbucket.org/rude/love'
     }
 
-    html_filters.push 'love/clean_html', 'love/entries', 'title'
-
-    options[:root_title] = 'love'
-    options[:initial_paths] = LOVE_MODULES
+    html_filters.push 'love/entries', 'love/clean_html', 'title'
 
+    options[:root_title] = 'LÖVE'
     options[:decode_and_clean_paths] = true
+    options[:container] = '#bodyContent'
 
-    # Add types to classes and their members
-    options[:list_classes] = true
+    options[:skip] = %w(Getting_Started Building_LÖVE Tutorial Tutorials Game_Distribution License
+      Games Libraries Software Snippets Version_History Lovers PO2_Syndrome HSL_color Guidelines)
 
-    options[:container] = '#mw-content-text'
-
-    options[:only_patterns] = [
-      /\A(love\z|love\.|[A-Z]|\([A-Z])/
-      # love
-      # love.* (modules and functions)
-      # Uppercased (classes and enums)
-      # (Uppercased) (generalized classes)
-    ]
-    options[:skip] = %w(
-      Getting_Started
-      Building_LÖVE
-      Tutorial
-      Tutorials
-      Game_Distribution
-      License
-      Games
-      Libraries
-      Software
-      Snippets
-      Version_History
-      Lovers
-      PO2_Syndrome
-      HSL_color
-    )
     options[:skip_patterns] = [
-      /_\([^\)]+\)\z/,
-      # anything_(language) (this might have to be tweaked)
-      /\ASpecial:/,
-      /\ACategory:/,
-      /\AFile:/,
-      /\AHelp:/,
-      /\ATemplate:/,
-      /\AUser:/,
-      /\ATutorial:/
-      # special pages are indistinguishable from instance methods
+      /_\([^\)]+\)\z/, # anything_(language)
+      /\A(Special|Category|File|Help|Template|User|Tutorial):/,
+      /\A\d/
     ]
 
     options[:replace_paths] = {
-      "Config_Files" => "love.conf"
+      'Config_Files' => 'love.conf',
+      'conf.lua' => 'love.conf',
+      'lua-enet' => 'enet'
     }
 
     options[:attribution] = <<-HTML
-      &copy; L&Ouml;VE Development Team<br>
+      &copy; 2006&ndash;2016 L&Ouml;VE Development Team<br>
       Licensed under the GNU Free Documentation License, Version 1.3.
     HTML
   end