trio: finish scraper and filters

Author: jmerle

.gitignore

@@ -7,4 +7,3 @@ public/fonts
 public/docs/**/*
 docs/**/*
 !docs/*.md
-vendor

assets/javascripts/templates/pages/about_tmpl.coffee

@@ -721,6 +721,11 @@ credits = [
     '2018 HashiCorp',
     'MPL',
     'https://raw.githubusercontent.com/hashicorp/terraform-website/master/LICENSE.md'
+  ], [
+    'Trio',
+    '2017 Nathaniel J. Smith',
+    'MIT',
+    'https://raw.githubusercontent.com/python-trio/trio/master/LICENSE.MIT'
   ], [
     'Twig',
     '2009-2018 The Twig Team',

lib/docs/filters/trio/clean_html.rb

@@ -3,6 +3,7 @@ class Trio
     class CleanHtmlFilter < Filter
       def call
         @doc = at_css('div[role="main"]')
+
         css('.section, [itemprop=articleBody]').each do |node|
           node.replace node.children
         end
@@ -11,40 +12,43 @@ def call
 
         css('dt').each do |node|
           node.name = 'h3'
+
           if node.parent.classes.include? 'field-list'
             node.name = 'h4'
             node['style'] = 'margin: 0'
+
             if node.text == 'Parameters' or node.text == 'Raises'
               node.next_element.css('strong').each do |n|
                 n.name = 'code'
               end
             end
           else
             code = doc.document.create_element 'code'
+
             if em = node.at_css('.property')
               code.inner_html = "<em>#{em.text.strip}</em> "
               em.remove
             end
+
             code.inner_html += node.inner_text.strip
             node.inner_html = code
           end
         end
 
         css('pre').each do |node|
+          node.content = node.content.strip
+
           classes = node.parent.parent.classes
           if classes.include? 'highlight-python3'
-            node['class'] = 'language-python'
             node['data-language'] = 'python'
           end
+
           node.parent.parent.replace(node)
         end
 
         css('.admonition').each do |node|
           node.name = 'blockquote'
           node.at_css('.admonition-title').name = 'h4'
-          # new_node = node.document.create_element 'blockquote'
-          # new_node.inner_html = node.inner_html
-          # node.replace new_node
         end
 
         doc

lib/docs/filters/trio/entries.rb

@@ -11,7 +11,6 @@ def get_type
 
       def additional_entries
         css('.descname').each_with_object [] do |node, entries|
-
           name = node.text
           if node.previous.classes.include? 'descclassname'
             name = node.previous.text + name
@@ -25,16 +24,19 @@ def additional_entries
               or dl.classes.include?('data')
             parent = dl.parent.previous_element
             cls = ''
+
             if n = parent.at_css('.descclassname')
               cls += n.text
             end
+
             if n = parent.at_css('.descname')
               if n.text == "The nursery interface"
                 cls += "Nursery."
               else
                 cls += n.text + '.'
               end
             end
+
             name = cls + name
           end
 

lib/docs/scrapers/trio.rb

@@ -1,8 +1,8 @@
 module Docs
   class Trio < UrlScraper
     self.type = 'simple'
-    self.release = '0.11'
-    self.base_url = 'https://trio.readthedocs.io/en/latest/'
+    self.release = '0.12.1'
+    self.base_url = 'https://trio.readthedocs.io/en/v0.12.1/'
     self.root_path = 'index.html'
     self.links = {
       home: 'https://trio.readthedocs.io/',
@@ -19,9 +19,13 @@ class Trio < UrlScraper
     ]
 
     options[:attribution] = <<-HTML
-      &copy; 2017-2019 Nathaniel J. Smith<br>
-      Licensed under MIT and Apache2.
+      &copy; 2017 Nathaniel J. Smith<br>
+      Licensed under the MIT License.
     HTML
 
+    def get_latest_version(opts)
+      doc = fetch_doc('https://trio.readthedocs.io/en/stable/', opts)
+      doc.at_css('.rst-other-versions a[href^="/en/v"]').content[1..-1]
+    end
   end
 end