Implementation of CouchDB docs.

Author: spamguy

lib/docs/filters/couchdb/clean_html.rb

@@ -0,0 +1,30 @@
+module Docs
+  class Couchdb
+    class CleanHtmlFilter < Filter
+      def call
+        css('h1').each do |node|
+          node.content = node.content.gsub(/\d*\. |\P{ASCII}/, '').split('.').last
+        end
+
+        css('h2', 'h3').each do |node|
+          node.content = node.content.gsub(/\P{ASCII}/, '').split('.').last
+        end
+
+        css('pre').each do |node|
+          node.content = node.content.strip
+
+          classes = node.parent.parent.classes
+          if classes.include? 'highlight-bash'
+            node['data-language'] = 'bash'
+          else
+            node['data-language'] = 'javascript'
+          end
+
+          node.parent.parent.replace(node)
+        end
+
+        doc
+      end
+    end
+  end
+end

lib/docs/filters/couchdb/entries.rb

@@ -0,0 +1,53 @@
+module Docs
+  class Couchdb
+    class EntriesFilter < Docs::EntriesFilter
+      SLUG_MAP = {
+        'api' => 'API',
+        'json' => 'JSON Structures',
+        'cluster' => 'Cluster Management',
+        'replication' => 'Replication',
+        'maintenance' => 'Maintenance',
+        'partitioned' => 'Partitioned Databases'
+      }
+
+      def get_name
+        at_css('h1').content.gsub(/\P{ASCII}/, '').split('.').last
+      end
+
+      def get_type
+        if slug.start_with?('ddocs/views')
+          'Views'
+        elsif slug.start_with?('ddocs')
+          'Design Documents'
+        else
+          SLUG_MAP[slug[/^(.+?)[-\/]/, 1]] || name
+        end
+      end
+
+      def additional_entries
+        needs_breakup = [
+          'JSON Structure Reference',
+          'Design Documents',
+          'Partitioned Databases'
+        ]
+
+        if needs_breakup.include?(name)
+          entries = []
+
+          css('section > section').each do |node|
+            h2 = node.at_css('h2')
+
+            if h2.present?
+              name = node.at_css('h2').content.split('.').last
+              entries << [name, node['id']]
+            end
+          end
+
+          entries
+        else
+          []
+        end
+      end
+    end
+  end
+end

lib/docs/scrapers/couchdb.rb

@@ -0,0 +1,39 @@
+module Docs
+  class Couchdb < UrlScraper
+    self.name = 'CouchDB'
+    self.type = 'couchdb'
+    self.root_path = 'index.html'
+    
+    self.links = {
+      home: 'https://couchdb.apache.org/',
+      code: 'https://github.com/apache/couchdb'
+    }
+
+    html_filters.push 'couchdb/clean_html', 'couchdb/entries'
+
+    options[:container] = 'div[itemprop=articleBody]'
+    options[:only_patterns] = [
+      /api\//,
+      /cluster\//,
+      /ddocs\//,
+      /replication\//,
+      /maintenance\//,
+      /partitioned-dbs\//,
+      /json\-structure*/
+    ]
+    options[:rate_limit] = 50 # Docs are subject to Cloudflare limiters.
+    options[:attribution] = <<-HTML
+      Copyright &copy; 2025 The Apache Software Foundation &mdash; Licensed under the Apache License 2.0
+    HTML
+
+    version '3.5' do
+      self.release = '3.5.1'
+      self.base_url = "https://docs.couchdb.org/en/#{self.release}"
+    end
+
+    def get_latest_version(opts)
+      doc = fetch_doc('https://couchdb.apache.org/', opts)
+      doc.at_css('.download-pane > h2').content.split(' ').last
+    end
+  end
+end

public/icons/docs/couchdb/16.png

@@ -0,0 +1 @@
+https://docs.couchdb.org/en/stable/_static/favicon.ico