Merge branch 'autolink-filter-text-parse' into 'master'

CHANGELOG

@@ -16,6 +16,7 @@ v 8.11.0 (unreleased)
   - Add support for using RequestStore within Sidekiq tasks via SIDEKIQ_REQUEST_STORE env variable
   - Optimize maximum user access level lookup in loading of notes
   - Add "No one can push" as an option for protected branches. !5081
+  - Improve performance of AutolinkFilter#text_parse by using XPath
   - Environments have an url to link to
   - Remove unused images (ClemMakesApps)
   - Limit git rev-list output count to one in forced push check

lib/banzai/filter/autolink_filter.rb

@@ -31,6 +31,14 @@ module Banzai
       # Text matching LINK_PATTERN inside these elements will not be linked
       IGNORE_PARENTS = %w(a code kbd pre script style).to_set
 
+      # The XPath query to use for finding text nodes to parse.
+      TEXT_QUERY = %Q(descendant-or-self::text()[
+        not(#{IGNORE_PARENTS.map { |p| "ancestor::#{p}" }.join(' or ')})
+        and contains(., '://')
+        and not(starts-with(., 'http'))
+        and not(starts-with(., 'ftp'))
+      ])
+
       def call
         return doc if context[:autolink] == false
 
@@ -66,16 +74,11 @@ module Banzai
       # Autolinks any text matching LINK_PATTERN that Rinku didn't already
       # replace
       def text_parse
-        search_text_nodes(doc).each do |node|
+        doc.xpath(TEXT_QUERY).each do |node|
           content = node.to_html
 
-          next if has_ancestor?(node, IGNORE_PARENTS)
           next unless content.match(LINK_PATTERN)
 
-          # If Rinku didn't link this, there's probably a good reason, so we'll
-          # skip it too
-          next if content.start_with?(*%w(http https ftp))
-
           html = autolink_filter(content)
 
           next if html == content