From c214b28f88c1b1cbce37d87c6e0276a122841712 Mon Sep 17 00:00:00 2001
From: Jess Hottenstein <jess@splitwise.com>
Date: Wed, 8 May 2024 16:11:35 -0400
Subject: [PATCH 1/2] Add new scrubber suitable for strip tags but adding
whitespace
---
lib/rails/html/scrubbers.rb | 28 ++++++++++++++++++++++++++++
test/scrubbers_test.rb | 14 ++++++++++++++
2 files changed, 42 insertions(+)
diff --git a/lib/rails/html/scrubbers.rb b/lib/rails/html/scrubbers.rb
index a1317ad..7afd7c0 100644
--- a/lib/rails/html/scrubbers.rb
+++ b/lib/rails/html/scrubbers.rb
@@ -195,5 +195,33 @@ def scrub(node)
end
end
end
+
+ # === Rails::HTML::SpaceBlockElementScrubber
+ #
+ # +Rails::HTML::SpaceBlockElementScrubber+ removes all undisplayable tags and contents.
+ #
+ # It preserves text contents and inserts a space around block level elements.
+ class SpaceBlockElementScrubber < Loofah::Scrubber # :nodoc:
+ def initialize
+ @direction = :bottom_up
+ end
+
+ def scrub(node)
+ if Loofah::Elements::LINEBREAKERS.include?(node.name)
+ replacement = if Loofah::Elements::INLINE_LINE_BREAK.include?(node.name)
+ " "
+ else
+ " #{node.content} "
+ end
+ node.add_next_sibling(Nokogiri::XML::Text.new(replacement, node.document))
+ node.remove
+ elsif node.text?
+ return CONTINUE
+ else
+ node.before node.children
+ node.remove
+ end
+ end
+ end
end
end
diff --git a/test/scrubbers_test.rb b/test/scrubbers_test.rb
index 8db2d85..ac979b6 100644
--- a/test/scrubbers_test.rb
+++ b/test/scrubbers_test.rb
@@ -200,6 +200,20 @@ def test_skips_text_nodes
end
end
+class SpaceBlockElementScrubberTest < ScrubberTest
+ def setup
+ @scrubber = Rails::HTML::SpaceBlockElementScrubber.new
+ end
+
+ def test_removes_all_tags_and_keeps_the_text_content_and_adds_spaces
+ assert_scrubbed %(<div><h1>A list!</h1><ul><li>An element!</li></ul><script>alert("hi!")</script></div>), " A list! An element! "
+ end
+
+ def test_skips_text_nodes
+ assert_node_skipped("some text")
+ end
+end
+
class ReturningStopFromScrubNodeTest < ScrubberTest
class ScrubStopper < Rails::HTML::PermitScrubber
def scrub_node(node)
From bd5ea3a5f6c6517fa664779456c485c816547478 Mon Sep 17 00:00:00 2001
From: Mike Dalessio <mike.dalessio@gmail.com>
Date: Sun, 19 May 2024 17:26:22 -0400
Subject: [PATCH 2/2] style: rubocop autocorrects
---
lib/rails/html/scrubbers.rb | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/lib/rails/html/scrubbers.rb b/lib/rails/html/scrubbers.rb
index 7afd7c0..e074e65 100644
--- a/lib/rails/html/scrubbers.rb
+++ b/lib/rails/html/scrubbers.rb
@@ -209,14 +209,14 @@ def initialize
def scrub(node)
if Loofah::Elements::LINEBREAKERS.include?(node.name)
replacement = if Loofah::Elements::INLINE_LINE_BREAK.include?(node.name)
- " "
- else
- " #{node.content} "
- end
+ " "
+ else
+ " #{node.content} "
+ end
node.add_next_sibling(Nokogiri::XML::Text.new(replacement, node.document))
node.remove
elsif node.text?
- return CONTINUE
+ CONTINUE
else
node.before node.children
node.remove