Release version 0.1.2 with new data-markdown-ignore attribute to ex…

CHANGELOG.md

@@ -7,6 +7,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [0.1.2] - 2026-05-08
+
+### Added
+- `data-markdown-ignore` boolean attribute to exclude any HTML element (and its children) from Markdown output
+
 ## [0.1.1] - 2026-04-30
 
 ### Fixed
@@ -21,6 +26,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Support for `Accept: text/markdown` header and `.md` URL suffix
 - Rails 6.1+ and Ruby 3.0+ compatibility
 
-[Unreleased]: https://github.com/paultursuru/mark-don/compare/v0.1.1...HEAD
+[Unreleased]: https://github.com/paultursuru/mark-don/compare/v0.1.2...HEAD
+[0.1.2]: https://github.com/paultursuru/mark-don/compare/v0.1.1...v0.1.2
 [0.1.1]: https://github.com/paultursuru/mark-don/compare/v0.1.0...v0.1.1
 [0.1.0]: https://github.com/paultursuru/mark-don/releases/tag/v0.1.0

Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    mark-don (0.1.1)
+    mark-don (0.1.2)
       nokogiri (>= 1.13)
       rails (>= 6.1)
       reverse_markdown (~> 2.1)
@@ -290,7 +290,7 @@ CHECKSUMS
   loofah (2.25.1) sha256=d436c73dbd0c1147b16c4a41db097942d217303e1f7728704b37e4df9f6d2e04
   mail (2.9.0) sha256=6fa6673ecd71c60c2d996260f9ee3dd387d4673b8169b502134659ece6d34941
   marcel (1.1.0) sha256=fdcfcfa33cc52e93c4308d40e4090a5d4ea279e160a7f6af988260fa970e0bee
-  mark-don (0.1.1)
+  mark-don (0.1.2)
   matrix (0.4.3) sha256=a0d5ab7ddcc1973ff690ab361b67f359acbb16958d1dc072b8b956a286564c5b
   mini_mime (1.1.5) sha256=8681b7e2e4215f2a159f9400b5816d85e9d8c6c6b491e96a12797e798f8bccef
   minitest (6.0.6) sha256=153ea36d1d987a62942382b61075745042a2b3123b1cd48f4c3675af9cc7d6f1

README.md

@@ -86,6 +86,22 @@ This is a **great** room with a [nice view](/view).
 - Breakfast at 8am
 ```
 
+### Excluding elements
+
+Add `data-markdown-ignore` to any HTML element to exclude it and its children from the Markdown output:
+
+```erb
+<nav data-markdown-ignore>
+  <%= link_to "Login", login_path %>
+  <%= link_to "Sign up", signup_path %>
+</nav>
+
+<h1>My Room</h1>
+<p>Visible content.</p>
+```
+
+The `<nav>` block is removed before conversion — only the visible content remains in the Markdown response.
+
 ## Discoverability tip
 
 Once a page has a `.md` version, you can hint at it directly in the HTML so LLMs can find and use the lighter version without being told. Add a hidden element to your view (invisible to human visitors, readable by crawlers):

lib/mark_don/converter.rb

@@ -8,6 +8,7 @@ class Converter
     def self.convert(html)
       doc = Nokogiri::HTML(html)
       STRIP_TAGS.each { |tag| doc.css(tag).remove }
+      doc.css('[data-markdown-ignore]').remove
 
       body = doc.at('body') || doc
       ReverseMarkdown.convert(body.inner_html, unknown_tags: :bypass, github_flavored: true).strip

lib/mark_don/version.rb

@@ -1,3 +1,3 @@
 module MarkDon
-  VERSION = '0.1.1'
+  VERSION = '0.1.2'
 end

spec/mark_don/converter_spec.rb

@@ -45,6 +45,20 @@
       expect(result).not_to include('<script>')
     end
 
+    it 'ignores elements with data-markdown-ignore attribute' do
+      html = '<p>Visible</p><nav data-markdown-ignore><a href="/login">Login</a></nav>'
+      result = described_class.convert(html)
+      expect(result).to eq('Visible')
+      expect(result).not_to include('Login')
+    end
+
+    it 'ignores nested content inside data-markdown-ignore' do
+      html = '<h1>Title</h1><div data-markdown-ignore><p>Hidden</p><ul><li>Also hidden</li></ul></div>'
+      result = described_class.convert(html)
+      expect(result).to eq('# Title')
+      expect(result).not_to include('Hidden')
+    end
+
     it 'strips style tags and their content' do
       html = '<style>body { color: red; }</style><p>Hello</p>'
       result = described_class.convert(html)