From 51d8118b43b15c3bf9941d4d5aef05ee8a9a0535 Mon Sep 17 00:00:00 2001
From: apainintheneck <apainintheneck@gmail.com>
Date: Sun, 20 Oct 2024 19:51:11 -0700
Subject: [PATCH] Fix encoding bug in tests on Ruby 3

We need to guess the HTML encoding here otherwise some tests fail.

```
Failures:

  1) Readability images should show one image, but outside of the best candidate
     Failure/Error: @input = @input.gsub(REGEXES[:replaceBrsRe], '</p><p>')

     ArgumentError:
       invalid byte sequence in UTF-8
     # ./lib/readability.rb:51:in `gsub'
     # ./lib/readability.rb:51:in `initialize'
     # ./spec/readability_spec.rb:80:in `new'
     # ./spec/readability_spec.rb:80:in `block (3 levels) in <top (required)>'

  2) Readability the cant_read.html fixture should work on the cant_read.html fixture with some allowed tags
     Failure/Error: @input = @input.gsub(REGEXES[:replaceBrsRe], '</p><p>')

     ArgumentError:
       invalid byte sequence in UTF-8
     # ./lib/readability.rb:51:in `gsub'
     # ./lib/readability.rb:51:in `initialize'
     # ./spec/readability_spec.rb:555:in `new'
     # ./spec/readability_spec.rb:555:in `block (3 levels) in <top (required)>'
```

Fixes https://github.com/cantino/ruby-readability/issues/87

It also adds the latest Ruby 3 version to CI to test for these sort of bugs regularly.
---
 .github/workflows/ruby.yml | 2 +-
 lib/readability.rb         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/.github/workflows/ruby.yml b/.github/workflows/ruby.yml
index 1d00469..16c65f8 100644
--- a/.github/workflows/ruby.yml
+++ b/.github/workflows/ruby.yml
@@ -12,7 +12,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        ruby-version: ['2.7']
+        ruby-version: ['2.7', '3.3']
 
     steps:
     - uses: actions/checkout@v2
diff --git a/lib/readability.rb b/lib/readability.rb
index 4e4309f..2279517 100644
--- a/lib/readability.rb
+++ b/lib/readability.rb
@@ -43,7 +43,7 @@ def initialize(input, options = {})
       @options = DEFAULT_OPTIONS.merge(options)
       @input = input
 
-      if RUBY_VERSION =~ /^(1\.9|2)/ && !@options[:encoding]
+      if RUBY_VERSION =~ /^(1\.9|2|3)/ && !@options[:encoding]
         @input = GuessHtmlEncoding.encode(@input, @options[:html_headers]) unless @options[:do_not_guess_encoding]
         @options[:encoding] = @input.encoding.to_s
       end