From 4a72f7e9d89124c32b39194a5409bb8dc733486c Mon Sep 17 00:00:00 2001
From: DrumsnChocolate <matbronk@gmail.com>
Date: Thu, 21 Nov 2024 21:29:37 +0100
Subject: [PATCH] write a regex for html images. might not be perfect

---
 app/helpers/markdown_helper.rb | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/app/helpers/markdown_helper.rb b/app/helpers/markdown_helper.rb
index bd20938b..44309a25 100644
--- a/app/helpers/markdown_helper.rb
+++ b/app/helpers/markdown_helper.rb
@@ -7,10 +7,26 @@ def camofy(markdown)
     markdown.gsub(markdown_img_regex) do
       "![#{Regexp.last_match(1)}](#{camo(Regexp.last_match(2))}#{Regexp.last_match(3)})"
     end
+
+    markdown.gsub(html_img_regex) do
+      quote_mark = Regexp.last_match(2)
+      "<img#{Regexp.last_match(1)} src=#{quote_mark}#{camo(Regexp.last_match(3))}#{quote_mark}#{Regexp.last_match(4)}"
+    end
   end
 
   def markdown_img_regex
     # ![alt text](url =widthxheight "title")
     /!\[([^\[]*)\]\(([^\ )]+)(( =[^)]?[^ ]+)?( [^)]?"[^)]+")?)?\)/
   end
+
+  def html_img_regex
+    # warning: this regex may not be perfect. They rarely are. If you find an edge case, improve this regex!
+    # <img...something... src="url"
+    # or, the alternative quotes: <img...something... src='url'
+    # or, even without
+    # note that we don't allow mismatched quotes like 'url" or shenanigans like that
+    # This regex contains two particularly useful features:
+    #  capturing groups, and lazy matching.
+    /<img([^>]*) src=(["']?)(.+?)\2([ >])/
+  end
 end