From bff36de130f0eac213d86ccc4be856d70ca51a24 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?R=C3=A9gis=20Hanol?= <regis@hanol.fr>
Date: Wed, 3 May 2017 18:29:25 +0200
Subject: [PATCH] FIX: HtmlToMarkdown should not convert empty/bad <img> tags

---
 lib/html_to_markdown.rb                  | 21 +++++++++++++--------
 spec/components/html_to_markdown_spec.rb |  6 ++++++
 2 files changed, 19 insertions(+), 8 deletions(-)

diff --git a/lib/html_to_markdown.rb b/lib/html_to_markdown.rb
index 60f88d9095f..6e72d65fd4e 100644
--- a/lib/html_to_markdown.rb
+++ b/lib/html_to_markdown.rb
@@ -134,20 +134,21 @@ class HtmlToMarkdown
   end
 
   def visit_img(node)
-    if @opts[:keep_img_tags]
-      @stack[-1].markdown << node.to_html
-    else
-      title = node["alt"].presence || node["title"].presence
-      @stack[-1].markdown << "![#{title}](#{node["src"]})"
+    if is_valid_url?(node["src"])
+      if @opts[:keep_img_tags]
+        @stack[-1].markdown << node.to_html
+      else
+        title = node["alt"].presence || node["title"].presence
+        @stack[-1].markdown << "![#{title}](#{node["src"]})"
+      end
     end
   end
 
   def visit_a(node)
-    href = node["href"]
-    if href.present? && (href.start_with?("http") || href.start_with?("www."))
+    if is_valid_url?(node["href"])
       @stack[-1].markdown << "["
       traverse(node)
-      @stack[-1].markdown << "](#{href})"
+      @stack[-1].markdown << "](#{node["href"]})"
     else
       traverse(node)
     end
@@ -203,4 +204,8 @@ class HtmlToMarkdown
     (lines + [""]).join("\n")
   end
 
+  def is_valid_url?(url)
+    url.present? && (url.start_with?("http") || url.start_with?("www."))
+  end
+
 end
diff --git a/spec/components/html_to_markdown_spec.rb b/spec/components/html_to_markdown_spec.rb
index e5a9c5b8e2c..1e622e93f18 100644
--- a/spec/components/html_to_markdown_spec.rb
+++ b/spec/components/html_to_markdown_spec.rb
@@ -61,6 +61,12 @@ describe HtmlToMarkdown do
     expect(HtmlToMarkdown.new(HTML_WITH_IMG, keep_img_tags: true).to_markdown).to eq(HTML_WITH_IMG)
   end
 
+  it "removes empty & invalid <img>" do
+    expect(html_to_markdown(%Q{<img>})).to eq("")
+    expect(html_to_markdown(%Q{<img src="">})).to eq("")
+    expect(html_to_markdown(%Q{<img src="foo.bar">})).to eq("")
+  end
+
   (1..6).each do |n|
     it "converts <h#{n}>" do
       expect(html_to_markdown("<h#{n}>Header #{n}</h#{n}>")).to eq("#" * n + " Header #{n}")