Switch to Nokogiri::HTML for html writer

By switching to the HTML parser from the XML parser, we can better handle malformed html documents and save out something that at least marginally resembles the input HTML.
approvals · Apr 5, 2014 · 05dc7f6 · 05dc7f6
1 parent f97c7a7
commit 05dc7f6
Show file tree

Hide file tree

Showing 3 changed files with 23 additions and 1 deletion.
diff --git a/lib/approvals/writers/html_writer.rb b/lib/approvals/writers/html_writer.rb
@@ -7,7 +7,7 @@ def extension
       end
 
       def format(data)
-        Nokogiri::XML(data.to_s.strip,&:noblanks).to_xhtml(:indent => 2, :encoding => 'UTF-8')
+        Nokogiri::HTML(data.to_s.strip,&:noblanks).to_xhtml(:indent => 2, :encoding => 'UTF-8')
       end
 
     end

diff --git a/spec/approvals_spec.rb b/spec/approvals_spec.rb
@@ -57,6 +57,17 @@ def hello.inspect
     Approvals.verify html, :format => :html, :namer => namer
   end
 
+  it "verifies a malformed html fragment" do
+    html = <<-HTML
+<!DOCTYPE html>
+<html>
+<title>Hoi</title>
+<script async defer src="http://foo.com/bar.js"></script>
+<h1>yo</h1>
+    HTML
+    Approvals.verify html, :format => :html, :namer => namer
+  end
+
   it "verifies xml" do
     xml = "<xml char=\"kiddo\"><node><content name='beatrice' /></node><node aliases='5'><content /></node></xml>"
     Approvals.verify xml, :format => :xml, :namer => namer

diff --git a/spec/fixtures/approvals/approvals_verifies_a_malformed_html_fragment.approved.html b/spec/fixtures/approvals/approvals_verifies_a_malformed_html_fragment.approved.html
@@ -0,0 +1,11 @@
+<!DOCTYPE html>
+<html xmlns="http://www.w3.org/1999/xhtml">
+  <head>
+    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
+    <title>Hoi</title>
+    <script async="" defer="defer" src="http://foo.com/bar.js"></script>
+  </head>
+  <body>
+    <h1>yo</h1>
+  </body>
+</html>
-Original file line number
+Diff line change
@@ Expand Up / @@ -7,7 +7,7 @@ def extension @@
           end
           def format(data)
-            Nokogiri::XML(data.to_s.strip,&:noblanks).to_xhtml(:indent => 2, :encoding => 'UTF-8')
+            Nokogiri::HTML(data.to_s.strip,&:noblanks).to_xhtml(:indent => 2, :encoding => 'UTF-8')
           end
         end
@@ Expand Down @@