HTML Filter: throws exception while parsing html encoding when input is w/ &quot

Hi team,

I have a service that uses Okapi html filter (underneath, jericho parser http://jericho.htmlparser.net/docs/index.html as mentioned in this ticket’s first comment) to segment html.

When the input looks like (this input is inside an <iframe> html element):

<html>
   <head>  
      <meta http-equiv=&quot;Content-Type&quot; content=&quot;html; charset=UTF-8&quot;>
      </meta>
   </head>
</html>

Okapi throws:

Warning: Unsupported encoding UTF-8" specified in document

In the above text, encoding is specified in META tag. Okapi is taking this encoding to perform filtering. " is quotation mark ". Replacing ", <meta http-equiv="Content-Type" content="html; charset=UTF-8">. Okapi takes UTF-8" along with the quotation. Ideally it should take only UTF-8.

This input works correctly <meta http-equiv="Content-Type" content=\"html; charset=UTF-8\";>. When the quote has escape character, it correctly takes only UTF-8.

This seems a bug with html parser. I wonder if this would be fixed, or there are any suggestions deal with it on our end? Ideally we want to avoid hard-coded logic check for these inputs w/ &quot .

Comments (2)