Running the checker on an unknown domain name such as:
 http://idontexistatall.org
...yields to the following result where HTTP_RESPONSE-1 is returned twice for some reason:

<?xml version="1.0" encoding="UTF-8"?>
<tests outcome="FAIL">
   <test name="MAIN_DOCUMENT" outcome="FAIL">
      <result name="HTTP_RESPONSE-1" outcome="FAIL">
         <info>FAIL: The request to the resource http://idontexistatall.org does not result in a valid HTTP response (because of network-level error, DNS resolution error, or non-HTTP response) </info>
      </result>
      <result name="HTTP_RESPONSE-1" outcome="FAIL"/>
   </test>
</tests>

The incriminated code is in the MainDocumentTest class, where the second HTTP_RESPONSE-1 result is being added. The comment there states "This is hard to add any other way".

The test condition that triggers the addition of the second HTTP_RESPONSE-1 is:
 if (results.getSaxonMokiDocument() == null ||
   results.getPrimaryDocument().getBody() == null)
In my case, the moki document is not null, and there's indeed no body to the primary document since it cannot be retrieved.

I think the whole thing was a needed hack that is no longer needed. I guess it's harmless and safer to keep the condition on the moki document being null, but don't see why we need the second test in there.