Community

Anyone can give me some context as to what they're trying to do, and why they're using GZIPInputStream? 

Weird, it looks like it's assuming that there exists a pre-indexed version of AnnotationRepository.

e.g. this is a notion from Papaki, where one would do standalone scan, and serialize pre-indexed repo instance into a jar

You need to check why it returns non-null input stream from this code:

protected InputStream getMcAnnInputStream(URL url)
   {
      try
      {
 
         String urlString = url.toExternalForm();
         if (urlString.endsWith("/") == false)
            urlString += "/";
         urlString += AnnotationRepository.MCANN_PATH;
         URL mcannURL = new URL(urlString);
         return mcannURL.openStream();
      }
      catch (IOException e)
      {
         if (log.isTraceEnabled())
            log.trace("Cannot access mcann.ser file: " + e);
         return null;
      }
   }