Do you mean files that can't be indexed because of processing errors (like encrypted PDFs)?

If so, I completely agree, they should be added nonetheless and I tend to think that the current situation, when they are simply skipped, is rather a bug than a missing feature.