AFAIK The indexer can only index plain text. So no it won't be able to understand compressed archives. How about defining a rule that unpacks the archived content, then the indexer will find it? Anyone have a better solution such as a transformer for zip archives?
