No modules or parsing needed, just a simple one line regex.

The only problem is that it doesn't prune every valid empty XML tag. If getting the correct result is important, use an XML parser.