Code review comment for lp:~ian-clatworthy/bzr/faster-tags

It's helpful when we're reporting on bzr progress to put an indicative performance number in the news like "5-10x faster". (Especially when it's as good as you've done here.) I realize it's not spads-approved and ymmv etc.

I might be missing something but it looks like in the previous code the dict 'tags' was indexed by tag name but here it's indexed by the enumeration integer?

If that's true, shouldn't a test have failed?