This is one more good reason to include groff.texinfo into the
testsuite of texinfo :-)I do always (try to) test groff.texinfo from your CVS repo before making a Texinfo release. I do not try to test every groff.texinfo change. I agree that the C/Lisp bias of Texinfo is unfortunate. Looking at groff.texinfo, I'm afraid I cannot easily understand why you need macros for this case in the first place, so don't have anything useful to suggest :(. Sorry. k
