Comment #6 on issue 3639 by [email protected]: tuplet of a single chord gives a
syntax error
http://code.google.com/p/lilypond/issues/detail?id=3639
Unfortunately, "reveal its type in the first token" is not an automatic
advantage since optional arguments are checked by calling predicates. So
you still need to make an explicit syntax rule and testing expression for
every possible "revealed in the first token" argument. It's become easier
and more reliable to do, but it's still a chore. It does not even need to
be "revealed in the first token": it's just that at the point you make a
decision, the information gathered so far must be compressible into a
single token (a MARKUP_IDENTIFIER usually compresses quite more than a
single token). And you may invent synthetic tokens not naturally occuring
in the input for that as long as you teach other parts of the parser how to
deal with them.
So it's all manageable but tiresome. I'll take a look at the \repeat
thing: it's probably doable by _not_ looking at potential \alternative
before making a decision.
--
You received this message because this project is configured to send all
issue notifications to this address.
You may adjust your notification preferences at:
https://code.google.com/hosting/settings
