If your score gets too long, though, and you have too many of one before the other, you can get overflows from this. Because the action of checking each scalar is a result of the output of the previous scalar, one "next" message can start an indefinitely long string of messages. I find using a 0 ms delay to work for this; the next string is started as a new command, or at least that's how it seems. Anyway, it works as in the attached patch.
This might not be an issue for you, if you are talking about two separate structures that exist for each note, IF you sort the score each time, because then there is never more than a couple of one before the next instance of the other. But if you start adding anything else to your score, like tempo markers, bar lines, or any sort of automation, you might run into this.
-Chuckk
On 6/5/06, Frank Barknecht fbar@footils.org wrote:
Hallo, Frank Barknecht hat gesagt: // Frank Barknecht wrote:
You forgot to attach the patch, but I guess what's happening is some error in traversing the data structures in the patch.
Okay, I've seen your patch now and indeed what's happeing is what I've suspected. One additional note: If you use [pointer] with one or more type arguments as suggested, you'l probably want to make sure to skip unwanted types and autoamtically advance the traversal with a "next" message. You can do this as in attached patch.
Ciao
Frank Barknecht _ ______footils.org_ __goto10.org__