On 03/19/2013 04:21 PM, Shane Brandes wrote:
I see now, but one would think that might cause more difficult
programming necessitating the keeping tracking of various R values
through out the piece as defined by a time signature as opposed to us
setting the value, which probably would also slow down lilypond having
to parse and hang on to that extra undeclared information. Of course
maybe i am wrong on that. But I do believe making sure the user is
forced to think out the value of R duration is probably more useful
than any minimal convenience, especially as Joram points out that
would make r and R functionality divergent.
FWIW the multimeasure rest syntax has been a bit of an annoyance with
what I've been working on recently (Mars). Not only do I have to type
something like R1*5/4*8 instead of just R*8 (the syntax someone else
proposed and which I too thought of), but because I've forgotten about a
time change when typing the duration. And thus entered, e.g. R1*5/4*8
instead of R1*5/2*8. The *really* annoying part of that particular
mistake is that R1*5/4*8 is actually valid: it gives a 4-measure rest in
5/2. It took a fair bit of head scratching to figure out why Lilypond
was getting the duration of the rest "wrong" the first time I made that
mistake. And I *still* make that mistake with some frequency -- I'm just
faster at spotting the cause of the bad output now.
To me it's quite counter-intuitive that a "full measure rest" can mark
any durations other than a multiple of a full measure, even if it does
sometimes produce a warning if you do something wrong, and the proposed
suggestion is a significant local improvement.
Evan
_______________________________________________
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user
