2011/4/18 Janek Warchoł
<address@hidden>
Dear Bruce,
> I've read this thread. It seems to be mainly concerned with making the
> brackets extend to the full duration of the notes, like it is shown in the
> snippets, but that is not what I want to do here. I've added a couple of
> '\overrides' ('shorten-pair' was mentioned in the thread) which achieve the
> desired outcome (see below), but the solution seems terribly fragile. A
> different manual '\override' will need to be put in place for each different
> tuplet.
>
> It is not clear what units 'shorten-pair' uses (I just used trial and
> error);
I'd guess it uses staffspaces (the distance between staff lines), it's
a commonly used Lily unit.
> or why putting a value of 0.0 in the first position actually
> shortens the bracket on the left-hand end.
Very often 0.0 isn't the default value. Here 0.0 means flush with the
stem (i suppose), see attached.
Ah yes, I think that is it. The line is centred on the stem, rather than the inside edge of the line (as I was expecting).
> Lilypond seems to be doing what I want for the upper bracket - how to make
> it automatically do it for the other?
If no property to control this is mentioned in the documentation, then
probably you have to either write a cheme function doing this, or
modify Lily code itself so this option would become available.
Could you please write a feature request describing what you'd like
LilyPond to do in detail (with examples from some published scores if
possible) and send it to address@hidden? Hopefully someone would
implement this, but i don't have any idea when.
Or if you know some c++ go ahead and add it yourself! Contributor's
Guide (http://lilypond.org/doc/v2.13/Documentation/contributor/index.html)
tells you how to start. If you need any help, ask :)
Thanks for your suggestion, I'll send a feature request, which will be a new experience. I don't have a scanner, so I'll just have to include some Lilypond examples, and maybe a quote from Elaine Gould. I'd love to be able to say I'd that I'd write some C++ code, but I'm a long way from being able to do that now. I have only an extremely rudimentary grasp of what the Scheme interface does. I'm guessing that it is possible to do it using a Scheme function, but I'll need to devote some serious time to reading through the Scheme source files.
You advised earlier that the clashing numerals is a bug that still exists. I had some difficulty finding the bug reference, but it's covered by
"Issue
509:
collision nested tuplet numbers".
cheers,
Janek