Phil Taylor wrote:
| John Atchley wrote:
| >Finally, just to throw an even bigger wrench in the works, I'm pretty sure I've
| >seen a bit of abc that names the voices, instead of numbering them. I.e.
| >something like V: alto to switch to the voice named alto. Apparently, there must
| >be some package out there that uses this convention.
|
| Yes I've seen that. I thought that came from an early version of
| abc2ps. You could do it with macros in BarFly if you really wanted
| to - just define V: alto to be equivalent to V:2 or whatever.
I suspect that you might have misinterpreted this. The current abc2ps
accepts V:alto, but it isn't a voice change. It means that the
current voice is to switch to alto clef. K:alto would do the same
thing, as would V:clef=alto.
Specifying the clef with either V: or K: seems to be very reasonable.
K: would be preferred by people typing a single voice (and not using
V: at all). But in multi-voice music, it's handy to be able to switch
clefs in one voice without affecting any others.
One complication that we might want to keep is that the tokens "bass"
and "alto" and "treble" imply "clef=". There are a number of ABC
tools that accept this now, and rather than break existing ABC, it
would be nice if we could just make these into alternatives. There
has also been a reasonable suggestion that we accept clef=G, clef=C
and clef=F with their expected meanings.
The Early Music gang would really like it if we could also place
clefs at other than the default position on the staff. One simple
syntax was to use middle=<note> to give the letter-to-line mapping.
This would make "clef=G middle=B" equivalent to "clef=treble" (or
just "treble"), while "clef=G middle=d" would be the French violin
clef. All three clefs are regularly placed at different positions by
at least a few kinds of musicians.
This would also provide a simple and elegant solution to the question
of how many commas to use in the bass clef. The current abc2ps
interprets bass clef as "clef=F middle=d", but some other programs
use "clef=F middle=D,". Both turn out to be reasonable and useful for
some users, so there's no reasonable way to limit it to one or the
other. The "middle=" syntax shouldn't be all that difficult to parse.
If you're doing clefs at all, you have to have a way to shift notes
up and down on the staff. This clause would just be a vertical offset
value, similar to what's needed to change clefs
These changes are also on my to-do list for my abc2ps clone. I could
use them all. I've also been thinking that, if I can get the middle=
thing to work, it could solve a problem with my Tune Finder. Since it
uses abc2ps to generate all the return formats except midi, people
who request music using the "clef=F middle=D," mapping see the bass
line written with zillions of leger lines below the staff. This isn't
very readable. A simple kludge (oops, I mean heuristic) of counting
the commas in the first staff or so could guess which convention was
used and adjust accordingly. It would still guess wrong on occasion,
but if the use of the middle= qualifier spread, this problem would
slowly decrease.
To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html