Dirk Laurie wrote
> The A command is the obvious candidate. Maybe AB, but since the
> suggested syntax for a score-level beam template will always start `A[`,
the B
> seems unnecessary. If instrument-level or voice-level templates are to be
> accommodated, the B may be needed after all.
> I don't think staff-level is relevant to PMX.
>
> The suggested syntax for a template is to remove everything except
duration
> and brackets from PMX code in which all durations are explicit. For
example,
>
> A[8888][8888]
>
> would tell PMX to beam
>
> | c8 d8 e8 f8 g8 a8 b8 c8 |
>
> as "[ c8 d8 e8 f8 ][ g8 a8 b8 c8 ]". One would need a punctuation
character to
> distinguish between '][' and '] [', though, e.g.
>
> A[8888],[8888]
>
> would produce "[ c8 d8 e8 f8 ] [ g8 a8 b8 c8 ]".
This may not be a very good example because the default beaming in 4/4 meter
covers that pattern. I suspect you're really more interested in things like
the earlier 7/4 example. And as I said earlier, I'm not sure where the [888]
pattern came from in that case. So let me change the example to something
that is not included in the default, but where I know there's a well defined
set of defaults for the meter, say
A{888],[888],[88]
in 4/4 meter. Think about what you would want to happen when unbeamable
patterns, or other beamable patterns are encountered, such as "cd4 c8 d e c
d","cd4 c8 r e c d", "c1 d e e d c c8 d e c d", "cd2 c8 d", "cd2 c1 d e d",
"c2 c8 d e d", etc, etc. What I really need is a definitive set a rules that
covers all possibilities. One possible set would be simple: "if the pattern
doesn't match exactly, then use the defaults." Another equally simple one
would be "if it doesn't match exactly, then don't beam anything." I could
probably do either of those, and probably even with more than one
user-defined pattern, entered in priority order. A possible rule that would
likely be much more difficult to implement is "use the specified pattern
only if there are no rests and if all the notes can be subdivided among
legal beams of the same duration as in the pattern". More complicated still
would be "beam only beamable subgoups that start and end at the same places
as the beams in the pattern" but the "only" in that form would leave
unbeamed the 1/16's in e.g. "c1 d r c8 d e c d". So you might want to "beam
only beamable subgroups that do not include notes spanning the subgroup
boundaries in the pattern". And for any of the above possible rules, or any
other you might come up with, you need to decide whether to revert to the
default if the specified pattern is not adopted. It goes on and on.
--Don
-------------------------------
[email protected] mailing list
If you want to unsubscribe or look at the archives, go to
http://tug.org/mailman/listinfo/tex-music
