On 12/29/08 5:45 PM, "Reinhold Kainhofer" <reinh...@kainhofer.com> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Am Dienstag, 30. Dezember 2008 01:27:05 schrieb Carl D. Sorensen:
>> On 12/29/08 3:56 PM, "Neil Puttock" <n.putt...@gmail.com> wrote:
>>> #(revert-auto-beam-setting '(end * * 2 4) 1 4)
>>
>> This reversion is necessary in all cases, and I think we should get rid of
>> it so it's not necessary, since the default beatLength will produce this
>> beaming.
>>
>> Once you have reverted the undesired auto-beam setting, you have three ways
>> to set the desired auto-beam setting:
>
>
>> 1) use #(override-auto-beam-setting '(end * * 2 4) 2 4)
>
> Hmm, I would have never guessed that I need to add a rule to end the beaming
> at the end of the measure (since lilypond will automatically break the beam
> there anyway)...
> What's the reason for this?
There's a priority for determining when auto-beams are ended.
1) Explicit rules
2) beatGrouping
3) beatLength
A rule from higher on the priority list takes precedence over one lower on
the priority list.
If no auto-beam rules are defined, then beatGrouping or beatLength are
used to determine when beams should end.
So adding a rule to end the beaming at the end of the measure eliminates the
"end at every beat" and "end at every beatGrouping" rules, because an
explicit statement of beaming has been given.
If no auto-beam-setting rules exist (which is what happens when you revert
the existing setting), and with beatGrouping not set to a valid value, the
default rule of ending on beatLength applies (which is why Neil's approach
to change beatLength works). However, if a valid beatGrouping exists, the
beatGrouping default overrides the beatLength default. That's why my
preferred solution of setting beatGrouping to (2) works.
IMO, 2/4 music has a beatLength of 1/4, so I prefer *not* to change
beatLength. I believe that the intent expressed is to group two quarter-note
beats in one beam. Hence, I think the most intuitive thing is to just set
beatGrouping to (2). But having beatGrouping work for autobeams was only
implemented in August, so lots of people have other ways to make autobeaming
work.
HTH,
Carl
_______________________________________________
lilypond-user mailing list
lilypond-user@gnu.org
http://lists.gnu.org/mailman/listinfo/lilypond-user
