lilypond-user
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Barcheck failure warnings for lyrics in parallel mode


From: Galen Menzel
Subject: Re: Barcheck failure warnings for lyrics in parallel mode
Date: Fri, 10 Aug 2018 12:16:48 -0700



On 9 Aug 2018, at 0:53, David Kastrup wrote:

"Galen Menzel" <address@hidden> writes:

On 8 Aug 2018, at 14:10, David Kastrup wrote:

"Galen Menzel" <address@hidden> writes:

So basically explicit bar markings and lyrics just don’t play
nicely
together?

And auto-length lyrics.  If you give the lyric syllables explicit
durations and not use \lyricsto/\addlyrics but just set
associatedVoice
and \lyricmode appropriately, you can write bar checks into the lyrics perfectly well. Note that bar checks indicate that your lyrics _are_
written to match a given rhythm at seminal points, so there is some
motivation to either put none of the rhythm into the lyrics (and let
them get autotimed using \addlyrics/lyricsto) or put all of it in and
use explicit durations.

Ok, that makes sense. Looks like this is just a rough edge, where the
requirements of auto-length lyrics are in conflict with the
requirements of parallel notation. I’ll look into explicit duration or
suppressing the warning.

You could just remove the bar checks from the lyrics selectively, like

lyrics_b = #(map-some-music (music-type-predicate 'bar-check) lyrics_b)

However, that would not get you rid of the warnings because of music
with unequal length.  The warnings are intended for locating problems:
removing them has its own problems. Have you considered specifying your
lyrics with durations, like \lyricmode { one 2. two 4 } (spaces before
durations are optional)?

This is what I ended up doing. It’s a somewhat unsatisfying solution, since it’s clearly something a machine could do. But it seemed to be worth it in this case, since searching for lyrics are the easiest way to jump to a bar without having to look up the bar number.

There seems to be a legitimate use case here for suppressing warnings related to only one specific voice in a parallel-music expression. Is there a way of doing this? If not, do you think the dev team would be open to such a feature request?

Thanks!

Galen



reply via email to

[Prev in Thread] Current Thread [Next in Thread]