David Kastrup wrote Thursday, February 23, 2017 2:29 PM

> Alexander Kobel <a-ko...@a-kobel.de> writes:
> 
>> It might be surprising that a non-existing voice does not even produce
>> a warning; but there *is* a (very sensible) difference.
>
[snip] 
> If searchForVoice is set (a context property) and this voice has
> a name, then this name is shortened by everything following its last
> hyphen and used as the corresponding voice name to synchronize to.
> 
> Really.  Fortunately, the default has been off for 6 years without
> anybody noticing or complaining:
> 
>    commit 4133b4e0ec939a427a70361ad838c65a84a0dfda
>    Author: Trevor Daniels <t.dani...@treda.co.uk>
>    Date:   Fri Apr 8 08:26:15 2011 +0100
> 
>        Search of context hierarchy for voice for lyrics is optional
> 
>         - add new context property, searchForVoice, to indicate
>           whether the extended search for a voice containing lyric
>           rhythms should be conducted
> 
>         - fixes 1579
> 
>         * lily/lyric-engraver.cc
> 
>           test searchForVoice in get_voice_to_lyrics () to decide
>           whether to conduct an extended search
> 
>         * ly/engraver-init.ly
> 
>           set default value of searchForVoice to #f
> 
>           (also in passing remove duplicated instrumentname
>            and shortInstrumentName settings)
> 
>         * scm/define-context-properties.scm
> 
>           add and document searchForVoice
> 
> Maybe we should remove that -- thing.

Probably sensible.  As I said at the time, I left it available in 
case any score or user was relying on it.  I'd no means of
determining whether or not that was the case then (nor have I now),
but if no one shouts soon that they are setting searchForVoice 
maybe it would be good to tidy this up.

Trevor
_______________________________________________
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Reply via email to