Custos at the beginning of the score

[Toinewh]

Food for thought.

Some scores have a nice hint for beginners: a custos placed just after the key for the dominante note of the piece.
This is a little tricky to reproduce. One can make a T(j+)ri(g)stí(gj)ti(h)a(gf) try, but get an automatic hyphenation that should not be. One can make a Tri(j+/[4]g)stí(gj)ti(h)a(gf) try, but get now a syllable centered on the custos taken for a single neume.

This drives me to a double suggestion.

• It could be nice to prevent hyphenation for a particular syllable.
• And it could be useful to choose which part of the neume should be use for centering the syllable, as it is possible to define the lyric part for centering with braces…

See below to have an idea of the expected result.
Test.pdf

[rpspringuel]

This kind of custos might be better as part of the initial clef, rather than the first syllable.

As for the actual suggestions you made, are there circumstances outside this particular use case in which one or both of these suggestions would be useful?

For the hyphens, this is probably possible as we already have a flag (\ifgre@showhyphenafterthissyllable) that is used to decide if they hyphen shows up. It would just be a matter of adding a way to manipulate this flag manually.

Changing the centering point in the neumes, however, is much harder because the gabc doesn't actually encode the neumes, it encodes the notes. What neume is associated with a sequence of notes is determined by context and we'd need some firm rules on how the forced centering note effects the choice of neume and how to center on notes that are in "weird" places in their neume (think the middle note of a porrectus).

[Toinewh]

I agree with your suggestion. For this particular purpose, the custos as a part of the key came in my mind a few days ago too. And it would be much better for handling, typically something like (c4:j) at the beginning of the score is quite readable…

For the hyphen, it could lead to another use, but I don't have any example in mind. In my opinion, the key-part option solves the issue.

Besides, maybe in GABC we could indicate with braces which part of the bunch of notes should be considered for a custom centring… Tri(j[g])stí(gj)ti(h)a(gf) But I'm not sure that this option would be very useful.