Re: LSR 1169 issue: missing point-and-click

[Jean Abou Samra]

Le mardi 20 juin 2023 à 11:43 +0200, Simon Albrecht a écrit :

Hello everyone,

it’s great to have LSR 1169 and it seems to be working as it should—many

thanks to Jean Abou Samra and Werner Lemberg!

However I’m sorry to write this mail because there is one issue I found

;) Any syllable but the last in a word doesn’t have point-and-click the

way it is implemented now. Maybe there’s a way to generate a transparent

markup just for point-and-click on the other syllables? I might be able

to work on this at some time, but I can’t say when, so in case someone™

wants to pick it up, I make it known here :)

>

Well...

There is no silver bullet here. In the original snippet I wrote, when two syllables are squashed, they would still be typeset separately, and then printed adjacently. Werner improved the snippet to make it typeset the two syllables together as a single run of text instead (with \concat), in order to enable font ligatures — and complex text layout, for languages where this is applicable — between the two syllables. But it might be fiddly, after making Pango lay out the combined text with ligatures, to extract back the parts. Off the top, I *think* it's possible (I'd have to check how to tell Pango to force an itemization boundary at a given point; I think it can do that, since it's also able to render arbitrary parts of the text in a different color), but I'm pretty sure it requires new C++ code.

In the very short term (i.e., what can we do in Scheme that works in 2.24), I'm afraid the only available improvement would be to make the whole syllable link back to one of the input syllables in the code, and giving up on having point-and-click separate the syllables.

By the way, this problem is more general, since it also affects properties like color and whiteout. (There's a reason it's in LSR and not in LilyPond proper yet.)

signature.asc
Description: This is a digitally signed message part