crescendo/diminuendo lines take vertical height from hairpin settings rather than dynamics text settings

Actually it *is* by design. Hairpin textlines were specifically designed to automatically line up as well as possible with dynamics. So they respond to the dynamics text style.

Well, I am not so sure - everything about this stuff has changed for master because of the new layout and auto-position of various elements. As far as I can tell, the crescendo lines are placed according the dynamic attached to the start note, if present and autoplaced. This is actually pretty reasonable behavior for that case. Less clear what the behavior should be if there is *not* dynamic on the start note though. Sure, in the old days we used text style for that, but I gather that's not even a thing any more.

At this point I'm inclined to call this "by design". Sure, somewhat different from 2.x in that hairpin lines now respond to hairpin height rather than dynamic height, but that's not inherently wrong. Does someone have a case that shows why a change is needed?

(also changed issue title to reflect what is actually going on - for 2.x they obeyed dynamics text height, for 3.0 they obey hairpin height, I think that actually makes more sense, although I'm sure I had a reason for doing it the way I did originally)

I'm inclined to agree that this is by design and should be kept that way. One problem with 2.x is that the default positions for dynamics didn't line up with the hairpins, so I always adjusted them. I don't see this problem in version 3.

OK, now that I look further, I see this really is a problem. The default height for hairpins is deliberately set to be 0.5sp higher than for dynamics, so that they will align on their vertical centers, and this does work correctly in both 2.3.2 and in 3.0. That 0.5sp offset is, I think, why I didn't just use the hairpin height for the hairpin lines - these are aligned based on the baseline, not on the vertical center. Using the dynamics text style solved that nicely.

Now that text styles are back, we could elect to use the same approach, or maybe solve it in how we do the layout. Like, what if we made the hairpin height setting the same as the dynamics height, but during layout offset the hairpins by half their height, or something like that? Having that hardcoded 0.5sp discrepancy in the style settings has always seemed a bit kludgy anyhow.

Here is the problem:

Select the first hairpin in Marc's example, and use the Inspector to change the Type (in the "Hairpin" section) from "Crescendo Hairpin" to "Crescendo Line". This is the result.

The text is aligned with the dynamics, but the line is not aligned with the baseline of the text.

Although I wrote much of the original code for these elements, and I do know a thing or two about style settings, there is a bit too much going on here in terms of the "property" settings for me to feel comfortable attempting to fix this.

Fixed in branch master, commit 487db7fe46

fix #164786 crescendo/diminuendo lines take vertical height from hairpin settings rather than dynamics text setting
The fix introduces new style values for crescendo/diminuendo lines. This moves them to the same position as dynamics.
Also fixed is a another bug: when you switch between above/below the proper offsets were not shown in
the inspector. Another bug was not initializing styled properties right after reading a score.

Automatically closed -- issue fixed for 2 weeks with no activity.