Anchoring a rehearsal mark to a barline instead of note head/break
Usually rehearsal marks are also used as section marks.
Those section marks should be anchored to a barline and not a note head.
This currently is only possible with manual adjustment(and readjustment in parts...)
Would it be easy to allow barlines as anchor targets for rehearsal marks?
| Attachment | Size |
|---|---|
| Bildschirmfoto vom 2014-04-14 22:20:11.png | 5.42 KB |
Comments
I've thought about that too, but it turns out it really isn't necessary. Just attach to the first note/rest of the measure, but have the X position in the text style set to an appropriate negative value. That's what I do in my own templates (including the ones I provided to ship with MuseScore: Jazz Lead Sheet & Jazz Combo). Manual adjustment is needed only very seldom - I use it only if the measure is the first in a system and I want the mark all the way left of the clef.
In reply to I've thought about that too, by Marc Sabatella
I have always used the same system Marc advocates. However, anchoring it to a bar line makes logical sense.
Looking at the source code, it would seem a rather simple change, with no obvious side effects. I think I'll have a try at it...
M.
In reply to Seems doable by Miwarre
What about rehearsal marks that apply to the last beat of a measure, kind of an anacrusis mid-piece?
Pull request adding this pushed onto Github: //github.com/musescore/MuseScore/pull/837 .
M.
In reply to Pull request adding this by Miwarre
How does it handle measures at the start of a system? I'd hope that barline still means barline, I guess; if the staff has no beginning barline, the anchor point would be the right edge of the staff. Not sure how'd you actually manage to create one of those, though - nothing to deag too. Also be sure they still work well with mmrests.
In reply to How does it handle measures by Marc Sabatella
If there is no bar line, nothing can be attached!
"if the staff has no beginning barline, the anchor point would be the right edge of the staff"
Do you mean the left edge of the staff or the right edge of the previous staff? Anyway, see above: no bar line? no party! (and either solution would look ugly!).
About multi measure rests: nice point, I didn't think of them; I'll try.
Thanks,
M.
In reply to If there is no bar line, by Miwarre
Yes, I meant left. So, what happens if you attach a rehearsal mark to a barline, then add a line break to that barline, so now it starts a new system and has no barline. Does the rehearsal mark appear at the end of previous system? Not good if so. I'd want it at the beginning of the next, right above the non-existent left barline :-).
Seems like a bit of a can of worms, actually. I'd kind of prefer an implementation where you continued to attach to a *note*, but it somehow knew to *display* aligned with the start of the *measure*. Don't we do anything like that for any other element types? Seems codas et al work kind of like that.
In reply to Yes, I meant left. So, what by Marc Sabatella
Telling from memory, I wold say that at least voltas are attached (or may be attached) to a measure boundary.
Anyway, would an algorithm like the following meet normal practice and user expectations?
1) If the rehearsal mark is on the first chord or rest of a measure, align it with the previous bar line (or with the potential location of the bar line, if in the first measure of a system);
2) otherwise, align it with the chord/rest it is on.
Note that, by design or by chance, whole-measure rests already make a rehearsal mark on them to display on bar line alignment (but a chord or any other type of rest do not).
Thanks,
M.
In reply to Telling from memory, I wold by Miwarre
Voltas can only be attached to measures, this sometimes leeds to the need to split a measure.
And yes, your proposal would meet my expectations :-)
For rehearsal marks at the beginning of a System, where there is no barline, it could/should show where a (repeats-) barline would appear if added, i.e. shortly after the timesig.
Wouldn't the same proposal also apply to Tempo Markings?
In reply to Voltas can only be attached by Jojo-Schmitz
Tempo markings have a similar issue, but a bit different. They really are supposed to appear directly over the note in most cases - and that one's reason one might want rehearsal marks further to the left, to avoid tempo markings (also chord symbols). But if a measure has a time signature, then a tempo marking attached to the first note should display above the time signature instead. Not all the way to the barline, though. Especially not if there is also a clef and/or key signature.
This sort of special casing in layout algorithms troubles me a little when I put my "software engineer" hat on, but as a user, I do think this would be very cool. Rehearsal marks align to beginning of measure if they are attached to first chordrest of measure, tempo markings align to time signature if they are attached to first chordrest of measure and previous segment is a time signature. Seems simple enough, and it would eliminate a whole bunch of the manual adjustments I normally expect to have to do.
But I just realized what I wrote is different from what Jojo wrote. For a first measure of a system, I expect a rehearsal mark to be at far left - aligned with the left edge of the staff, to the left of clef, key, and time. So you can see it in a quick scan of the margin. Whereas Jojo said he would prefer it where the repeat barline would otherwise appear - to the *right* of the clef, key, and time signature. I recognize this is a personal decision and fully expect there to still sometimes be a need for manual adjustment, so I'd be fine with it either way.
In reply to Tempo markings have a similar by Marc Sabatella
The need to manually adjust a tempo marking when there is a time signature bothers me since long (I thought it was a private perversion of mine, but I see I'm not alone...).
About the difference for system-initial rehearsal mark (at staff left edge or right after clef and key sig), for what is worth, looking at Beethoven, Steichquartett Op. 18, no 1, as published by Universal Ed. ca. 1920 (Kalmus reprint, ca. 1937), available here , there are several examples of system-initial rehearsal marks ('D' in p. 4, 'F' in p. 6, 'R' and 'S' in p. 14, ...), all consistently left aligned to the right edge of the key signature.
Of course, it is just one source, but from a reputed publisher anyway, in a time when everything was still engraved by hand and established practice followed.
If anybody has different examples, possibly at least as much authoritative...
M.
In reply to Any 'prior knowledge' example? by Miwarre
The examples I would be more likely to refer to would be much more recent - jazz fakebooks and big band charts. Looking over what I have access to, I'd say directly over the key key sig and all the way left are about equally common. Complicating this is the unfortunate convention among some publishers to omit clefs. So there is basically no difference between aligned to key sig versus barline with these editions.
Elaine Gould, btw, doesn't cover this. She just says rehearsal marks need to be left of tempo marks.
In reply to Telling from memory, I wold by Miwarre
I have decided to revisit this, and implemented (in my own test builds only so far) something like what we've discussed above. That is, rehearsal marks and tempo texts still attach to chord/rests only, but if they happen to be attached to the first chord/rest of the measure, then the layout algorithm moves them to the left as appropriate. See discussion and screen shot here:
http://musescore.org/en/node/25247#comment-119341
As I mention, I'm still not sure how to handle rehearsal marks at the beginning of systems, but the basic approach seems to work well.
I am withdrawing the pull request, partly following Marc's comments above, partly because the current implementation seems not very consistent:
1) Display: If a rehearsal mark is attached to a whole measure rest it is displayed on the previous bar line, but if it is attached to the initial (non-whole measure) rest of a measure, it is displayed above the rest.
2) Multi-measure rests: Assuming at least 2 parts, A and B; if the rehearsal mark is attached to a mid-measure chord or rest (in part A) and another part B is nothing in that measure (i.e. a whole measure rest), the extracted part for B breaks the multi-measure rest before the measure, but does not separate the measure itself (i.e. the measure is included in the next MM rest) and does not show the rehearsal mark at all.
So, it seems to me better to improve the current implementation before adding new features. This may also be the occasion for improving the display of measure-initial rehearsal marks (for instance, on the measure initial bar line, as Marc suggested) and dropping this request.
Thanks,
M.