Joe Neeman escreveu:
> On 11/28/06, *Han-Wen Nienhuys* <address@hidden
> <mailto:address@hidden>> wrote:
>
> Joe Neeman escreveu:
> > On 11/28/06, *Werner LEMBERG* <address@hidden <mailto:address@hidden>
> <mailto:
address@hidden <mailto:address@hidden>>> wrote:
> >
> >
> > Joe,
> >
> >
> > here's a quite important problem regarding skyline positioning and
> > outside-staff-priority. Hopefully, you can fix that easily
> since this
> > feature rocks!
>
> > I'm not on a lilypond-enabled computer right now, so it could be
> caused
> > by a bug, but I can't tell from your example.
> >
>
> marks are typeset at score level. I'd be surprised if they partook
> in skyline stuff.
>
>
> How's this? I just recurse into smaller boxes if I find a
> vertical-axis-group (before, the System was treating the
> VerticalAlignment as a single grob and therefore the skyline was just a
> big box).
Sounds like a good idea. There's one thing I'm not comfortable with
(and I was bitten by this doing the separation item stuff), but I'm
not sure it could be a problem in this case.
It is possible to ask for the X,Y extent of an axis group which is X
but not Y. In that case, the [Y] dimension will be bogus, which may
affect the result of the skylining.
I haven't been affected by this yet, but I'm not really sure what you mean. If the grob doesn't have a valid Y-extent, it should return an empty interval which is then filtered out by the Skyline constructor. Or am I missing something?