Fix 1563: System start bars interpreted collapse-height as absolute length.

Fix 1563: System start bars interpreted collapse-height as absolute length.

If you increased the staff-space, this meant sometimes the collapse-height
would not be enough to hide the start-bar for a staff, while in other cases
it was enough...

This patch interprets the collapse-height in multiples of the staff-space.

However, I think that the notion of a collapse-height (as a length) for
hiding/showing the system start delimiter is not the best approach in
general.

[pkx166h, 2011-07-12 23:40:54]

Pass Make but I ge a difference in one reg test

see

http://code.google.com/p/lilypond/issues/detail?id=1563#c3

where I have posted the screenshot

[Keith, 2011-07-16 03:57:40]

Looks good to me.

The differences in beam-skip.ly were not due to this patch.

On 2011/07/12 23:40:54, J_lowe wrote:
> Pass Make but I get a difference in one reg test
>

[Neil Puttock, 2011-07-18 20:24:43]

http://codereview.appspot.com/4693043/diff/1/lily/system-start-delimiter.cc
File lily/system-start-delimiter.cc (right):

http://codereview.appspot.com/4693043/diff/1/lily/system-start-delimiter.cc#n...
lily/system-start-delimiter.cc:117: staffspace =
Staff_symbol_referencer::staff_space (sp);
A mostly theoretical gripe, I guess: what about collapse-height comparisons
where multiple staves are involved?  If staff-space settings differ, this only
picks up the last stave's value.

[Reinhold, 2011-07-20 17:02:02]

On 2011/07/18 20:24:43, Neil Puttock wrote:
> lily/system-start-delimiter.cc:117: staffspace =
> Staff_symbol_referencer::staff_space (sp);
> A mostly theoretical gripe, I guess: what about collapse-height comparisons
> where multiple staves are involved?  If staff-space settings differ, this only
> picks up the last stave's value.

Yes, that was a deliberate decision. After all, the collapse-height is mainly
intended to hide/show the bracket for single-staff systems.
In my eyes it would make more sense to show/hide the bracked depending on the
number or staves involved rather than the grob height.
I can't imagine a situation, where using the last staff's staff-space would lead
to unwanted output. Usually, collapse-height will be either 5.0 (the default) or
something smaller as advocated in the documentation and the LSR. In these cases,
the last staff alone (plus some spacing to the staff above for 5.0) already
suffices to show the bracket, even if the other staves have a different
staff-space.

[Neil Puttock, 2011-07-21 17:50:06]

On 2011/07/20 17:02:02, Reinhold wrote:
> On 2011/07/18 20:24:43, Neil Puttock wrote:
> > lily/system-start-delimiter.cc:117: staffspace =
> > Staff_symbol_referencer::staff_space (sp);
> > A mostly theoretical gripe, I guess: what about collapse-height comparisons
> > where multiple staves are involved?  If staff-space settings differ, this
only
> > picks up the last stave's value.
> 
> Yes, that was a deliberate decision. After all, the collapse-height is mainly
> intended to hide/show the bracket for single-staff systems.
> In my eyes it would make more sense to show/hide the bracked depending on the
> number or staves involved rather than the grob height.

I agree.

> I can't imagine a situation, where using the last staff's staff-space would
lead
> to unwanted output. Usually, collapse-height will be either 5.0 (the default)
or
> something smaller as advocated in the documentation and the LSR. In these
cases,
> the last staff alone (plus some spacing to the staff above for 5.0) already
> suffices to show the bracket, even if the other staves have a different
> staff-space.

OK, LGTM.

Cheers,
Neil

[Reinhold, 2011-07-22 12:52:17]

On 2011/07/21 17:50:06, Neil Puttock wrote:
> OK, LGTM.

Pushed to git as commit 1c4662c4773eb0c646011c82a2dd2f7c5f92a283