Emit not-quite-cross-staff beams in the right context.

Emit not-quite-cross-staff beams in the right context.

This is related to 1043 and possibly other bugs.  Previously,
if a staff change happened immediately after the termination of
an auto-engraved cross-staff beam, then the beam was parented
to the wrong staff.  Now, every beam is parented to the context
in which it began.

[Keith, 2011-05-28 18:43:19]

I would love to test, but don't know enough of the internals to see what this
does.

[joeneeman, 2011-05-31 08:19:42]

On 2011/05/28 18:43:19, Keith wrote:
> I would love to test, but don't know enough of the internals to see what this
> does.

If you take your example from comment 5 of issue 1043, without manual beaming,
the Beam's Y-parent will be the VerticalAxisGroup of the top staff and
Beam::calc_cross_staff will return true. With manual beaming (or with this
patch), the Beam's Y-parent will be the VerticalAxisGroup of the bottom staff
and Beam::calc_cross_staff will return false.

[mike_apollinemike.com, 2011-05-31 08:55:08]

On May 31, 2011, at 9:19 AM, joeneeman@gmail.com wrote:

> Reviewers: Keith,
> 
> Message:
> On 2011/05/28 18:43:19, Keith wrote:
>> I would love to test, but don't know enough of the internals to see
> what this
>> does.
> 
> If you take your example from comment 5 of issue 1043, without manual
> beaming, the Beam's Y-parent will be the VerticalAxisGroup of the top
> staff and Beam::calc_cross_staff will return true. With manual beaming
> (or with this patch), the Beam's Y-parent will be the VerticalAxisGroup
> of the bottom staff and Beam::calc_cross_staff will return false.
> 
> 
> Description:
> Emit not-quite-cross-staff beams in the right context.
> 
> This is related to 1043 and possibly other bugs.  Previously,
> if a staff change happened immediately after the termination of
> an auto-engraved cross-staff beam, then the beam was parented
> to the wrong staff.  Now, every beam is parented to the context
> in which it began.
> 

Hey Joe,

Lines 1207-1209 in beam.cc are a kludge that I believe your patch resolves.  You
may want to consider removing these lines and testing to see if it still passes
regtests.  If so, I think you can nix these lines permanently.

Cheers,
MS

[joeneeman, 2011-05-31 11:55:35]

> Hey Joe,
> 
> Lines 1207-1209 in beam.cc are a kludge that I believe your patch resolves. 
You
> may want to consider removing these lines and testing to see if it still
passes
> regtests.  If so, I think you can nix these lines permanently.
>

After fixing my patch, I can indeed remove those lines.

Cheers,
Joe

[Keith, 2011-06-01 05:59:21]

On 2011/05/31 08:19:42, joeneeman wrote:
> the Beam's Y-parent will be the VerticalAxisGroup of the bottom staff
> and Beam::calc_cross_staff will return false.

Old patch didn't; but new patch does.  LGTM

[joeneeman, 2011-06-04 18:19:04]

On 2011/06/01 05:59:21, Keith wrote:
> On 2011/05/31 08:19:42, joeneeman wrote:
> > the Beam's Y-parent will be the VerticalAxisGroup of the bottom staff
> > and Beam::calc_cross_staff will return false.
> 
> Old patch didn't; but new patch does.  LGTM

Thanks, pushed. Also removed the 3 lines suggested by Mike.

[Bertrand Bordage, 2011-06-05 12:01:58]

If I understand well, there's no need anymore for manual beams in cross-staff
cases ?
If so, I suggest you remove the manual beams from
input/regression/beam-collision-cross-staff.ly.
You should also remove the knownissues in keyboard.itely and the last part of
changes.tely's entry.

Thanks,
Bertrand