Tie break-visibility of OctavateEight to the one of the associated clef

Tie break-visibility of OctavateEight to the one of the associated clef

It does not make sense to handle break-visibility of OctavateEight
grobs independently from the clef they are assigned to.
This patch lets OctavateEight always inherit break-visibility from
the Clef grob and ignores any break-visibility assigned to
OctavateEight.

[Neil Puttock, 2010-12-28 22:06:43]

Hi Reinhold,

I'd prefer a callback to copy the Clef's setting, then it's documented
automatically (and can be changed if absolutely necessary).

Cheers,
Neil

[Reinhold, 2010-12-28 23:02:25]

On 2010/12/28 22:06:43, Neil Puttock wrote:
> I'd prefer a callback to copy the Clef's setting, then it's documented
> automatically (and can be changed if absolutely necessary).

Hmm, I can't really think of a situation where you want to show or hide the
octavation eight independently of the associated clef (actually, I never
understood why it was a separate grob in the first place). 

The only situation might be to hide the 8 in a "treble_8" clef for a tenor
voice. However, the break-visibility of OctavationEight only applies to implicit
clefs (i.e. NOT to clef changes!). For explicit clefs, the
explicitClefVisibility applies to both the clef and the octavation (it is
hardcoded to be explicitly assigned to both in the engraver!). So, there is
currently no way to hide the 8 in all treble_8 clefs, anyway. The initial clef
will always have the 8 visible.

Of course, your suggestion of a callback (and the removal of the hardcoded
assignment of explicitClefVisibility to the octavate_ member) would allow the
user to hide ALL octavation eights, even for explicit clefs.

[Reinhold, 2010-12-28 23:43:46]

Okay, new patch uploaded, which uses a callback to inherit the OctavateEight's
break-visibility from the parent clef. This has the advantage that one can still
override the visibility independent from the clef (although I don't think there
is a real-life use for that)

Okay to push?

[Neil Puttock, 2010-12-29 00:00:37]

On 2010/12/28 23:02:25, Reinhold wrote:

> Hmm, I can't really think of a situation where you want to show or hide the
> octavation eight independently of the associated clef

Neither can I.  :)

> (actually, I never
> understood why it was a separate grob in the first place).

It makes it easier to tweak the appearance separately from the clef.

[Neil Puttock, 2010-12-29 00:01:45]

On 2010/12/28 23:43:46, Reinhold wrote:
> Okay, new patch uploaded, which uses a callback to inherit the OctavateEight's
> break-visibility from the parent clef. This has the advantage that one can
still
> override the visibility independent from the clef (although I don't think
there
> is a real-life use for that)

Looks OK, though it would be simpler to define the callback in output-lib.scm.

Cheers,
Neil

[Reinhold, 2010-12-29 00:09:04]

On 2010/12/29 00:01:45, Neil Puttock wrote:
> Looks OK, though it would be simpler to define the callback in output-lib.scm.

I had the same idea, so that patch is already uploaded ;-) This has the
advantage that it might also be used for other grobs

[Neil Puttock, 2010-12-29 00:55:06]

LGTM.

http://codereview.appspot.com/3838042/diff/2002/scm/output-lib.scm
File scm/output-lib.scm (right):

http://codereview.appspot.com/3838042/diff/2002/scm/output-lib.scm#newcode238
scm/output-lib.scm:238: (let ((parent (ly:grob-parent grob 0)))
(ly:grob-parent grob X)