Footnotes correctly printed on grobs at first timestep.

Footnotes correctly printed on grobs at first timestep.

[MikeSol, 2012-06-10 15:58:25]

This fixes Issue 2574 but also deals with the footnote-break-visibility regtest,
which currently does not register the property change (this may have something
to do with the footnote engraver being on the Score level).

Cheers,
MS

[dak, 2012-06-10 16:09:03]

On 2012/06/10 15:58:25, MikeSol wrote:
> This fixes Issue 2574 but also deals with the footnote-break-visibility
regtest,
> which currently does not register the property change (this may have something
> to do with the footnote engraver being on the Score level).

I'll readily admit that the footnote engraver being at Score level may be a
source for new problems.  However, automatic footnotes get a number for each
time they hit an engraver, and footnotes may occur at pretty much any level.  So
I don't really see a way around having the engraver registered at Score level by
default.

Moving it to lower levels may be a user-level option, but it comes at the cost
of some elements no longer being available for footnoting.  So it would be good
if we could get the Score-level footnote engraver working well.

[MikeSol, 2012-06-10 16:22:25]

On 2012/06/10 16:09:03, dak wrote:
> On 2012/06/10 15:58:25, MikeSol wrote:
> > This fixes Issue 2574 but also deals with the footnote-break-visibility
> regtest,
> > which currently does not register the property change (this may have
something
> > to do with the footnote engraver being on the Score level).
> 
> I'll readily admit that the footnote engraver being at Score level may be a
> source for new problems.  However, automatic footnotes get a number for each
> time they hit an engraver, and footnotes may occur at pretty much any level. 
So
> I don't really see a way around having the engraver registered at Score level
by
> default.
> 
> Moving it to lower levels may be a user-level option, but it comes at the cost
> of some elements no longer being available for footnoting.  So it would be
good
> if we could get the Score-level footnote engraver working well.

It's worth mentioning in the change log and perhaps a convert-ly NOT_SMART rule.
 A once-over of the footnote regtests wouldn't hurt either if you have a bit of
time just to make sure they're doing what they claim to be doing.

[dak, 2012-06-10 16:24:52]

http://codereview.appspot.com/6306064/diff/1/input/regression/footnote-break-...
File input/regression/footnote-break-visibility.ly (right):

http://codereview.appspot.com/6306064/diff/1/input/regression/footnote-break-...
input/regression/footnote-break-visibility.ly:21: \once \override Score .
FootnoteItem #'break-visibility = ##(#f #f #t)
This looks like an unrelated fix that should just get pushed to staging
independently.  I am surprised that it did not show in the regtest comparison.

http://codereview.appspot.com/6306064/diff/1/lily/system.cc
File lily/system.cc (right):

http://codereview.appspot.com/6306064/diff/1/lily/system.cc#newcode257
lily/system.cc:257: end_of_line_visible = item->break_status_dir () == LEFT;
While I readily admit that this is breaking with the traditions set forth in the
existing code, could you add a comment about what you are doing here?

http://codereview.appspot.com/6306064/diff/1/lily/system.cc#newcode280
lily/system.cc:280: if (find (out.begin (), out.end (), at_bat) != out.end ())
And here?  Thanks.

[dak, 2012-06-10 16:30:24]

On 2012/06/10 16:22:25, MikeSol wrote:
> 
> It's worth mentioning in the change log and perhaps a convert-ly NOT_SMART
rule.

The change log describes changes relative to the last stable release.  A release
which did not even have footnotes.  If we wanted to put something in the change
log, mentioning the existence of footnotes in the first place would be a good
candidate.

I am not sure that the level of the footnote engraver is not something so fuzzy
that it makes little sense to even devise convert-ly rules.  At best, one could
try changing overrides for Staff.FootnoteItem appropriately in files that don't
tamper with Footnote_engraver, but that's all very hickety-wobbledy.

[dak, 2012-06-10 16:49:56]

http://codereview.appspot.com/6306064/diff/7001/lily/system.cc
File lily/system.cc (right):

http://codereview.appspot.com/6306064/diff/7001/lily/system.cc#newcode285
lily/system.cc:285: TODO
Ugh.  At least the comment now mentions that something is likely rotten
elsewhere.

Thanks!