Re: Spacing for s4

[Thomas Morley]

2016-12-04 17:50 GMT+01:00 Carl Sorensen <address@hidden>:
>
>
> On 12/3/16 7:12 AM, "PMA" <address@hidden> wrote:
>
>>
>>With such a simple solution (my using 'r' instead & setting
>>rests to transparent), I'm not inclined to pursue the issue.
>
> I suspect (although I haven't tried it) that using 'r' and setting the
> stencil to #f would produce exactly the same output as using s.
>
>>
>>I do wonder, though, if spacers are not to make space (for
>>themselves, so of course nudging the next event rightward),
>>why they are called "spacers" at all.
>
> Yes, that is an interesting question.
>
>
> Until I reread the manual I had forgotten that spacer rests are really
> called "invisible rests", which exactly describes what they do.  Perhaps
> we should change the documentation to have most of the text read
> "invisible rest".
>
> Thanks,
>
> Carl



Well, "spacer rests" don't create a NoteColumn, as opposed to Rests
(visible or not) and no grob is assigned to them.
No bug but a feature.

This becomes important once we want to deal with fancy glissandi.
(There may be other use cases apart from glissandi. It's just the one
I use sometimes.)

In the following example only the first measure puts out a glissando at all:

{
  c''4\glissando s c'''2
  \omit Rest
  c''4\glissando r c'''2
  \override Rest.after-line-breaking = #ly:grob-suicide!
  c''4\glissando r c'''2
  \override Rest.before-line-breaking = #ly:grob-suicide!
  c''4\glissando r c'''2
}

Currently I've no idea how we could keep that behavior _and_ giving
spacers some extend of there own, without creating a grob for them.

For the name, "spacer" may be ill-named, internally they are called
SkipEvents, which comes closer. Maybe "timer" would even more
appropiate, but sounds even more strange.

Cheers,
  Harm