Re: chord name collide with staff

[address@hidden]

On 5 févr. 2013, at 22:15, address@hidden wrote:

> 
> On 5 févr. 2013, at 21:55, address@hidden wrote:
> 
>> 
>> On 5 févr. 2013, at 21:32, "address@hidden" <address@hidden> wrote:
>> 
>>> So, find #1...
>>> 
>>> If you replace all of the empty strings in the fourth lyric group by:
>>> \repeat unfold 33 \skip 1
>>> or
>>> \repeat unfold 33 " "
>>> The problem goes away.  So the skyline code doesn't like something about 
>>> the empty string...
>>> 
>> 
>> Find #2...
>> 
>> There is a vertical axis group of empty lyrics that is never suicided in the 
>> example with "" that is suicided in the example with \skip 1.
>> So, the finding of distances is somehow thrown off when a vertical axis 
>> group is empty.
>> An easy fix to this (just tested it) is to add to the end of 
>> Axis_group_interface::skyline_spacing the following:
>> 
>> if (skylines.is_empty ()) {
>> for (DOWN_and_UP (d))
>>  skylines[d].set_minimum_height (0.0);
>> }
>> 
>> That way, a proper distance can be measured between the empty skyline and 
>> the other one.
>> 
>> However, this seems kludgy - ideally, we should never let this vertical axis 
>> group survive to this point.  I'll try to find a way to scrub it earlier...
>> 
> 
> Find #3...
> 
> The problem is in the new axis-group-engraver code.  There are grobs being 
> added to the items-worth-living list via 
> Hara_kiri_group_spanner::add_interesting_item that are not with \skip 1.
> 
> These grobs probably got vetted out in 2.16.2.
> 
> So, the issue is to see why they are now added as interesting items but were 
> not before.
> 

Find #4...

Find #3 was wrong.  There should have been no suicides in this precise case.  
LilyPond just does not know how to do spacing on empty skylines.  This makes 
sense, as it is difficult to estimate the distance between something and 
nothing.  So, I'll use my previous solution of flattening the skyline to 0 with 
a nice comment.

Cheers,
MS