Re: Path to LilyPond 2.18

[Keith OHara]

On Wed, 05 Jun 2013 22:38:17 -0700, Mike Solomon <address@hidden> wrote:
>  Could someone please send me minimal examples of things going wrong?

The issues on the tracker that are not marked Fixed have minimal examples of 
things still going wrong.

> On 6 juin 2013, at 06:52, Keith OHara <address@hidden> wrote:
>
> > Issue 3359, the infinite-recursion crash with the failed cross-staff tie,
> > also worries me.  It shows that one of the properties with 'pure' in the
> > name, 'pure-relevant-grobs' includes objects from different staves.
>
> Grobs from different vertical axis groups being in a pure properties array 
> shouldn't, in theory, cause a problem, as at that point the Y-common is a 
> vertical alignment and one would just use pure minimum translations instead of 
> minimum translations.

So, a Tie starting in the upper staff is for some reason put on the 
'pure-relevant-grobs' list of the lower staff.   When LilyPond estimates the 
'pure-height' of the lower staff while planning page breaks, she wants to 
include every 'pure-relevant-grob' in that height, including that Tie.  She 
knows where the Tie is attached on the upper Staff, so just needs the 
'pure-minimum-translations'  between the staves to get the position of the Tie 
relative to the lower staff.  All as you say.

The 'pure-minimum-translation' between the upper and lower staff is 
conservative, allowing enough space that all the 'pure-relevant-grobs' for the 
lower staff lie below the the 'pure-relevant-grobs' for the upper staff --- 
that is, enough space so the 'pure-heights' do not overlap.

So this concept of the 'pure-height' of the lower staff depends on the 
'pure-minimum-translation'  between two staves, which in turn depends on the 
'pure-height'  of the lower staff.


This is easier with a picture.  For issue 3385 could you outline the order of 
placement decisions, including tentative placements and/or pure placements, 
that you had in mind ?