Re: Cross-staff and skyline work

[David Kastrup]

Keith OHara <address@hidden> writes:

> mike <at> mikesolomon.org <mike <at> mikesolomon.org> writes:
>
>> Just a quick note to say that I'm eye-deep in work but I'm reading the 
> LilyPond list every other day.  I don't
>> want anything related to my work to get bogged down by my absence.  If you 
> have specific questions about how
>> something works (or doesn't), don't hesitate to send me an e-mail and I'll 
> do my best to respond.
>> 
>> I'll be back around the 28th.
>
> There are just the two outstanding issues so far from 
> "Uses only unpure-pure containers to articulate unpure-pure relationships"
>
> I'm having trouble finding how it cause the crash with issue 3359,
> and having trouble getting a clean make check after reverting your patch.
> Probably I'll successfully revert it before you are back.

On another side note: I am getting a rather obstinate visual variation
in tie length in laissez-vibrer-tie-beam depending on "unknown
circumstances", namely code changes not really related.  I could imagine
this to be some circular dependency.  At the end of scm/output-lib.scm
there is

;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
;;  laissez-vibrer tie
;;
;;  needed so we can make laissez-vibrer a pure print
;;
(define-public (laissez-vibrer::print grob)
 (ly:tie::print grob))

but it would not appear like the current code actually does anything any
more that would serve to establish a difference between
laissez-vibrer::print and ly:tie::print, and I suspect that this might
cause the oscillating regtest.

-- 
David Kastrup