Re: music starting with afterGrace breaks layout

[David Kastrup]

David Kastrup <address@hidden> writes:

> Thomas Morley <address@hidden> writes:
>
>> 2017-02-22 18:44 GMT+01:00 Klaus Rettinghaus
>> <address@hidden>:
>>> Hi there,
>>>
>>> when the music starts with an \afterGrace the layout is broken:
>>>
>>> \relative { \time 2/4 \afterGrace g'2 { f8[ g] } f2 }
>>>
>>>
>>> You have to put something before that to make it work:
>>> \relative { \time 2/4 s2*0 \afterGrace g'2 { f8[ g] } f2 }
>>>
>>> This is present in 2.18.2. and 2.19.55 (Mac).
>>>
>>> Cheers,
>>> Klaus
>>
>> Hi Klaus,
>>
>> thanks for the report.
>>
>> \afterGrace returns simultaneous music.
>> If no context exists before, LilyPond creates them. In the case of
>> afterGrace Staves for each musical expression in the created list.
>> Another example would be: << c c >>
>>
>> So I don't think it qualifies as a bug, but expected (though
>> undesired) behaviour. Not sure whether it's fixable in the
>> afterGrace-definition.
>
> Easy enough: wrap a (descend-to-context ... 'Bottom) around its
> definition.  However, this breaks \displayLilyMusic in bewildering ways.
>
> Still trying to figure out what's happening there.

Tracker issue: 5079 (https://sourceforge.net/p/testlilyissues/issues/5079/)
Rietveld issue: 314590043 (https://codereview.appspot.com/314590043)
Issue description:
  Let \afterGrace start a Bottom context  Previously, an \afterGrace
  as the first expression of music outside of an existing Bottom
  context caused quite weird results.   Also contains commits:  with-
  music-match: lists should match completely  Previously music
  properties that were matched to a list of wildcards could contain
  fewer elements than the list they were matched to, leading to an
  error when the wildcard variable was being assigned to.  Now the
  list lengths have to be matched exactly: a shorter pattern than the
  actual list length of the matched property cannot match any more.
  Fix \displayLilyMusic \partcombine...  Strictly speaking this
  oversight in issue 4356 did not actually manifest itself as a bug
  because of another bug masking it.

-- 
David Kastrup