Hmm. I'm running a local Linux build of 2.24.0 (commit 916ae4111441a9426fbb9c11f7c42735eccf0cac). My configure command was just running autogen.sh with --prefix. Does that make a difference?
--T On Fri, Dec 16, 2022 at 07:31:46PM +0000, Paul Hodges wrote: > Your snippet compiles correctly (with the errors) on my Windows > install of 2.24.0 > Paul > > From: H. S. Teoh via LilyPond user discussion <lilypond-user@gnu.org> > To: <lilypond-user@gnu.org> > Sent: 16/12/2022 18:33 > Subject: Re: LilyPond 2.24.0 released! > > On Thu, Dec 15, 2022 at 10:42:42PM +0100, Jonas Hahnfeld via LilyPond > user discussion wrote: > > We are proud to announce the release of GNU LilyPond 2.24.0. > [...] > > This latest release broke one of my scores. I have reduced the problem > to the following snippet: > > \version "2.18.2" > \score { > \tuplet 3/2 4 { \repeat tremolo 6 { <cis ais g e>8( } } > } > > Lilypond produces the following error: > > Processing `2.ly' > Parsing... > Interpreting music... > warning: omitting tuplet bracket with neither left nor right bound > 2.ly:3:55: warning: unterminated slur > \tuplet 3/2 4 { \repeat tremolo 6 { <cis ais g e>8 > ( } } > Preprocessing graphical objects... > /usr/src/lilypond/share/lilypond/2.25.0/ly/init.ly:65:2: error: > Guile signaled an error for the expression beginning here > # > (let ((book-handler (if (defined? 'default-toplevel-book-handler) > In procedure ly:grob-array-ref: Wrong type argument in position 1 > (expecting Grob_array): () > > I have narrowed down the offending commit using git bisect to the > following: > > commit fd92bef058ee0fbbdaafdee0cc056316f6d4b9f9 (HEAD, > refs/bisect/bad) > Author: Jean Abou Samra <j...@abou-samra.fr> > Date: Tue Oct 25 00:48:09 2022 +0200 > > Move Tuplet_bracket::parallel_beam to objects callbacks > > Not only does this avoid redoing the same computation several > times, but it > makes what happens much clearer, because previously, different > calls to > Tuplet_bracket::parallel_beam would non-obviously return > different values > depending on the caller callback since some of the callers are > executed before > line breaking and some after. > > Prior to this commit, the original score compiled correctly, albeit with > the same warnings about omitting tuplet brackets. The output is correct > (a slashed stem with the tuplet number written above the chord). The > current release, however, does not produce any output because lilypond > aborts after displaying the above error message. > > How do I obtain the same output without triggering this error? > > --T -- Recently, our IT department hired a bug-fix engineer. He used to work for Volkswagen.