Re: Broken Forbid_line_break_engraver with Completion_heads

[Trevor Daniels]

Carl

Why do you want a tied note to forbid a line break?
It is preferable that it permits it, surely?

Trevor

----- Original Message ----- 
From: "Carl D. Sorensen" <address@hidden>
To: "LilyPond Development" <address@hidden>
Cc: "Han-Wen Nienhuys" <address@hidden>
Sent: Friday, August 15, 2008 3:07 AM
Subject: Broken Forbid_line_break_engraver with Completion_heads


The code below

%%%%%
\paper{
 ragged-right=##f
 line-width=3\cm
 indent=0\cm
}

\new Voice \with {
\remove "Note_heads_engraver"
\consists "Completion_heads_engraver"
%\remove "Forbid_line_break_engraver"
}
\relative c'  {
 e4 e e e2 % Line break should be forbidden here, but it's not
 f4 f f
}

%%%%%%%%%%%

caused both measures to be compressed onto one line up until 2.11.45-1.

In all later releases, each measure gets its own line, which I think the
Forbid_line_break_engraver should prevent.

The behavior changed at commit 0030e7e6ab44b0050238ba2e5e220b184cde70b,
when completionBusy was added to the Melisma and the "length of the
generated note head" was adjusted so the "busyGrobs list is correctly
calculated".

I think that this means there is now no busyGrob crossing the measure
boundary, to Forbid_line_break_engraver is free to break.

I suspect that the fix is to have Forbid_line_break_engraver check both
busyGrobs and completionBusy, but I don't know the code well enough to make
this change.

The commit that changed the behavior was made by Han-Wen.

If you'd like, I'll file this as a bug report.

Thanks,

Carl



_______________________________________________
lilypond-devel mailing list
address@hidden
http://lists.gnu.org/mailman/listinfo/lilypond-devel