bug-lilypond
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: bug report


From: David Kastrup
Subject: Re: bug report
Date: Tue, 05 Nov 2013 18:34:03 +0100
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3.50 (gnu/linux)

Colin Campbell <address@hidden> writes:

> In looking over your files, I see you are making a lot of changes to
> spacing, particularly system-system values, and perhaps this is the
> cause of your problems starting around measure 89. Because there has
> been a lot of work done in the spacing mechanisms of LilyPond since
> version 2.16, and the latest development version 2.17.96 is very close
> to the next stable release, it would probably be a good idea for you
> to upgrade your LilyPond to 2.17.95, and to read the sections in the
> documentation which deal with spacing, quite carefully.

You can basically throw out all changes to the spacing variables without
a change in the essential problem.  The problem is that this example
contains wagonloads of cruft that has nothing to do with the issue
itself.

The issue looks like a close relative to
<URL:http://code.google.com/p/lilypond/issues/detail?id=3383>.

> As a final thought, and a friendly nod to the people who develop and
> support Frescobaldi: the latest version has the ability to show the
> various spacing values in a preview mode, among other, quite wonderful
> ways of inspecting your work. Frescobaldi has been my favourite
> LilyPond editor for years, and it just gets better and better at
> helping me find my problems.

Adding -ddebug-skylines to the commandline (probably something that
Frescobaldi does with some specific setting) shows that this example has
severely corrupted skylines.

Unfortunately, the example is so contorted and contains so much unused
and unnecessary stuff that it is quite unsuited for proper debugging.

The example already shows severely corrupted skylines in 2.16.2.  It
would be interesting to see whether the patch I proposed for issue 3383
(but which lead to infinite loops for some cases, and so was junked in
favor of a simpler patch by Keith) would make a difference here.

-- 
David Kastrup




reply via email to

[Prev in Thread] Current Thread [Next in Thread]