lilypond-devel
[Top][All Lists]
Advanced

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

Re: bug with 2.15.39 fixed with 2.15.42: good job but no reg test


From: Keith OHara
Subject: Re: bug with 2.15.39 fixed with 2.15.42: good job but no reg test
Date: Tue, 10 Jul 2012 05:35:08 +0000 (UTC)
User-agent: Loom/3.14 (http://gmane.org/)

Frédéric Bron <frederic.bron <at> m4x.org> writes:

> 
> >> The following snippet does not give 1/32 notes grouped together as
> >> expected with 2.15.39:

> >>  2.15.42 works as expected.

This was an un-related side effect to fixing the problems with the rules
for ending beams mid-measure in 3/4.  Any exceptions not involving eighth-
notes were ignored.  No-one had reported this bug until now.

> As requested, I had a look a the regtests and:
> - only one test has beamExceptions different from empty #'():
> les-nereides.ly.

> Would you recommend adding a test or just make autobeam-3-4-rules.ly a
> bit longer?

The exact same bug, specific to 3/4 time, is not very likely to happen 
again.  Best to write a new regression test to check the function of
'beamExceptions' in general.

> If so, what should I add in addition to modifying the test file? any
> result of the test file compilation?

Add only the new test file, but check carefully that `make test` succeeds
with the new test included, and take care that the header is correct
and that no errors are generated, so that `make doc` will also succeed.




reply via email to

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