Re: Fwd: 2.17.6: assertion failed with \glissando

[Ian Hulin]

Hi Colin, Keith and Walter,
Sorry to double-post, but I replied on the -user list by mistake!

On 22/11/12 09:22, Colin Hall wrote:
> 
> Hi,
> 
> Werner Lemberg reported an assertion failure on his own build of
> 2.17.6 here:
> 
> http://lists.gnu.org/archive/html/bug-lilypond/2012-11/msg00047.html
>
>  I'm forwarding this to the devs in case it is of interest.
> 
> Cheers, Colin.
> 
> 
> ----- Forwarded message from Colin Hall <address@hidden>
> -----
> 
> Date: Thu, 22 Nov 2012 09:19:45 +0000 From: Colin Hall
> <address@hidden> To: Werner Lemberg <address@hidden> Cc:
> address@hidden Subject: Re: 2.17.6: assertion failed with
> \glissando
> 
> On Wed, Nov 21, 2012 at 05:08:52AM +0000, Keith OHara wrote:
>> Werner LEMBERG <wl <at> gnu.org> writes:
>> 
>>> This time I tried with current git (commit 436824b8):
>>> 
>>>> ../flower/include/interval.hh:226: T Interval_t<T>::center()
>>>> const [with T = double]: Assertion `!is_empty ()' failed.
> 
> Thanks for the report, Werner.
> 
>>> I'm on a GNU/Linux box, running `autogen.sh' with 
>>> `--disable-optimising', and compiling with g++ 4.6.2.
>>> 
>> 
>> The released builds are not compiled this way, so no-one can
>> reproduce your bug without using Linux to create a debug-enabled
>> build.  I would guess that readers of bug-lilypond are more
>> interested in helping to record bugs that affect released builds
>> (I am , at least).
> 
> Keith is correct. This is a bug report against unreleased code so 
> we can't create a tracker for it.
> 

Colin, you're right to tell the -user list we shouldn't raise a
tracker against the releases prior to current development, but it
looks like we have got a problem in the development cycle.

> I'm forwarding this thread to lilypond-devel as it may be of
> interest to the devs.
> 
Werner, if this is reproducible it will need a tracker and patch and
everything to go through the review process.  (I realize you probably
know this).  Just want to make sure all your sleuthing doesn't get lost.

Cheers,

Ian