On Fri, 31 Aug 2012, Julien Cristau wrote:
> > While building a squeeze backport of new release of python-brian package
> > I just found out that there is a g++ 4.4 compatibility issue with weave
> > of scipy shipped in squeeze. Fix is trivial -- I adopted it from a
> > package in Ubuntu (links are in the patch). I tested that it
> > resolves my brian build failures. See attached the debdiff for a
> > proposed upload. Would it be acceptable? Thank you in advance
> Does that affect any package in squeeze?
not AFAIK (I cannot know for sure since many packages lack build-time
unittesting so even if package uses affected inline weave'ing I would
not know). otherwise we would have spotted it by now ;)
> If not then I don't think this
> is required, as it can easily enough be worked around in users?
not sure if easily since the majority of magic is hidden away from
"users", e.g. actual code would look like:
(Pdb) print code
int numspikes=0;
for(int i=0;i<N;i++)
if(V(i)>Vt)
spikes(numspikes++) = i;
return_val = numspikes;
and then bug shows itself up via using blitz "typeconverter" which
shipped with scipy aren't g++ 4.4 compatible.
may be there is a 'user-level' workaround but I am not aware of it... anyone?
Altogether acceptance of the patch would not fix any FTBFS in the stable
archive but should make few scipy users of 'stable' happier. Due to
minimalistic patch, I do not think that it should have negative side-effects.
--
Yaroslav O. Halchenko
Postdoctoral Fellow, Department of Psychological and Brain Sciences
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834 Fax: +1 (603) 646-1419
WWW: http://www.linkedin.com/in/yarik
--
To UNSUBSCRIBE, email to debian-release-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120831191911.gv2...@onerussian.com
