Re: Sundials is way outdated
Hi James,
welcome in the Debian Science team. ;-)
On Mon, Feb 06, 2017 at 01:25:54PM +1100, James Tocknell wrote:
> sundials-config was removed in sundials 2.6, so I'd just remove the
> references to it. There's a patch
> https://github.com/aragilar/debian-packaging-sundials/blob/master/debian/patches/Add-pkgconfig-files.patch
> which adds pkg-config support, as otherwise you have to know what the
> correct paths are.
It would be great if you would move this patch to the Debian Science git
since pkgconfig support is important.
> Questions about the current packaging: Why is
> 0001-Added-missing-linkages-to-the-DSOs.patch needed, shouldn't the
> necessary linking already be set up? Also, what's the need to override
> dh_install for the docs (see
> https://anonscm.debian.org/cgit/debian-science/packages/sundials.git/tree/debian/rules#n29),
> shouldn't it just work?
I'd be happy if any unneeded patch would be removed.
> Do we want to rename parallel to mpi in the packaging, to reduce confusion
> (as there is now two other parallel methods)?
+1
> Also, do we want to keep the current "tests", given 1) unrelieable, and 2)
> only cover the C serial version (the Fortran and parallel versions appear
> to be commented out, presumably because they keep failing)? Wouldn't be
> better to write a simple set of tests which solve (for example) the simple
> harmonic oscillator, and which run against the different features sundials
> has (c.f. https://github.com/bmcage/odes/tree/master/scikits/odes/tests)? I
> can do that when I get back home.
Lets do a sensible test suite. There is no point in fighting
unrelieable tests but it might make sense to point out this to upstream
if the tests they provide might fail randomly.
Kind regards
Andreas.
--
http://fam-tille.de
Reply to: