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

Re: Sagemath: going further



Hi,

Le 03/02/2015 22:56, Tobias Hansen a écrit :
Am 03.02.2015 um 21:09 schrieb Julien Puydt:
Hi,

Le 03/02/2015 17:25, Tobias Hansen a écrit :
Hi Julien,

I like that the patch is still quite small. Are you aware of the similar
patches Fedora is using?
http://pkgs.fedoraproject.org/cgit/sagemath.git/tree/

No... perhaps I should have a look.

The sagemath.spec is also interesting. It has short comments for the
patches, lists the build and runtime dependencies etc.


I'll see if I get around to setting up a packaging git repo where the
sage repo is included as a git submodule and your patch is applied via
quilt. I think that way the packaging can be handled quite nicely. What
do you think? I just need you to put the commands for building into
debian/rules then. (Or if you want you can also set up everything.)

I don't feel confortable with packaging sage myself, so my goal is to
grease things around so when someone gets to that work, things will go
smoothly.

This is not about packaging, it's about doing things in git. For
example, I applied your patch and ran 'make'. It starts to download and
build patch, pkgconfig, bzip2. Now I think: these should be taken from
Debian, am I doing something wrong or is it supposed to be like that? I
have to ask you. If there's a debian rules build target I know that I'm
building with the same command as you.

I created the repository at

http://anonscm.debian.org/cgit/debian-science/packages/sagemath.git/

There are some useful commands mentioned in debian/README.Debian

Please consider using it (updating the patch) and putting the build
command into debian/rules (is it just make?)

Again I'm not even considering letting this produce a Sage package yet,
and even if we do, it can't do any harm unless it's uploaded to Debian.

I honestly don't want to play with git branches at that point : I still edit my patch by hands at times :-P

Still, it would be nice if my results could be reproduced, so I'll probably write a long mail to this list to explain what I have done.

Snark on #debian-science


Reply to: