Sage package name convention

33 views
Skip to first unread message

Thomas Feulner

unread,
Oct 23, 2013, 3:57:46 AM10/23/13
to sage-...@googlegroups.com
I have produced a new Sage package (http://www.algorithm.uni-bayreuth.de/de/team/Feulner_Thomas/codecan-1_0_spkg.zip) which applies to Sage 5.11.
Now, for the current version 5.12 of Sage the source code results in some runtime errors, which I would like to fix (there are only a few lines I have to change).

I am not sure about the naming conventions. Should I bump up the version number of my spkg to 1.1 or should I patch my changes, which will result in codecan-1.0.p1.spkg?

In general, do I have to (somehow) test the Sage version which is used for installing my package?

Thomas
 

David Joyner

unread,
Oct 23, 2013, 7:09:19 AM10/23/13
to sage-devel
On Wed, Oct 23, 2013 at 3:57 AM, Thomas Feulner
<thomas....@uni-bayreuth.de> wrote:
> I have produced a new Sage package
> (http://www.algorithm.uni-bayreuth.de/de/team/Feulner_Thomas/codecan-1_0_spkg.zip)
> which applies to Sage 5.11.
> Now, for the current version 5.12 of Sage the source code results in some
> runtime errors, which I would like to fix (there are only a few lines I have
> to change).
>
> I am not sure about the naming conventions. Should I bump up the version
> number of my spkg to 1.1 or should I patch my changes, which will result in
> codecan-1.0.p1.spkg?

If you mean replace the patch with a new patch, then bumping to 1.1 is fine,
IMHO.


>
> In general, do I have to (somehow) test the Sage version which is used for
> installing my package?

One way to do it might be this:
You can install the latest, which now is 5.13.b1 I think, test it on your system
using sage -testall. If it tests clean, install your patch to a clone, and
run sage -b then sage -testall. If again it tests clean, then just
replace the patch
and say it was tested against 5.13.b1 on the ticket.

If you wish, while you are at it feel free to add comparisons with
your code and the
existing Sage code which was basically written by Robert Miller (in
the binary case).
That not only gives an extra test but also might help potential users
understand your syntax better.

>
> Thomas
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-devel+...@googlegroups.com.
> To post to this group, send email to sage-...@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-devel.
> For more options, visit https://groups.google.com/groups/opt_out.
Reply all
Reply to author
Forward
0 new messages