certainly, I can relicense whatever was under v3 to v2, if needed. Dima On Dec 14, 10:00 pm, Keshav Kini <keshav.k...@gmail.com> wrote: > Hi Minh, > > On Dec 14, 9:07 pm, Minh Nguyen <nguyenmi...@gmail.com> wrote: > > > > > > > [...] > > Your patch adds the following source files to the Sage library: > > > * STABIL.c > > * STABIL.h > > > both of which are explicitly licensed under the terms of GPLv3. A > > policy of the Sage project is to include only code licensed under the > > terms of the GPLv2+ in the Sage library. That means that any code that > > goes into the Sage library must be covered by the GPL version 2 or any > > later version, but not explicitly version 3. Here are some options I > > can think of to remedy the situation: > > > (1) Relicense your C source files under the terms of the GPLv2+, after > > which you should be able to get your C source files into the Sage > > library. > > That would be fine with me, but the original implementation by Dmitrii > Pasechnik and Luitpold Babel is licensed under GPLv3+, and while my > code doesn't technically use any lines from the original > implementation, it borrows heavily from the organizational structure > (after all, it uses the same algorithm), so if I understand correctly > it would be improper to relicense my own code, even if the original > authors permitted it - even the original authors cannot now relicense > their GPLv3+ code as GPLv2. (Please correct me if this understanding > is incorrect.) > > > (2) Create a Sage package (also known as an spkg) and write an > > interface to your spkg. Anything in the interface that is licensed > > under GPLv2+ can be included in the Sage library. For everything else, > > you need to include them in your spkg. See this [1] from sage-devel on > > options for packaging your code so that the spkg can be distributed > > with Sage. > > I actually have created an spkg as well, which works fine. However, > distributing a mere two functions consisting of less than 40 KB of > code seemed like sort of a waste, especially considering that the user > would then need to import a base-level package "wlrefine" instead of > immediately being able to use it. Or is that what you meant by > "writing an interface to" the the spkg? Just putting a .py file in > $SAGE_ROOT/devel/sage/sage/graphs/ that loads the module from the > spkg? I notice that you didn't list the .pyx file the patch also adds > (wlrefine.pyx) - I guess that could be called an "interface" to the C > code. Is that what you mean? > > > > There is a comment in STABIL.c which says something about line > > > continuation and doctests, but this should hopefully be fixed by my > > > patch at trac #10458 (http://trac.sagemath.org/sage_trac/ticket/10458 > > > ). > > > See the following sage-devel thread for discussion on this matter: > > >https://groups.google.com/group/sage-devel/browse_thread/thread/d6125... > > Yes, thanks for publicizing my trac ticket! :) > > Yours, > Keshav
-- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
