Hello, those libraries are very cool! I was unaware of their existence, thanks a lot for the pointers. My only worry is that these libraries as far as I can tell are "computational", not "symbolic", so we may lose some nice symbolic analysis if we don't reimplement the algorithms in terms of SAGE objects.
How is this generally dealt with? I assume other components of SAGE also face the dichotomy of (i) use external library that is fast and well tested, but computational (ii) implement within SAGE to be able to perform symbolic computation On Thursday, 23 July 2020 at 13:20:51 UTC+5:30 Thierry (sage-googlesucks@xxx) wrote: > Hi, > > this looks great. One important point that might be missing in the > discussion, especially if you do not have any code written yet, would be > to review existing optimized libraries, that could be interfaced within > Sage, instead of reinventing the wheel, see e.g. > https://doc.sagemath.org/html/en/tutorial/interfaces.html > https://doc.sagemath.org/html/en/thematic_tutorials/cython_interface.html > > To mention a few libraries of possible interest: > > https://www.cgal.org/ > https://polymake.org/ > https://dgtal.org/ > > Ciao, > Thierry > > > > > On Tue, Jul 21, 2020 at 10:23:14PM -0700, Siddharth Bhat wrote: > > Hi, > > > > I've posted two "meta-tickets" with list out the basic discrete > > differential geometric objects I would be interested to add first. I > tried > > to model it after the SageManifolds meta-ticket > > <https://trac.sagemath.org/ticket/18528> > > > > 1. Discrete Exterior calculus > > <https://trac.sagemath.org/ticket/30196#ticket>: > > https://trac.sagemath.org/ticket/30196#ticket > > 2. Discrete Meshes and duals < > https://trac.sagemath.org/ticket/30197#ticket>: > > https://trac.sagemath.org/ticket/30197#ticket > > > > I'm not sure I did this right :) So, should I: > > 1. break out _each_ of the bullet points into a separate ticket > > 2. start working on the "most basic" of them? (For example, let's say > > adding discrete k-forms) > > 3. Submit code, have it reviewed > > 4. Have code accepted > > 5. Pick new bullet point, goto step 2? > > > > Thanks a lot, > > ~Siddharth > > > > > > On Monday, 20 July 2020 15:05:51 UTC+5:30, Eric Gourgoulhon wrote: > > > > > > Hi, > > > > > > Le lundi 20 juillet 2020 01:17:39 UTC+2, Travis Scrimshaw a écrit : > > >> > > >> Hi Siddharth, > > >> That sounds like a good idea. > > >> > > > > > > +1 ! > > > > > > What you will need to do is create a number of tickets to add in the > > >> corresponding functionality. Once you have a proposal (which you can > ask > > >> questions on said tickets), it will be reviewed, where suggestions > and > > >> comments will be given on possible design decisions. > > >> > > > > > > If not done already, have a look at the Sage Developer's Guide > > > <https://doc.sagemath.org/html/en/developer/>. FWIW, a kind of > summary > > > adapted to smooth manifolds development is here > > > <https://sagemanifolds.obspm.fr/contrib.html>. > > > > > > Best wishes, > > > > > > Eric. > > > > > > > -- > > 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 view this discussion on the web visit > https://groups.google.com/d/msgid/sage-devel/854a399f-21d9-4ead-b70f-fab1fbadd4a0o%40googlegroups.com > . > > -- 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+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/426155c4-944f-41e8-8b13-e13e08efd4fbn%40googlegroups.com.
