Thanks everybody! I've gone ahead and changed the list method to return a copy rather than a reference. This seemed like the most minimal change I could make.
Robert, how would I go about returning a list with copy-on-write semantics? Is there a standard container that does this? All the best, Joris On Wednesday, September 26, 2012 12:09:52 AM UTC+1, Robert Bradshaw wrote: > > +1 to a copy. There could be another method, say points(), that > returns an (immutable) tuple. Even more sophisticated, one could > return a list with copy-on-write semantics. > > On Tue, Sep 25, 2012 at 3:38 PM, mhampton <hamp...@gmail.com <javascript:>> > wrote: > > > > It seems safer to return a copy and not a reference. The documentation > does > > make it clear that it returns a reference, so it can't really be called > a > > bug. My guess is that there isn't any code out there that exploits that > > though, so I would vote to not bother with a deprecation warning. > > > > -Marshall Hampton > > > > > > On Tuesday, September 25, 2012 5:28:26 AM UTC-5, Joris Vankerschaver > wrote: > >> > >> Hi everybody, > >> > >> I made two small changes to the Sage interface for cubic spline > >> interpolation (found in sage/gsl/interpolation.pyx). The first is at > >> > >> http://trac.sagemath.org/sage_trac/ticket/13519 > >> > >> and concerns a small bug which prevents splines from being recomputed > when > >> the interpolation points are changed. The patch just adds two lines to > the > >> relevant methods, along with a bunch of doctests, and should be a > breeze to > >> review. > >> > >> The other patch is at > >> > >> http://trac.sagemath.org/sage_trac/ticket/13520 > >> > >> and simply exposes two methods from GSL for spline integration and > >> computation of the derivatives. > >> > >> One thing that I noticed when working on this patch is that Spline.list > >> returns a reference to the spline interpolation points (this is > actually > >> noted in the documentation), allowing one again to change the > interpolation > >> points without forcing the spline to be recomputed. This issue was > brought > >> up in #12036 when doctest coverage was addressed, and I've opened a > ticket > >> about this at > >> > >> http://trac.sagemath.org/sage_trac/ticket/13530 > >> > >> My question is, can I just change Spline.list() to return a copy of the > >> list of interpolation points, rather than a reference? Strictly > speaking, > >> this would alter the interface and so a DepreciationWarning of some > sort > >> would be appropriate but I can't imagine that anybody is relying on > this > >> (buggy) behavior. Still, maybe there are other reasons for not making > this > >> change -- please let me know if this is the case. > >> > >> With best wishes, > >> Joris Vankerschaver > >> > > -- > > You received this message because you are subscribed to the Google > Groups > > "sage-devel" group. > > To post to this group, send email to > > sage-...@googlegroups.com<javascript:>. > > > To unsubscribe from this group, send email to > > sage-devel+...@googlegroups.com <javascript:>. > > Visit this group at http://groups.google.com/group/sage-devel?hl=en. > > > > > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To post to this group, send email to sage-devel@googlegroups.com. To unsubscribe from this group, send email to sage-devel+unsubscr...@googlegroups.com. Visit this group at http://groups.google.com/group/sage-devel?hl=en.
