On Tue, Jul 23, 2013 at 12:07 PM, Joachim Durchholz <[email protected]> wrote: > What about the other outstanding decisions? > > Do we need the ability to assign to lazy class attributes? > I think probably no, but I may be missing something. > > Do we need the ability to reset a lazy class attribute? > This would be useful to cut down on memory usage. It would require something > where some large constants are needed just once and never again, AND if the > calculation is memory-constrained, AND the user knows what he's doing... > okay, doesn't sound too likely, but again I don't have a very good overview > over the set of common use cases.
If the attributes are only cached once per class, then it's not a big deal. If it's once per instance, then that's another story. > > I'm so anal about these questions because if I don't have answers, I might > find myself with several weeks sunk into a technique that works only in 95% > of use cases. > > Am 23.07.2013 17:50, schrieb Aaron Meurer: > >> You shouldn't have to do anything with mpmath here. I would just >> ignore it completely. > > > Can't say I'm surprised :-) ... but I *had* to ask. > > What about the other oldstyle classes? > They're relevant only > - IF they have class or instance attributes > - AND might be part of a cyclic dependency > - AND changing them to newstyle classes means trouble. I think these old style classes are old style either because whoever wrote them didn't know to subclass from object, or because there is some kind of performance boost by doing so (I think). I don't see any reason to keep them around, especially considering that in Python 3, all classes are new style. > > I can slap on that (object) suffix to their declaration and see what the > unit tests do, but the unit tests take 40,000 seconds and maybe somebody > already knows about this. Yeah, just do it. I doubt any tests will fail, or that their running time will be different, but who knows. If you don't like running the tests, create a pull request with your work so far and let Travis do it. Aaron Meurer > >>> assumptions.ask.Q >>> core.multidimensional.vectorize >>> core.mul.NC_Marker >>> >>> ntheory.generate.Sieve >>> parsing.maxima.MaximaHelpers >>> parsing.sympy_tokenize.Untokenizer >>> plotting.pygletplot.plot.ScreenShot >>> series.gruntz.Node >>> utilities.autowrap.CodeWrapper > > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at http://groups.google.com/group/sympy. > For more options, visit https://groups.google.com/groups/opt_out. > > -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy. For more options, visit https://groups.google.com/groups/opt_out.
