I've removed the BiConsumer interface in NUMBERS-166. I also renamed the "accept" method to "apply" to better indicate its active role in modifying the arguments.
If there aren't any other changes required, I'll plan on starting the release tonight or tomorrow sometime. To confirm, all modules will be released with the exception of commons-numbers-complex-streams. Regards, Matt J On Sat, Jul 10, 2021 at 7:17 AM Alex Herbert <alex.d.herb...@gmail.com> wrote: > On Sat, 10 Jul 2021 at 09:19, Gilles Sadowski <gillese...@gmail.com> > wrote: > > > Le ven. 9 juil. 2021 à 16:21, Alex Herbert <alex.d.herb...@gmail.com> a > > écrit : > > > > > > [...] > > > SortInPlace > > > > > > The latter was recently ported from Commons Math. For that class there > > is a > > > warning about using varargs for the signature of BiConsumer<double[], > > > double[][]>. It is using accept(double[], double[]...). I am not sure > if > > > this is a real problem or not. The varargs is not allowed in the > > BiConsumer > > > signature. So perhaps it should be removed from the implementation of > the > > > method. > > > > The varargs API is handy for sorting related arrays without the user > having > > to create a temporary array of arrays. > > > > I don't also know whether the warning (I didn't notice it) is a problem, > > but > > if it is, I'd rather remove the "implements" clause than the method. [I > > don't > > have a real-life example for using the interface.] > > > > +1 to removing the interface. The vargargs has more benefit than the > interface declaration. > > Alex >
