On May 28, 1:53 pm, Aaron Brady <castiro...@gmail.com> wrote: > On May 27, 11:07 pm, Steven D'Aprano <st...@remove-this- > > cybersource.com.au> wrote: > > On Wed, 27 May 2009 12:58:02 +0000, Albert van der Horst wrote: > > > >>And how is reduce() supposed to know whether or not some arbitrary > > >>function is commutative? > > > > Why would it or need it? A Python that understands the ``par'' keyword > > > is supposed to know it can play some tricks with optimizing reduce() if > > > the specific function is commutative. > > > Fine. Move the smarts out of reduce() into the compiler. How is the > > compiler supposed to know if an arbitrary function is commutative? > > Unit testing.
I think this kind of gets to the heart of it, here- parallelization is not a toy, runs with scissors, and will eat your dog, but so will everything else if you don't know what you're doing. I just don't see this as being a valid argument- maybe I'm wrong. In the spirit of continuing to be wrong, would it be possible to just fork off and synchronize the passed-in variables at the end of the block via a container in shared memory? That sounds to me like the simple, safe route, if a route must be taken. Geremy Condra -- http://mail.python.org/mailman/listinfo/python-list
