On Thu, 23 Jul 2009 18:29:14 -0400 George Sakkis <george.sakkis@gmail.com> wrote:
Overall it seems like a non-solution to a non-problem.
It's definitely a real solution. The problem may not be very important, though.
Why don't we add a dict.universal(), list.universal(), etc. while we're at it.
Because a set of "all the things in the universe" makes sense as an entity, whereas dict's or lists of "all the things in the universe" don't. For lists, this implies some sort of ordering on everything in the universe, which we've already abandoned as a bad idea. Dicts, on the other hand, aren't simple collections, but maps, so you don't have _a_ dict of everything in the universe, you have len(universe)**2 of them. The def statement gives us the ability to express such maps. Which may be how set.universal() stacks up: We can already create a class that has the appropriate behaviors. It's not clear that that is sufficiently useful to justify adding it to the language. One thing that bugs me is the universal set is a *constant. It's value doesn't change, so it shouldn't be a function or method, but an attribute. On the other hand, adding negative sets to the language provides an easy way to express that constant, and certainly have more uses in general. If we had to add one of the two, it'd be these. But I'm not convinced that these have enough uses to justify adding them, either. <mike -- Mike Meyer <mwm@mired.org> http://www.mired.org/consulting.html Independent Network/Unix/Perforce consultant, email for more information. O< ascii ribbon campaign - stop html mail - www.asciiribbon.org