On Tuesday, March 1, 2016 at 5:23:36 PM UTC-8, kcrisman wrote: > > If I recall correctly, Jason Grout's example for this is > > x+y-y > > which becomes x but in principle had two variables, so how do you know > which one is "right"? >
Hm, I'm not sure that's a problem here. Indeed if both are defined then one would hope that f1(a)+f2(a)+f3(a) = (f1+f2+f3)(a) which is still OK in your example. If one of them errors out then there's not a direct problem. Anyway, I'm -1 on supporting calling expressions with one variable (and one argument) and/or supporting calling expressions with an empty argument list (which would also be unambiguous) and/or calling constant expressions with no arguments at all. It's easy to explain "expressions do not support calling syntax. You need a function for that". It's hard to explain "expressions don't really support calling, except in some edge cases that you will initially encounter most". By having the muddling exceptions, you make it harder for novice users to start learning the difference between a symbolic expression and a symbolic function; a difference they'll need later anyway. If you don't want to talk about functions at all, just use "subs". -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
