On Wed, Jul 9, 2008 at 8:17 PM, Alan Gauld <[EMAIL PROTECTED]> wrote: > No I meant in syntactic terms. > We usually define an LC as > > [ expr for vars in sequence if expr ] > > or somesuch imprecise gobbledy gook ;-). > > Now we can define the generator expr (syntax) as > > expr for vars in sequence if expr > and the LC as > > [ gen expr ]
The gen exp needs the parens. You could possibly have an intermediate term that can be put inside () or []. The actual formal syntax definitions for the two are slightly different: http://docs.python.org/ref/lists.html http://docs.python.org/ref/genexpr.html Presumably this means there is something that is syntactically allowed in one form and not the other, but I can't figure out what it might be. >> <code> >> squares = (x * x for x in range (10)) >> l = [squares] > > But doesn't that generate a tuple (because of the parens)? No, the parens are required for, and create, a generator expression. In [23]: squares = (x * x for x in range (10)) In [24]: squares Out[24]: <generator object at 0x126ae90> Kent _______________________________________________ Tutor maillist - Tutor@python.org http://mail.python.org/mailman/listinfo/tutor
