RetroSearch Browse

Mon Mar 8 15:45:30 EST 2004 · http://mail.python.org/pipermail/python-dev/2004-March/042961.html

On Monday 08 March 2004 03:05 pm, Skip Montanaro wrote:
 > Must take a single argument, which itself must be a callable,
 > right?

I wrote:
 > I'd expect w2() to be passed whatever w1() returns, regardless of
 > whether it's callable.  It should raise an exception if it gets
 > something it can't handle.

On Monday 08 March 2004 03:27 pm, Skip Montanaro wrote:
 > Yes.  I was thinking of the case where we wanted it to return something
 > useful which could be bound to the name "foo".  I suppose if you've had

And I certainly expect that's the typical case; I was mostly reacting to your 
use of the word "must" rather than the idea.  When I read "must", that tells 
me someone is going to check that in the mechanism rather than just passing 
it on.

 > too much caffeine you could dream up a case where w1() returns an AST
 > based on the original foo and w2() does something with it to cook up a new
 > object, but I suspect that would be pretty rare.

Defining foo to some useful object doesn't imply that it's callable, or that 
it can't be further transformed in useful ways.

  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at acm.org>
PythonLabs at Zope Corporation

Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from http://mail.python.org/pipermail/python-dev/2004-March/042961.html below:

[Python-Dev] PEP 318 - generality of list; restrictions on elements