Two questions about efficiency

[Fredrik Lundh]

"Steve M" wrote:

> 2. Is there any reason to prefer one of the following over the other?
> Im interested in both a consensus on style and readability, and
> considerations of efficiency at runtime.
>
> for x in L:
>   if f(x):
>     pass
>
> for x in L:
>   if f(x):
>     continue

did you really mean to write that?

as written, both constructs are pointless, and the "if f(x)"-statement can
be replaced with just "f(x)" in both cases.  and if you add more code inside
the if statement, or after it, the constructs do different things.

did you perhaps mean:

    for x in L:
        if f(x):
            do stuff

    for x in L:
        if not f(x):
            continue
        do stuff

if so, my rule of thumb is to use alternative 1 if "stuff" is short, and alter-
native 2 if "stuff" is long, or if you have multiple tests before you get to
the real stuff.

ymmv, as usual.

and yes, if "stuff" is trivial, and builds a new list, consider using a list
comprehension:

    L2 = [stuff for x in L if f(x)]

</F>