Guido's regrets: filter and map

[maney at pobox.com]

Robin Becker <robin at jessikat.fsnet.co.uk> wrote:
> In article <arqrso$b2j$1 at wheel2.two14.net>, maney at pobox.com writes
> ..... anything with a lambda is already doomed as the lambda time

Right, and that's the way I would normally have had to use them (before
list comprehensions came along)... and of course, that's why you're
asking for more non-lambda facilities.  The problem is that once you go
beyond a few basic operations, most or all of which have already been
implementated, there's a vast range of tests and translations, far too
many to implement them all in case they shuold be needed.

> dominates. As for the del I use it to remove the possibility that memory
> use from the first test will cause problems for the second etc

Of course.  I only remarked it because I thought it was a small error
that it was contained inside range of code that was timed, and then
because moving the del outside those ranges caused an easily visible
change for the map and comprehension test, but not for the explicitly
coded loop.