On 27/08/14 15:57, Albert-Jan Roskam wrote:
if *fewer* than 1 in 10 operations will raise an exception, then use
try...except; but if *more* than 1 in 10 operations will raise an
exception, and it is safe to do so, then LBYL may be appropriate.
Thanks a lot, Steven! This kind of stuff should be in a book.
It is in lots of books but they aren't Python books; they are generic
software engineering books. Because this is not really a language
specific type design issue, its general software design.
> Do you know of any real life examples where the code is written
> in such a way that it could switch from AFTP to LBYL
It would be very unlikely since it would almost always be a case of
premature optimisation. You would need to be inside a very tight loop to
make that kind of optimisation worthwhile in real world scenarios.
(And the cost of evaluating it would probably outweigh any gains.)
Much more important is to make the code correct and maintainable than to
worry about slight performance issues before they are a proven issue.
Building those kind of self modifying control flows makes code much more
error prone and difficult to modify.
The vast majority of performance issues in software are to do with
blockages in IO processing(network, disk access, human input, etc)
or database processing (big data or complex searches) or in traversing
complex/deep data structures. Very few real-world performance issues
are related to simple code execution issues.
--
Alan G
Author of the Learn to Program web site
http://www.alan-g.me.uk/
http://www.flickr.com/photos/alangauldphotos
_______________________________________________
Tutor maillist - Tutor@python.org
To unsubscribe or change subscription options:
https://mail.python.org/mailman/listinfo/tutor
