Getting values out of a CSV

Steve Holden steve at holdenweb.com
Sun Jul 15 08:00:06 EDT 2007


Alex Popescu wrote:
> On Jul 14, 5:55 am, "Gabriel Genellina" <gagsl-... at yahoo.com.ar>
> wrote:
>> So, as always, one should measure in each specific case if optimization is  
>> worth the pain [...].
>>
> 
> I hope I am somehow misreading the above sentence :-). IMO synonim
> language contructs
> should result in the same performance or at least have clear/
> documented performance.

That's a fine opinion, how would you enforce it? Should we go throught 
he interpreter slowing down the faster to each pair of alternative 
constructs? ;-) It's inevitable there'll be differences in execution 
time between equivalent constructs, and in that case you have to test to 
find the better in your specific situation.

The real issue here is that in 95% or more of the source of most 
programs speed/performance isn't that much of an issue anyway.

> I don't think we really want to see in code something like:
> 
> if threshold:
>   do_it_with_list_function
> else:
>   do_it_with_list_comprehension
> 
This would most certainly be a premature optimization which, as has been 
repeated many times on this list, is the root of much evil in 
programming. As Gabriel mentioned, you only need to do it if it's "worth 
the pain", which in most case it won't be. It isn't worth spending even 
five minutes to shave a minute off the performance of a ten-minute 
program that is only run once a week, for example.

Ultimately we have to be pragmatic: circumstances alter cases, and it's 
usually not worth spending the time to improve execution speed except 
for the most critical parts (the innermost nested loops) of production 
programs.

regards
  Steve
-- 
Steve Holden        +1 571 484 6266   +1 800 494 3119
Holden Web LLC/Ltd           http://www.holdenweb.com
Skype: holdenweb      http://del.icio.us/steve.holden
--------------- Asciimercial ------------------
Get on the web: Blog, lens and tag the Internet
Many services currently offer free registration
----------- Thank You for Reading -------------




More information about the Python-list mailing list