Question

Two questions about speed/efficiency/best practices in Python. Which of the following are ‘better’ (faster, less memory-intensive, etc.):

1.  

   for x in list:
       #do something to x
   

   or

   for x in xrange(len(list)):
       #do something to list[x]
   

2.  

   for string in list_of_strings:
       for string2 in other_string_list:
           if string == string2:
               #do something
   

   or

   import re
   for string in list_of_strings:
       if re.match('%s'%(string),other_strings): #or re.search(etc)
            #do something
   

Not terribly pressing, I’m mostly just curious. I guess I could get some kind of raw data using timeit() or whatever, but I’d appreciate more depth than just “this one is faster than that one on your computer.”

Answer 1

You can’t really compare these.

for x in mylist:
    # do something to x

is the usual idiom, but whatever you do to x won’t affect mylist (unless x is mutable). If your goal is to modify mylist during the iteration, then

for x in xrange(len(list)):
    #do something to list[x]

is almost always bad form. A better way would be to use

for i, x in enumerate(mylist):
    # now you can work with x and/or change mylist[i] if you need to

But usually, you’re even better off using a list comprehension or a generator expression:

newlist = [foo(item) for item in mylist if bar(item)]

All that depends on your use case.

As for your second question, using a regex for a plain string equality comparison is overkill. Nesting two for loops is horrible, too:

for string in one_list:
    if string in other_list:
        # do something

would be a little better, but I’m pretty sure that that can be improved if, again, more is known about your actual use case.