On 2014-06-25 00:35, Wolfgang Maier wrote:
On 25.06.2014 00:55, Alex Kleider wrote:
I was surprised that the use of dictionaries was suggested, especially
since we were told there were many many files.
The OP was talking about several thousands of files, which is, of
course, too many for manual processing, but is far from an impressive
number of elements for a Python dictionary on any modern computer.
Dictionaries are fast and efficient and their memory consumption is a
factor you will have to think about only in extreme cases (and this is
definitely not one of them). What is more, your sequential approach of
always comparing a pair of elements hides the fact that you will still
have the filenames in memory as a list (at least this is what
os.listdir would return) and the difference between that and the
proposed dictionary is not that huge.
What's more important in my opinion is that while the two approaches
may look equally potent for the given example, the dictionary provides
more flexibility, i.e., the code is easier to adjust to new problems.
Think of the afore-mentioned situation that you could also have three
parts of a file instead of two. While your suggestion would have to be
rewritten almost from scratch, very little changes would be required
to the dictionary-based code.
Best,
Wolfgang
Thanks for elucidating this. I didn't know that "several thousand"
would still be considered a small number. If this is the case, then
certainly your points are well taken.
Gratefully,
alex
_______________________________________________
Tutor maillist - [email protected]
To unsubscribe or change subscription options:
https://mail.python.org/mailman/listinfo/tutor
