#timeit

How to: Why is it slower to iterate over a small string than a small list?

Why is it slower to iterate over a small string than a small list?

I was playing around with timeit and noticed that doing a simple list comprehension over a small string took longer than doing the same operation on a list of small single character strings. Any explanation? It’s almost 1.35 times as much time.

>>> from timeit import timeit >>> timeit("[x for x in 'abc']") 2.0691067844831528 >>>…

I recently wrote a script that I wanted to save the data on exit, and then read it in again. This is pretty trivial to do - use pickle everyone said. And I did.

But then I realised that I might want to quickly edit this data and restart the script. Using pickle makes this a bit more difficult - I would have to import the pickle file into a python shell, make the edit and save it again. What I really want to be able to do is edit the file directly... enter JSON.

So JSON needs no introduction, but that is a little frustrating about it is that there is no timestamp support. I found this out pretty quickly by running a simple json.dumps(my_dict) on my data and getting the following error:

TypeError: datetime is not JSON serializable

So after a quick Google search it turns out the standard JSON library in Python supports custom encoding and decoding. Simple I thought... but then I spent the next hour trawling through different implementation of what I wanted and no advise on what was best.

So I did what any good Python programmer would do - get all the possible solutions from the net, pretty them up and run them through timeit.