3

In coding a primality tester, I came across an interesting thought. When you want to do something if the result of an operation turns out to be 0, which is the better ('pythonic') way of tackling it?

# option A - comparison
if a % b == 0:
    print('a is divisible by b')

# option B - is operator
if a % b is 0:
    print('a is divisible by b')

# option C - boolean not
if not a % b:
    print('a is divisible by b')

PEP 8 says that comparisons to singletons like None should be done with the is operator. It also says that checking for empty sequences should use not, and not to compare boolean values with == or is. However, it doesn't mention anything about checking for a 0 as a result.

So which option should I use?

Improve this question
2
  • 1
    I prefer option C, though I don't know whether it is (more) Pythonic Commented Jan 18, 2013 at 9:37
  • I prefer A, more readable. Looking at Python 3.3 source code, I count 362 times the string " == 0:", and never " is 0:". More difficult to check automatically for option C, but I suspect option A is preferred. Commented Jan 18, 2013 at 9:44

4 Answers 4

Reset to default
6

Testing against 0 is (imo) best done by testing against 0. This also indicates that there might be other values than just 0 and 1.

If the called function really only returns 0 on success and 1 on fail to say Yes/No, Success/Failure, True/False, etc., then I think the function is the problem and should (if applicable) be fixed to return True and False instead.

Improve this answer
Sign up to request clarification or add additional context in comments.

3 Comments

0and 1 (or another int) as return values are very common when executing external programs, e.g., subprocess.call has exactly these return values.
@ThorstenKranz: Be that as it may, it's nice to be explicit wherever possible, especially if you have control over the return values. Returning True or False tells whoever calls the function that it only returns 2 values. 0 and 1 are a bit more ambiguous.
Sure, I didn't want to say "use 1 and 0". I literally hate statements like while 1:. We're in the 21st century, and we have bool-type. Just wanted to point out that there are situations were the return values are 0 and 1.
1

just personal : I prefer the not a % b way because it seems to be highly readable. But now, to lower the confusion level in the code, I will use the == 0, as it express what you expect to test exactly in a more accurate way. It's the "care for debug" approach.

Improve this answer

Comments

1

0 isn't guaranteed to be a singleton so don't use is to test against it: currently C Python re-uses small integers so there is probably only one int with the value 0 plus a long if you're still on Python 2.x, and any number of float zeroes not to mention False which all compare equal to 0. Some earlier versions of Python, before it got a separate bool type used a different int zero for the result of comparisons.

Use either == (which would be my preference) or just the not, whichever you prefer.

Improve this answer

Comments

1

A and C are both valid and very pythonic.

B is not, because

  1. 0 semantically is not a singleton (it is in cPython, but that is an implementation detail).
  2. It will not work with float a or b.
  3. It is actually possible that this will not work in some other implementation of Python.
Improve this answer

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.