When is preferable to use if 'key' in dict: instead of if dict.get('key') is not None:?

Question

I've been using the following syntax in my code for checking the existence of a key:

if dict.get("key") is not None:
    print "key found"

However, I have seen in other people's code the following syntax:

if 'key' in dict:
   print "key found"

I was just wondering which way is best to express a dictionary key null check and if there is an advantage (if any) to using one over the other.

I saw the existing question addressing the similar question of what are the alternatives to using .get() but just want to know if there is any advantage between the two examples I provided.

Why would you want to send the key to a function and check a default value rather than simply telling it to do what you actually want? — TigerhawkT3, Apr 01 '17 at 09:38
@TigerhawkT3 that is a good question, and it is probably suited to someone with more Python experience than me. — Willem van Ketwich, Apr 01 '17 at 10:01

score 1 · Accepted Answer · edited Apr 02 '17 at 11:37

1

If there is a key entry with the value None the two expressions will return different results:

>>> d = {'key': None}
>>> d.get("key") is not None
False
>>> 'key' in d
True

If you want to check whether the key exists in the dictionary, the in operator should be used; it also makes your intention clearer. Additionally, it also works for other sequence types (eg. list, tuple, etc.)

BTW, don't use dict as a variable name; it shadows the builtin function/type dict.

edited Apr 02 '17 at 11:37

Braiam

4,345
11
47
69

answered Apr 01 '17 at 09:36

falsetru

314,667
49
610
551

Thanks for this. I am aware of the dict keyword. I used it for demonstration purposes. – Willem van Ketwich Apr 01 '17 at 09:59
@WillemvanKetwich, FYI, `dict` is not a keyword. If it is a keyword, you couldn't use it as a variable name. – falsetru Apr 01 '17 at 10:18
Apologies - data structure. Interesting sidenote about other sequence types, BTW. – Willem van Ketwich Apr 01 '17 at 10:25

letmutx · Answer 2 · 2017-04-01T09:42:47.220

1

This is better.

if 'key' in dict:
   print "key found"

Also, as @falsetrue noted, if you have a None value in your dict, if d.get('key') is not None becomes False which may not be the intended behaviour.

dict.get is useful when you use the value in an expression and want a default value to be used when the key is not in the dict.

d['key'] = 4
a = 5 + d.get('key', 0) # returns 0 if `key` is not in `d`

edited Apr 01 '17 at 09:42

answered Apr 01 '17 at 09:36

letmutx

1,356
11
22

gzc · Answer 3 · 2017-04-01T09:39:32.447

1

These two ways is different.

dict.get("key") is not None means dict has "key" and dict["key"] is not None.

"key" in dict means dict has "key" and don't care it's value is None or other.

edited Apr 01 '17 at 09:39

answered Apr 01 '17 at 09:38

gzc

6,154
5
32
59

1

What I explain is `dict.get("key") is not None` rather than a single `dict.get("key")`. – gzc Apr 01 '17 at 09:41

When is preferable to use if 'key' in dict: instead of if dict.get('key') is not None:?

3 Answers3