I was quite surprised when
[] is not []
evaluated to True.
What is happening in this code? What really not and is statements are doing?
views:
131answers:
5
+9
A:
a is not b is a special operator which is equivalent to not a is b.
The operator a is b returns True if a and b are bound to the same object, otherwise False. When you create two empty lists you get two different objects, so is returns False (and therefore is not returns True).
Mark Byers
2010-09-15 14:22:21
*if a and b are the same object* - I would like *are bound to the same object* even better.
Space_C0wb0y
2010-09-15 14:30:21
@Space_C0wb0y: Fixed.
Mark Byers
2010-09-15 14:32:32
+8
A:
is is the identity comparison.
== is the equality comparison.
Your statement is making two different lists and checking if they are the same instance, which they are not. If you use == it will return true and because they are both empty lists.
unholysampler
2010-09-15 14:26:24
+1
A:
is checks for identity. [] and [] are two different (but equivalent) lists. If you want to check if both the lists are empty you can use their truth value (false for empty strings, collections, and zeros).
if not ([] and []):
print 'Spanish Inquisition'
the only time that is is guaranteed to return True is for singletons such as None. Like the Highlander, there can be only one instance of None in your program - every time you return None it's the very same "thing" as the none referred to if you type print None.
[], OTOH, is not guaranteed to be anything except an empty list and evaluate to False in a boolean context.
Wayne Werner
2010-09-15 14:26:44
@Wayne: As far as I know, it's true for everything. When you assign `b` to `a` in python, you are assigning `a` to be a reference to the object currently referenced by `b`. So their identity will be identical. It's python, so it may be possible to do some weird thing like overriding assignment to break this, but in general, I'd say yes.
recursive
2010-09-16 14:09:59
I guess that makes sense - and is different from `a = "hi"; b = "hi"` which `is` behavior is *not* guaranteed aside from returning True or False.
Wayne Werner
2010-09-16 14:30:24
+1
A:
The best way to describe WHY that happens is this:
Here is your example
>>> x = []
>>> y = []
>>> print(x is y)
... False
x and y are actually two different lists, so if you add something to x, it does not appear in y
>>> x.append(1)
>>> print(x)
... [1]
>>> print(y)
... []
So how do we make (x is y) evaluate true?
>>> x = []
>>> y = x
>>> print(x is y)
... True
>>> x.append(10)
>>> print(x)
... [10]
>>> print(y)
... [10]
>>> print(x is y)
... True
if you want to see if two lists have the same contents...
>>> x = []
>>> y = []
>>> print(x == y)
... True
>>> x.append(21)
>>> print(x)
... [21]
>>> print(y)
... []
>>> print(x == y)
... False
>>> y = [21]
>>> print(x == y)
... True
Jiaaro
2010-09-15 14:42:39