To allow the use of in
for custom classes the class must either provide the magic method __contains__
or, failing that, an __iter__
-method.
Suppose you have a class containing a list
of list
s:
class ListList:
def __init__(self, value):
self.value = value
# Create a set of all values for fast access
self.setofvalues = set(item for sublist in self.value for item in sublist)
def __iter__(self):
print('Using __iter__.')
# A generator over all sublist elements
return (item for sublist in self.value for item in sublist)
def __contains__(self, value):
print('Using __contains__.')
# Just lookup if the value is in the set
return value in self.setofvalues
# Even without the set you could use the iter method for the contains-check:
# return any(item == value for item in iter(self))
Using membership testing is possible using in
:
a = ListList([[1,1,1],[0,1,1],[1,5,1]])
10 in a # False
# Prints: Using __contains__.
5 in a # True
# Prints: Using __contains__.
even after deleting the __contains__
method:
del ListList.__contains__
5 in a # True
# Prints: Using __iter__.
Note: The looping in
(as in for i in a
) will always use __iter__
even if the class implements a __contains__
method.