The __repr__() and __str__() methods
Python has two string representations of an object. These are closely aligned with the built-in functions repr(), str(), print(), and the string.format() method.
- Generally, the
str()method representation of an object is commonly expected to be more friendly to humans. This is built by an object's__str__()method. - The
repr()method representation is often going to be more technical, perhaps even a complete Python expression to rebuild the object. The documentation says:For many types, this function makes an attempt to return a string that would yield an object with the same value when passed to
eval().This is built by an object's
__repr__()method. - The
print()function will usestr()to prepare an object for printing. - The
format()method of a string can also access these methods. When we use{!r}or{!s}formatting, we're requesting__repr__()or__str__(), respectively.
Let's look at the default implementations first.
The following is a simple class hierarchy:
class Card:
insure= False
def __init__( self, rank, suit ):
self.suit= suit
self.rank= rank
self.hard, self.soft = self._points()
class NumberCard( Card ):
def _points( self ):
return int(self.rank), int(self.rank)
We've defined two simple classes with four attributes in each class.
The following is an interaction with an object of one of these classes:
>>> x=NumberCard( '2', '♣') >>> str(x) '<__main__.NumberCard object at 0x1013ea610>' >>> repr(x) '<__main__.NumberCard object at 0x1013ea610>' >>> print(x) <__main__.NumberCard object at 0x1013ea610>
We can see from this output that the default implementations of __str__() and __repr__() are not very informative.
There are two broad design cases that we consider when overriding __str__() and __repr__():
- Non-collection objects: A "simple" object doesn't contain a collection of other objects and generally doesn't involve very complex formatting of that collection
- Collection objects: An object that contains a collection involves somewhat more complex formatting