edgegraph.structure.singleton.add_mapping

Contents

edgegraph.structure.singleton.add_mapping#

edgegraph.structure.singleton.add_mapping(identifier, obj)#

Adds another mapping to a semi-singleton instance.

This function can be used to add another mapping to an instance of a semi-singleton object. It can be useful for situations where the same object can have multiple “names” (or, sets of arguments), but it is inefficient to implement this in the metaclass hash function. Using this function instead allows a memory-time tradeoff, tipping the balance towards more memory usage in favor of a faster hash function (and therefore, object instantiation).

>>> from edgegraph.singleton import semi_singleton_metaclass, add_mapping
>>> class SemiSingleton(metaclass=semi_singleton_metaclass()):
...     def __init__(self, foo, bar=False):
...         self.foo = foo
...         self.bar = bar
...
>>> s3 = SemiSingleton(37, True)  # different arguments -- different object
>>> s3
<__main__.SemiSingleton object at 0x01234567>
>>> s2 is s3
False
>>> add_mapping(((39,), {'bar': True}), s3)
>>> s4 = SemiSingleton(39, True)  # we get s3 again, despite different args
>>> s4
<__main__.SemiSingleton object at 0x01234567>
>>> s3 is s4
True
Parameters:

  • identifier (Hashable) – Two-tuple of arguments and keyword arguments that will be passed to the class’s __init__ method. Typically, the first element is another tuple representing positional arguments, and the second elemtn is a dictionary representing keyword arguments.

  • obj (object) – The object to map the extra identifier to. Henceforth after this function, this object will be reachable by the identifier given as well as any others it may have already had.

Contents