Expand functools features to methods, classmethods, staticmethods and even for (unofficial) hybrid methods.
For now, methodtools only provides methodtools.lru_cache.
Use methodtools module instead of functools module. Than it will work as you expected.
from methodtools import lru_cache
class A(object):
# cached method. the storage lifetime follows `self` object
@lru_cache()
def cached_method(self, args):
...
# cached classmethod. the storage lifetime follows `A` class
@lru_cache() # the order is important!
@classmethod # always lru_cache on top of classmethod
def cached_classmethod(self, args):
...
# cached staticmethod. the storage lifetime follows `A` class
@lru_cache() # the order is important!
@staticmethod # always lru_cache on top of staticmethod
def cached_staticmethod(self, args):
...
@lru_cache() # just same as functools.lru_cache
def cached_function():
...
PyPI is the recommended way.
$ pip install methodtools
- To browse versions and tarballs, visit:
- https://pypi.python.org/pypi/methodtools/
Note
If you are working on Python 2, install also functools32.
- Documentation
- This project is derived from Ring, a rich cache interface using the same method handling technique.
- To learn more about bound method dispatching, see also wirerope.