Rotwang wrote:
> I've written a
> decorator to eliminate repeated calls by storing a dictionary whose
> items are arguments and their results:
> The problem is that the dictionary key
> stored depends on how the function was called, even if two calls should
> be equivalent; hence the original function gets called more often than
> necessary.
I think you can normalize the arguments with inspect.getcallargs(). A
sketch:
>>> import inspect
>>> def memo(f):
... def g(*args, **kw):
... callargs = tuple(sorted(inspect.getcallargs(f, *args,
**kw).items()))
... try:
... return cache[callargs]
... except KeyError:
... result = cache[callargs] = f(*args, **kw)
... return result
... cache = {}
... return g
...
>>> @memo
... def f(x, y=2):
... print "calling f(%s, %s)" % (x, y)
... return x + y
...
>>> f(1, 2)
calling f(1, 2)
3
>>> f(1)
3
>>> f(y=2, x=1)
3
--
http://mail.python.org/mailman/listinfo/python-list
