Python Closures and Currying

A closure is a function that can be customized due to bound variables (easily done using objects). If currying is applied then is a function that remembers and reuse the result of its previous execution and/or the arguments previously passed. class sorter(): def __init__(self,reverse=None): self.reverse = reverse or False self.result = [] def __call__(self,target,reverse=None): self.reverse = reverse if reverse is not None else self.reverse self.result += target return sorted(self.result,reverse=self.reverse) This function is a closure: In [8]:l = [1,4,65,2,3] In [9]:sorter(True)(l) Out[9]:[65, 4, 3, 2, 1] In [10]:sorter(False)(l) Out[10]:[1, 2, 3, 4, 65] And also applies currying: In [30]:s = sorter(True) In [31]:s(l) Out[31]:[65, 4, 3, 2, 1] In [32]:s([6,7,8],False) Out[32]:[1, 2, 3, 4, 6, 7, 8, 65] In [33]:s([1,65,8]) Out[33]:[1, 1, 2, 3, 4, 6, 7, 8, 8, 65, 65]    Other types of currying closures use sequential argument initialization (each setter method returns self): result = function.setArg1(x).setArg2(y).setArg3(3).calculate() Partial doesn't work fine with this kind of closures, as partial doesn't propagate inner attributes of the function object. I way to implement curry closures could use call(self,*args,**kw) to manage argument identification.