What data model methods implement argument keyword unpacking?

Question

I have a class that I would like to be able to unpack into an argument list using the *args and **kwargs syntax.

class MyFoo:
    x = 1
    y = 2 # unpack x and y
    z = 3 # do not unpack z

def bar(x, y):
    print(x, y)

def baz(a, b=2, x=100, y=100, z=5):
    print(a, b, x+2, y+2, z)

foo = MyFoo()

bar(*foo) # should print "1 2"
baz(1, **foo) # should print "1 2 3 4 5"

I can implement *foo by defining __iter__:

def __iter__(self):
    return iter([self.x, self.y])

However I haven't been able to figure out how to implement **foo. What methods need to be implemented to do this?

Answer 1

We just need to override keys to get the names of attributes to be unpacked, and then __getitem__ (preferably as a classmethod) to access their value. Check below:

class MyFoo(object):
  x = 1
  y = 2 # unpack x and y
  z = 3 # do not unpack z

  def __iter__(self):
      return iter([self.x, self.y])

  def keys(self):
      return ['x', 'y']

  @classmethod
  def __getitem__(cls, key):
    return cls.__dict__[key]



def bar(x, y):
  print(x, y)

def baz(a, b=2, x=100, y=100, z=5):
  print(a, b, x+2, y+2, z)

foo = MyFoo()

bar(*foo) # should print "1 2"
baz(1, **foo) # should print "1 2 3 4 5"