Get all object attributes in Python?

I use __dict__ and dir(<instance>)

Example:

class MyObj(object):
  def __init__(self):
    self.name = 'Chuck Norris'
    self.phone = '+6661'

obj = MyObj()
print(obj.__dict__)
print(dir(obj))

# Output:  
# obj.__dict__ --> {'phone': '+6661', 'name': 'Chuck Norris'}
#
# dir(obj)     --> ['__class__', '__delattr__', '__dict__', '__doc__',
#               '__format__', '__getattribute__', '__hash__', 
#               '__init__', '__module__', '__new__', '__reduce__', 
#               '__reduce_ex__', '__repr__', '__setattr__', 
#               '__sizeof__', '__str__', '__subclasshook__', 
#               '__weakref__', 'name', 'phone']

Use the built-in function dir().

What you probably want is dir().

The catch is that classes are able to override the special __dir__ method, which causes dir() to return whatever the class wants (though they are encouraged to return an accurate list, this is not enforced). Furthermore, some objects may implement dynamic attributes by overriding __getattr__, may be RPC proxy objects, or may be instances of C-extension classes. If your object is one these examples, they may not have a __dict__ or be able to provide a comprehensive list of attributes via __dir__: many of these objects may have so many dynamic attrs it doesn't won't actually know what it has until you try to access it.

In the short run, if dir() isn't sufficient, you could write a function which traverses __dict__ for an object, then __dict__ for all the classes in obj.__class__.__mro__; though this will only work for normal python objects. In the long run, you may have to use duck typing + assumptions - if it looks like a duck, cross your fingers, and hope it has .feathers.