__iter__ is the preferred way to iterate through an iterable object. If it is not defined the interpreter will try to simulate its behavior using __getitem__. Take a look here

To get the result you are expecting, you need to have a data element with limited len and return each in sequence:

class foo:
    def __init__(self):
        self.data=[10,11,12]

    def __getitem__(self, arg):
        print('__getitem__ called with arg {}'.format(arg))
        return self.data[arg]

bar = foo()
for i in bar:
    print('__getitem__ returned {}'.format(i)) 

Prints:

__getitem__ called with arg 0
__getitem__ returned 10
__getitem__ called with arg 1
__getitem__ returned 11
__getitem__ called with arg 2
__getitem__ returned 12
__getitem__ called with arg 3

Or you can signal the end of the ‘sequence’ by raising IndexError (although StopIteration works as well…):

class foo:
    def __getitem__(self, arg):
        print('__getitem__ called with arg {}'.format(arg))
        if arg>3:
            raise IndexError
        else:    
            return arg

bar = foo()
for i in bar:
    print('__getitem__ returned {}'.format(i))   

Prints:

__getitem__ called with arg 0
__getitem__ returned 0
__getitem__ called with arg 1
__getitem__ returned 1
__getitem__ called with arg 2
__getitem__ returned 2
__getitem__ called with arg 3
__getitem__ returned 3
__getitem__ called with arg 4

The for loop is expecting either IndexError or StopIteration to signal the end of the sequence.

You asked,

What is the difference between __getitem__ and __iter__?

What is __getitem__()?

Suppose that arr is a list. Maybe arr = ["A", "B", "C"]

__getitem__ is a function which is executed when you write:

• elem = arr[0]
• elem = arr[1]
• elem = arr[2]
• etc…

Suppose that we have lyst = [0, 111, 222, 333]

Each of the columns in the table below are equivalent in behavior:

+---------------------+------------------------------+--------------+
| `elem = lyst[0]`    | `elem = lyst.__getitem__(0)` | `elem = 0`   |
| `elem = lyst[1]`    | `elem = lyst.__getitem__(1)` | `elem = 111` |
| `elem = lyst[2]`    | `elem = lyst.__getitem__(2)` | `elem = 222` |
| `elem = lyst[3]`    | `elem = lyst.__getitem__(3)` | `elem = 333` |
+---------------------+------------------------------+--------------+

You can define your own __getitem__ method in any of the classes you write. An example a custom-made class with a __getitem__ method is shown below:

from functools import *  
from copy import *

class BaseArray:
   pass

class Array(BaseArray):
    def __init__(self:BaseArray, obj:object):
        if hasattr(obj, `__iter__`):
            it = iter(obj)
            lyst = list(it)
        lyst = self._init_helper(obj)
        self._lyst = lyst  
        self._lvi  = len(lyst) - 1 
        # `lvi` == `last valid index`

    @singledispatchmethod
    def _init_helper(self:BaseArray, obj:object):
        raise NotImplementedError()

    @_init_helper.register
    def _init_helper(self:BaseArray, arr:BaseArray):
        self._lyst = deepcopy(arr._lyst) 

    @_init_helper.register
    def _init_helper(self:BaseArray, lyst:list):
        self._lyst = deepcopy(lyst) 

    def __getitem__(self:BaseArray, index:int):
        if index > self._lvi:
            # `lvi` == `last valid index`
            raise IndexError("index is too large")
        return self._lyst

    def __setitem__(self:BaseArray, index:int):
        if index > self._lvi:
            # `lvi`== `last valid index`
            raise IndexError("index is too large")
        self._lyst[index] = index   

    def __iter__(self:BaseArray):  
        raise NotImplementedError()

You might or might not care about C++, but in C++ __getitem__ is known as operator[].

Many different languages have something like python’s __getitem__ method. If you become comfortable with the inner-workings __getitem__, it will help you write code in other programming languages as well.

# This code is written in C++, not python
int& Array::operator[](int index)
{
    \ In C++ `this` is like the `self` parameter in python
    if (index >= this->size) {
        throw std::invalid_argument( "index is too large" );
        \ The `throw` keyword from C++ 
        \ is known as `raise` in python
        exit(0);
    }
    return this->ptr[index];
}

What is __iter__()?

Like __getitem__, __iter__() is a class method.

__iter__() is usually used in for loops.

Suppose that cookie_jar is a list like the following:

["oatmeal 1", "chocolate chip 1", "oatmeal 2"]

The following two pieces of code are syntactically different, but are semantically equivalent:

+------------------------------+-----------------------------------------+
|                              | it = iter(cookie_jar)                   |
| for cookie in cookie_jar:    | while True:                             |
|      print(cookie)           |     try:                                |
|                              |         cookie = next(it)               |
|                              |     except (IndexError, StopIteration): |
|                              |         break                           |
|                              |     print(cookie)                       |
+------------------------------+-----------------------------------------+

Also, both of the loops shown above do the same thing as the following:

cookie_jar = ["oatmeal 1", "chocolate chip 1", "oatmeal 2"]
it = cookie_jar.__iter__()
while True:
    try:
        cookie = it.__next__()
    except (StopIteration, IndexError):
        break
    print(cookie)

Many python container classes, such as list, tuple, etc… accept any iterable as input.

That is, you can pass anything as input to __init__ provided that that thing has an __iter__ method.

tuppy_the_tuple =  ("A", "B", "C")
lizzy_the_list  =  ["A", "B", "C"]
steve_the_string = "ABC"

chests = [tuppy_the_tuple, lizzy_the_list, steve_the_string]
for chest in chests:
    larry_the_new_list = list(chest)
    print(larry_the_new_list)   


# treasure chests are an example of a "container"
#         __________
#        /____;;___
#       | /         /
#       `. ())oo() .
#        |( ()*^^()^
#        | |---------|
#         |    ))   |
#         |_________|

An example of how to define your own __iter__ method is shown below:

class BaseCookieJar:
    pass

class CookieJar(BaseCookieJar):
  
    class CookieIterator:

        def __init__(self, cookie_jar:BaseCookieJar):
            self._cookie_jar = cookie_jar
            self._index = 0 
 
        def __iter__(self):
            return self

        def __next__(self):
            if self_index >= len(self._cookie_jar):
                raise StopIteration()
            cookie = self._cookie_jar[self._index]
            self._index += 1
            return cookie 
        # End of Iterator class
    # Resume `CookieJar` class

    def __init__(*ignore_me, **ignore_me_too):  
        self._lyst = [0, 1, 2, 3, 4]

    def __iter__(self):
        return type(self).CookieIterator(self)

    def __len__(self):
        return len(self._lyst)  

What happens if you define your own __getitem__() method but do not define a __iter__() method?

Suppose that you write a class with a __getitem__() method, but no __iter__() method.

Your class will inherit a default __iter__() method. We can emulate the default implementation of __iter__ with the following analog:

IteratorClass = type(
    iter(
        type(
            "TempClass",
            (object,),
            {
                "__getitem__": lambda *args: None,
            }
        )()
    )
)


class ClassChest:
    """
        Something like this happens when you have
            ___getitem__()
        but you do not have...
            ___iter__()
    """
    # class ChestIterator(IteratorClass):
    # We are not allowed to subclass `iterator`
    #     TypeError: type 'iterator' is not an acceptable base type
    class ChestIterator:
        def __init__(self, chest):
            self._chest = chest
            self._idx = 0
            
        def __next__(self):
            idx = self._idx
            try:
                gold_coin =  self._chest[idx]
            except IndexError:
                raise IndexError from None
            gold_coin =  self._chest.__getitem__(idx)
            self._idx = 1 + self._idx
            return gold_coin
        
        def __iter__(self):
            return self
        # End of Iterator
    # Resume Class Chest class
            
    def __iter__(self):
        return type(self).ChestIterator(self)

    def __getitem__(self, idx: int):
        if idx > 4:
            raise IndexError
        return idx

instance_chest = ClassChest()

for shiny_object in instance_chest:
    print("mi treazure is == ", shiny_object)


while True:
    iterator = iter(instance_chest)
    iterator = instance_chest.__iter__()
    try:
        shiny_object = next(iterator)
        shiny_object = iterator.__next__()
    except (StopIteration, IndexError):
        break
    print("mi treazure is == ", shiny_object)

the console output is:

mi treazure is ==  0
mi treazure is ==  1
mi treazure is ==  2
mi treazure is ==  3
mi treazure is ==  4

Some notes about your code

You wrote,

class foo:
    def __getitem__(self, *args):
        print(*args)

bar = foo()
for i in bar:
    print(i)

In your code, None is printed because you failed to specify the return value of __getitem__

Both passages of code in the following table are equivalent:

+-------------------------------+-------------------------------+
|     no `return` statement     |         `return None`         |
+-------------------------------+-------------------------------+
| def __getitem__(self, *args): | def __getitem__(self, *args): |
|     print(*args)              |     print(*args)              |
|                               |     return None               |
+-------------------------------+-------------------------------+

Let us modify your code a bit:

class Klass:
    def __getitem__(self, idx:int):
        print("Inside __getitem__ `idx` is ", idx)
        return idx

    # WARNING: iter(self).__next__() can cause
    # infinite loops if we do not do one of the following:  
    #     * raise `StopIteration`
    #     * raise `IndexError`

obj = Klass()
for item in obj:
    print("Inside of the `for` loop `item` is:", item)

We will have an infinite loop:

[...]
Inside of the `for` loop `item` is: 41875
Inside __getitem__ `idx` is  41876
Inside of the `for` loop `item` is: 41876
Inside __getitem__ `idx` is  41877
Inside of the `for` loop `item` is: 41877
Inside __getitem__ `idx` is  41878
Inside of the `for` loop `item` is: 41878

We can stop the looping by raising a StopIteration exception.

class Klass:
    def __getitem__(self, idx:int):
        if idx > 3: 
            raise StopIteration() 
        print("Inside __getitem__ `idx` is ", idx)
        return idx

obj = Klass()
for item in obj:
    print("Inside of the `for` loop `item` is:", item)