Generators

Generator is non-exiting function that stores it's state in memory and can return result values many times when called.

Generator is the result of generator function

Generators function looks like regular function but instead of return they use yield which will return provided value and "freeze" generator state. Next time generator is called it will be resumed from the next statement after yield.

After creation of generator we can call it's __next__() method to get next value (next() for Python2). If no items will be available (function will end) StopIteration exception will be thrown to indicate this.

So, function is a "generator function" if it has yield statement.

The result of generator function is called generator.

Generator is specific kind of object that has __next__ method which returns "next" value until raising StopIteration exception meaning that it is the end of data.

Generator is particular case of more generic type - "iterator" which is defined as object with __next__ and __iter__ methods.

🪄 Code:

def sum_range(x, y):
    print("Starting generator...")
    for i in range(x + y):
        yield i
        print("Next iteration!")
    print("Stopping generator...Bye-bye!")

gen = sum_range(2, 2)
print("First __next__:", next(gen))
print("Second __next__:", next(gen))

for x in gen:  # Recommended way of looping
    print("For loop:", x)

try:
    print("__next__ on 'empty':", gen.__next__())
except StopIteration:
    print("Generator is already empty (raised StopIteration)")

📟 Output:

Starting generator...
First __next__: 0
Next iteration!
Second __next__: 1
Next iteration!
For loop: 2
Next iteration!
For loop: 3
Next iteration!
Stopping generator...Bye-bye!
Generator is already empty (raised StopIteration)

Generator expressions

Very similar to list comprehension. Instead of sqaure brackets [] for generator expressions it's needed to use parentheses ().

Important thing to understand - generator is the result of calling generator function or the result of running generator expression

🪄 Code:

gen = (x for x in range(1,10) if not x % 2) # get generator using generator expression
print(gen)

for i in gen:
    print(i)
gen = (x for x in range(1,10) if not x % 2) # get generator using generator expression

print(list(gen))

📟 Output:

<generator object <genexpr> at 0x7f3e79f23c78>
2
4
6
8
[2, 4, 6, 8]

🪄 Code:

def gen_func():
    for x in range(1, 10):
        if not x % 2:
            yield x
gen = gen_func() #  get generator object by using generator function 
print(gen)
for i in gen:
    print(i)

📟 Output:

<generator object gen_func at 0x103dcf518>
2
4
6
8