notes.md

Chapter 17: Iterators, Generators, and Classic Coroutines — Notes

1. Iterable vs. Iterator

These two terms are constantly confused, but the distinction is mathematically strict in Python.

• Iterable: Any object that has an __iter__ method returning an Iterator. Lists, tuples, and strings are Iterables. They hold data.
• Iterator: Any object that implements __next__ (to return the next item or raise StopIteration) AND __iter__ (which just returns self). Iterators are the engine that powers the for loop.

2. The yield Keyword

The moment you write the word yield anywhere inside a function, Python completely changes the nature of that function.

• It is no longer a normal function. It is a Generator Function.
• When called, it does NOT execute its body. It immediately returns a Generator Object.
• Execution only starts when you call next() on that generator object.
• yield literally suspends the execution state (local variables, instruction pointers) in CPython and returns control to the caller.

3. The yield from Keyword

Introduced in Python 3.3, yield from subgen() does more than just for x in subgen(): yield x. It opens a completely transparent, two-way communication channel between the outermost caller and the innermost sub-generator. If the caller calls .send(), .throw(), or .close(), those calls bypass the delegating generator entirely and are passed directly to the sub-generator.

4. Classic Coroutines

A generator naturally produces data. A coroutine is a generator used to consume data. By writing val = yield, you tell the generator to pause and wait. The caller can then push data into the suspended generator using coroutine.send(data). Crucial Rule: You must always "prime" a coroutine by calling next(coro) or coro.send(None) before you can send actual data into it. You have to advance the generator to the first yield so it is ready to receive.