Skip to content

README.md

Latest commit

 

History

History
50 lines (39 loc) · 2.41 KB

README.md

File metadata and controls

50 lines (39 loc) · 2.41 KB
Raw

Chapter 17 — Iterators, Generators, and Classic Coroutines

Theme: yield fundamentally alters how a function behaves, changing it from a standard C-stack subroutine into a suspendable, resumable coroutine. Generators enable O(1) memory processing via lazy evaluation.

What You'll Learn

Concept Explanation
Iterable An object containing data. It implements __iter__ to return an Iterator.
Iterator An engine that fetches data. It implements __next__.
Generator A function containing yield. It returns an Iterator implicitly.
yield from Delegates iteration strictly to a sub-generator, bridging the outer caller to the inner sub-gen.
Coroutines Generators used to consume data via .send(data) instead of just producing it.

Key Files

  • examples.py — Demonstrates lazy evaluation, yield from mechanics, and a classic .send() coroutine.
  • exercises.py — The Generator Exhaustion trap and the Coroutine Priming trap.
  • mini_project.py — A realistic data pipeline constructed entirely out of chained generators to process data exactly like Unix pipes (|).
  • benchmarks.py — Proves via tracemalloc that generating 2 million items eagerly in a list costs ~17 MB of RAM, while doing it via a lazy generator costs practically 0 MB.
  • notes.md — The mechanical difference between Iterators and Iterables.
  • pitfalls.md — The fatal mistake of returning lists inside what should be a generator.
  • interview_questions.md — L3 to L6 interview prep.
  • architecture_notes.md — The CPython YIELD_VALUE opcode and PyFrameObject suspension.

30-Second Rules

# Rule 1: A generator is strictly single-use. If you need to loop twice, create it twice.
gen = (x for x in range(10))
list(gen) # [0, 1, ..., 9]
list(gen) # [] <-- EMPTY!

# Rule 2: If a function returns a large list, refactor it to `yield`.
# BAD: O(N) memory
def get_lines(file):
    return file.readlines()

# GOOD: O(1) memory
def get_lines(file):
    for line in file:
        yield line

# Rule 3: Always prime a coroutine by calling next(coro) before sending data.
coro = my_coro()
next(coro) # Essential!
coro.send(data)

Reference: Fluent Python 2nd ed., Chapter 17 — pages 625–690