Dev Blog

Summary ​

Conclusions ​

A function combines a name, parameters, a contract, and a result. Its signature determines how an argument is passed; modifying a shared object differs from reassigning a local name. A closure retains state in an enclosing function. Recursion is suitable when a problem naturally splits into smaller ones, but requires a base case, limits, and control of repeated calculations. Annotations make these agreements visible to readers and tools.

Self-check questions ​

  1. How does a parameter differ from an argument?
  2. When does a def body run, and when are default values evaluated?
  3. How do return, print, and implicit None differ?
  4. What does return low, high actually return?
  5. How do the / and * markers work?
  6. How does collecting arguments differ from unpacking?
  7. Why does a default list retain changes between calls?
  8. Why does reassigning a parameter not replace the caller's list?
  9. What is the LEGB name lookup order?
  10. When are global and nonlocal needed?
  11. What data does a closure retain?
  12. What proves that recursion terminates?
  13. Why does Fibonacci caching not eliminate stack depth?
  14. How do you annotate a function argument?
  15. How do deferred annotations differ from string annotations?
  1. Functions and parameters: https://docs.python.org/3.14/tutorial/controlflow.html.
  2. Scope: https://docs.python.org/3.14/reference/executionmodel.html.
  3. Typing: https://docs.python.org/3.14/library/typing.html.
  4. Deferred annotations: https://docs.python.org/3.14/library/annotationlib.html.