Summary - Advanced decorators
- Python course
- Advanced
- Lesson 6/83
- 5 min
Combine lru_cache, partial and class decorator to write a cache layer.
Summary - Advanced decorators
This group covered advanced decorators. A parameterized decorator has three layers: factory → decorator → wrapper. A class decorator holds state through __call__.
The functools module gave us three tools: wraps preserves metadata, lru_cache caches results, partial binds arguments, and reduce folds a sequence into one value.
The final task combines a parameterized decorator with lru_cache: repeat count and caching side by side.
Practice Combine repeat and lru_cache
Write the parameterized repeat(n) decorator (factory, deco, wrapper; with @wraps(fn)). wrapper calls fn(*args) n times and returns the last result. Decorate slow_square(n) with @lru_cache(maxsize=None) and @repeat(3) (lru_cache on top). Print the result of slow_square(4).
Starter code
from functools import wraps, lru_cache
# repeat(n): factory → deco → wrapper; call n times, return result
# decorate wrapper with the wraps decorator
# also decorate slow_square with lru_cache(maxsize=None) on top
# finally print the result of slow_square(4)
@repeat(3)
def slow_square(n):
return n * nSample solution
from functools import wraps, lru_cache
def repeat(n):
def deco(fn):
@wraps(fn)
def wrapper(*args):
result = None
for _ in range(n):
result = fn(*args)
return result
return wrapper
return deco
@lru_cache(maxsize=None)
@repeat(3)
def slow_square(n):
return n * n
print(slow_square(4))Expected output
16
Questions and answers
Review the lesson questions together with their answers. To test yourself, solve the lesson interactively.
1 In what order does the combined decorator chain run?
- lru_cache first, then repeat
- repeat first, then lru_cache Correct answer
- In parallel at the same time
- Only repeat; lru_cache is ignored
Explanation: @repeat(3) is applied first; each call then caches the wrapped function's result via lru_cache.
2 Combine the decorator with lru_cache
from functools import wraps, lru_cache
@____(maxsize=None)
@repeat(3)
def slow_square(n):
return n * nfrom functools import wraps, lru_cache
@lru_cache(maxsize=None)
@repeat(3)
def slow_square(n):
return n * nExplanation: @lru_cache(maxsize=None) is on top; repeat(3) below is applied first.
3 Match the group concepts
| Term | Meaning |
|---|---|
repeat(n) |
Parameterized factory |
class Counter |
Holds state |
lru_cache |
Result cache |
partial |
Binds arguments |