🐍 Python Tip #86: functools.wraps()

When you create decorators in Python, there's a small problem that can easily be overlooked: the decorator can replace the original function's name and documentation.

functools.wraps() fixes that by preserving important information about the original function.

---

Importing wraps

from functools import wraps

---

The Problem Without wraps()

def my_decorator(func):
def wrapper():
print("Before function")
func()
return wrapper

@my_decorator
def greet():
"""Say hello to the user."""
print("Hello!")

print(greet.name)
print(greet.doc)

Output:

wrapper
None

The original function's metadata has been replaced by the wrapper's metadata.

---

Using wraps()

from functools import wraps

def my_decorator(func):
@wraps(func)
def wrapper():
print("Before function")
func()
return wrapper

@my_decorator
def greet():
"""Say hello to the user."""
print("Hello!")

print(greet.name)
print(greet.doc)

Output:

greet
Say hello to the user.

Now Python knows that the wrapped function is still greet.

---

Why Is This Useful?

Decorators are common in real Python projects.

They're used for things like:

✓ Logging
✓ Authentication
✓ Timing functions
✓ Caching
✓ Validation
✓ Permission checks
✓ Framework features

Using @wraps(func) makes your decorators behave more like the original functions.

---

A More Practical Example

from functools import wraps
import time

def timer(func):
@wraps(func)
def wrapper(*args, **kwargs):
start = time.perf_counter()

result = func(*args, **kwargs)

end = time.perf_counter()

print(f"{func.name} took {end - start:.6f} seconds")

return result

return wrapper

@timer
def calculate():
return sum(range(100000))

print(calculate())

The decorator measures the function's execution time while @wraps() preserves the original function's metadata.

---

Quick Rule

When writing your own decorators, a good habit is:

@wraps(func)

Put it directly above the wrapper function.

---

Challenge:

Create a decorator called logger that:

1. Prints "Function started" before running a function.
2. Runs the original function.
3. Prints "Function finished" afterward.
4. Uses @wraps(func).

Then decorate a function called hello() and check its __name__.

It should still print:

hello

🐍 Python Tip #87: collections.ChainMap

Ever needed to work with multiple dictionaries as if they were one?

ChainMap from the collections module lets you combine multiple dictionaries into a single view without actually merging or copying them.

---

Importing ChainMap

from collections import ChainMap

---

Basic Example

from collections import ChainMap

defaults = {
"theme": "light",
"language": "English"
}

user_settings = {
"theme": "dark"
}

settings = ChainMap(user_settings, defaults)

print(settings["theme"])
print(settings["language"])

Output:

dark
English

Python checks the dictionaries from left to right.

So if a key exists in user_settings, that value is used first.

---

Why Is This Useful?

Imagine an application has:

1. User settings
2. Project settings
3. Default settings

Instead of copying everything into one dictionary, you can use:

settings = ChainMap(
user_settings,
project_settings,
defaults
)

If the user has a custom value, it takes priority.

If not, Python checks the next dictionary.

---

Adding a New Dictionary

You can add another mapping with new_child():

from collections import ChainMap

defaults = {"color": "blue"}
user = {"name": "Ali"}

settings = ChainMap(user, defaults)

temporary = settings.new_child(
{"color": "red"}
)

print(temporary["color"])

Output:

red

The new dictionary gets the highest priority.

---

Updating Values

If you change a value:

settings["name"] = "Sara"

the change is made in the first dictionary in the chain.

ChainMap doesn't merge everything into one new dictionary.

It keeps references to the original mappings.

---

Real-World Example

from collections import ChainMap

system_config = {
"debug": False,
"language": "English"
}

app_config = {
"debug": True
}

config = ChainMap(app_config, system_config)

print(config["debug"])
print(config["language"])

Output:

True
English

This pattern can be useful for configuration systems where different levels of settings override each other.

---

Why use ChainMap?

✓ Combines multiple dictionaries without copying them
✓ Useful for configuration systems
✓ Supports priority-based lookups
✓ Keeps the original dictionaries separate
✓ Useful when working with layered settings

---

Challenge:

Create three dictionaries:

defaults = {"theme": "light", "font": "Arial"}
user = {"theme": "dark"}
temporary = {"font": "Courier"}

Use ChainMap so that:

1. temporary has the highest priority.
2. user comes next.
3. defaults is the fallback.

Then print the values of:

theme
font

Expected output:

dark
Courier

🐍 Python Tip #88: contextvars

Ever needed a variable whose value is local to the current task or execution context?

Python's contextvars module is especially useful in asynchronous applications where multiple tasks can be running at the same time.

---

Importing contextvars

import contextvars

---

Creating a Context Variable

import contextvars

user = contextvars.ContextVar("user")

user.set("Yehya")

print(user.get())

Output:

Yehya

ContextVar stores a value that belongs to the current execution context.

---

Using a Default Value

import contextvars

language = contextvars.ContextVar(
"language",
default="English"
)

print(language.get())

Output:

English

If you haven't assigned a value yet, the default is returned.

---

Changing the Value

import contextvars

username = contextvars.ContextVar(
"username",
default="Guest"
)

print(username.get())

username.set("Yehya")

print(username.get())

Output:

Guest
Yehya

---

Using ContextVar in Functions

import contextvars

current_user = contextvars.ContextVar(
"current_user",
default="Guest"
)

def show_user():
print("Current user:", current_user.get())

current_user.set("Ali")

show_user()

current_user.set("Sara")

show_user()

Output:

Current user: Ali
Current user: Sara

---

Why Is This Useful?

ContextVar is commonly useful in:

✓ Async applications
✓ Web servers
✓ Request-specific data
✓ Logging context
✓ User/session information
✓ Code where multiple tasks need separate values

For example, a web application might store a request ID in a ContextVar so different parts of the application can access it without passing the ID through every function.

---

Important Difference

A normal global variable is shared:

current_user = "Ali"

A ContextVar is designed so different execution contexts can have different values.

This becomes especially useful with asyncio and other concurrent code.

---

Challenge:

Create a ContextVar called:

current_user

Give it a default value of:

"Guest"

Then:

1. Print the default value.
2. Change it to "Yehya".
3. Create a function that prints the current user.
4. Call the function.

Expected output:

Guest
Yehya

🐍 Python Tip #89: types.SimpleNamespace

Sometimes you have a few related values and want to access them with dot notation instead of dictionary brackets.

SimpleNamespace from Python's types module gives you a simple object for storing attributes.

---

Importing SimpleNamespace

from types import SimpleNamespace

---

Basic Example

from types import SimpleNamespace

user = SimpleNamespace(
name="Yehya",
age=16,
language="Python"
)

print(user.name)
print(user.age)
print(user.language)

Output:

Yehya
16
Python

Instead of:

user["name"]

you can use:

user.name

---

Changing Values

The attributes can be changed normally:

user.age = 17

print(user.age)

Output:

17

You can also add new attributes:

user.country = "Ethiopia"

print(user.country)

Output:

Ethiopia

---

Creating One From a Dictionary

You can also convert dictionary data into a namespace:

from types import SimpleNamespace

data = {
"name": "Sara",
"score": 95
}

student = SimpleNamespace(**data)

print(student.name)
print(student.score)

Output:

Sara
95

---

Converting Back to a Dictionary

Use vars():

data = vars(student)

print(data)

Output:

{'name': 'Sara', 'score': 95}

---

When Is It Useful?

SimpleNamespace can be handy for:

✓ Small data containers
✓ Configuration objects
✓ Temporary structured data
✓ Making dictionary data easier to access
✓ Quick prototypes

It's not a replacement for dataclass when you need a more structured data model.

---

Real Example

from types import SimpleNamespace

server = SimpleNamespace(
host="localhost",
port=8000,
debug=True
)

print(f"Server: {server.host}:{server.port}")
print("Debug:", server.debug)

Output:

Server: localhost:8000
Debug: True

---

Challenge:

Create a SimpleNamespace called student with:

name = "Yehya"
grade = 11
subject = "Python"

Then:

1. Print the student's name.
2. Change the grade to 12.
3. Add a new attribute called score.
4. Print all the information.

Can you do it without using a dictionary?

🐍 Python Tip #90: functools.cache()

If a function keeps receiving the same arguments and doing the same expensive calculation again and again, caching can save a lot of work.

Python's functools.cache() automatically remembers previous results and reuses them when the same arguments appear again.

---

Importing cache

from functools import cache

---

Basic Example

from functools import cache

@cache
def square(number):
print("Calculating...")
return number ** 2

print(square(5))
print(square(5))
print(square(10))

Output:

Calculating...
25
25
Calculating...
100

Notice that "Calculating..." only appears once for 5.

The second square(5) gets the stored result instead of running the function again.

---

A More Useful Example

from functools import cache

@cache
def fibonacci(n):
if n < 2:
return n

return fibonacci(n - 1) + fibonacci(n - 2)

print(fibonacci(10))

Output:

55

Without caching, recursive Fibonacci calculations repeat many of the same calculations.

With @cache, previously calculated results are reused.

---

Checking the Cache

You can inspect the cache with:

fibonacci.cache_info()

For example, you might see:

CacheInfo(hits=8, misses=11, maxsize=None, currsize=11)

hits = results retrieved from the cache.

misses = calculations that had to actually run.

currsize = number of stored results.

---

Clearing the Cache

If you want to remove all stored results:

fibonacci.cache_clear()

After that, the function will calculate the results again when called.

---

Important Note

@cache is basically an unlimited cache.

That means stored results can continue growing as new arguments are used.

If you need a limited cache size, functools.lru_cache() is usually more appropriate.

---

Why use cache()?

✓ Avoids repeating expensive calculations
✓ Extremely useful with recursive functions
✓ Simple to add with @cache
✓ Can make repeated function calls much faster
✓ Part of Python's standard library

---

Challenge:

Create a cached function:

@cache
def factorial(n):

It should calculate:

5! = 120

Then call:

factorial(5)
factorial(5)
factorial(6)

Finally, use:

factorial.cache_info()

and check how many cache hits you get.