property() Decorator Complexity¶

The property() decorator allows you to define custom getter, setter, and deleter methods for class attributes. It enables attribute-like access with custom logic.

Complexity Analysis¶

Operation	Time	Space	Notes
Create property	O(1)	O(1)	Decorator application
Get value	O(1)	O(1)	Call getter function
Set value	O(1)	O(1)	Call setter function
Delete value	O(1)	O(1)	Call deleter function
Descriptor lookup	O(d)	O(1)	d = MRO depth
Total operation	O(d)	O(1)	MRO traversal may occur

Basic Usage¶

Simple Getter¶

# O(1) - define read-only property
class Circle:
    def __init__(self, radius):
        self._radius = radius

    @property
    def radius(self):
        """Get radius - O(1)"""
        return self._radius

# O(1) - access like attribute
circle = Circle(5)
r = circle.radius  # 5 (not a method call)

# Cannot set - read-only
# circle.radius = 10  # AttributeError

Getter and Setter¶

# O(1) - define read-write property
class Temperature:
    def __init__(self, celsius):
        self._celsius = celsius

    @property
    def celsius(self):
        """Get temperature - O(1)"""
        return self._celsius

    @celsius.setter
    def celsius(self, value):
        """Set temperature - O(1)"""
        if value < -273.15:
            raise ValueError("Temperature below absolute zero")
        self._celsius = value

# O(1) - get and set
temp = Temperature(20)
print(temp.celsius)  # 20

temp.celsius = 25    # O(1) - calls setter
print(temp.celsius)  # 25

Computed Property¶

# O(1) - property computes value
class Rectangle:
    def __init__(self, width, height):
        self.width = width
        self.height = height

    @property
    def area(self):
        """Calculate area - O(1)"""
        return self.width * self.height

# O(1) - computed on access
rect = Rectangle(5, 10)
a = rect.area  # 50 (computed, not stored)

# Change dimensions
rect.width = 6
a = rect.area  # 60 (recomputed)

Getter, Setter, and Deleter¶

# O(1) - complete property control
class Data:
    def __init__(self):
        self._value = None

    @property
    def value(self):
        """Get value - O(1)"""
        print("Getting value")
        return self._value

    @value.setter
    def value(self, val):
        """Set value - O(1)"""
        print(f"Setting value to {val}")
        self._value = val

    @value.deleter
    def value(self):
        """Delete value - O(1)"""
        print("Deleting value")
        del self._value

# O(1) - use property
d = Data()
d.value = 42      # Prints: Setting value to 42
v = d.value       # Prints: Getting value, returns 42
del d.value       # Prints: Deleting value

Complexity Details¶

Descriptor Protocol¶

# O(1) - property uses descriptor protocol
class WithProperty:
    @property
    def value(self):
        return 42

# When accessing:
# 1. Lookup 'value' in class dict - O(1)
# 2. Found descriptor (property object)
# 3. Call property.__get__() - O(1)
# 4. Calls getter function - O(1)

obj = WithProperty()
val = obj.value  # O(1)

MRO Traversal¶

# O(d) - inherited properties
class Parent:
    @property
    def value(self):
        return "parent"

class Child(Parent):
    pass

# O(d) - lookup via MRO
obj = Child()
val = obj.value  # O(d) - traverse MRO to find property

Property Override¶

# O(d) - child can override property
class Parent:
    @property
    def value(self):
        return "parent"

class Child(Parent):
    @property
    def value(self):
        return "child"

# O(d) - finds Child property first
parent_val = Parent().value  # "parent"
child_val = Child().value    # "child"

Performance Patterns¶

Property vs Direct Attribute¶

# Direct attribute - O(1)
class Direct:
    def __init__(self):
        self.value = 42

# Property - O(1) + function call overhead
class WithProperty:
    def __init__(self):
        self._value = 42

    @property
    def value(self):
        return self._value

# Both O(1), but property has ~5% overhead
direct = Direct()
val1 = direct.value  # O(1)

with_prop = WithProperty()
val2 = with_prop.value  # O(1) + overhead

Property vs Method Call¶

# Method - O(1) with explicit call
class WithMethod:
    def __init__(self):
        self._value = 42

    def get_value(self):
        return self._value

# Property - O(1) with implicit call
class WithProperty:
    def __init__(self):
        self._value = 42

    @property
    def value(self):
        return self._value

# Both O(1), property looks cleaner
obj1 = WithMethod()
val1 = obj1.get_value()  # Explicit call

obj2 = WithProperty()
val2 = obj2.value  # Attribute access

Caching in Properties¶

# O(1) with caching - compute once
class Cached:
    def __init__(self):
        self._computed = None

    @property
    def value(self):
        """Compute once and cache - O(1)"""
        if self._computed is None:
            self._computed = expensive_computation()
        return self._computed

    def expensive_computation(self):
        # ... slow operation ...
        return 42

# vs recompute every time - O(f) where f = computation time
class NoCache:
    @property
    def value(self):
        """Recompute every time"""
        return expensive_computation()

# Caching significantly improves performance

Common Use Cases¶

Validation on Set¶

# O(1) - validate input
class Person:
    def __init__(self, age):
        self._age = None
        self.age = age  # Calls setter

    @property
    def age(self):
        """Get age - O(1)"""
        return self._age

    @age.setter
    def age(self, value):
        """Set age with validation - O(1)"""
        if not isinstance(value, int) or value < 0:
            raise ValueError("Age must be non-negative integer")
        self._age = value

# O(1) - automatic validation
p = Person(30)
p.age = 35  # Valid

try:
    p.age = -5  # Invalid - raises ValueError
except ValueError as e:
    print(e)

Computed Properties¶

# O(1) - compute related value
class Distance:
    def __init__(self, kilometers):
        self.km = kilometers

    @property
    def miles(self):
        """Convert to miles - O(1)"""
        return self.km * 0.621371

# O(1) - accessed like attribute
d = Distance(10)
miles = d.miles  # 6.21371

Lazy Loading¶

# O(1) first access, O(n) computation
class Document:
    def __init__(self, filename):
        self.filename = filename
        self._content = None

    @property
    def content(self):
        """Load on first access - O(1) or O(n)"""
        if self._content is None:
            with open(self.filename) as f:
                self._content = f.read()  # O(n) - load file
        return self._content  # O(1) - return cached

# O(1) - first call loads file
# O(1) - subsequent calls return cached
doc = Document('large_file.txt')
text1 = doc.content  # O(n) - loads
text2 = doc.content  # O(1) - cached

Read-Only Properties¶

# O(1) - immutable attributes
class ImmutablePoint:
    def __init__(self, x, y):
        self._x = x
        self._y = y

    @property
    def x(self):
        """Get x - O(1), read-only"""
        return self._x

    @property
    def y(self):
        """Get y - O(1), read-only"""
        return self._y

# O(1) - get values
p = ImmutablePoint(1, 2)
x = p.x  # 1

# Cannot set - no setter defined
# p.x = 5  # AttributeError

Type Conversion¶

# O(1) - convert on access
class Temperature:
    def __init__(self, celsius):
        self._celsius = celsius

    @property
    def celsius(self):
        """Get Celsius - O(1)"""
        return self._celsius

    @property
    def fahrenheit(self):
        """Get Fahrenheit - O(1)"""
        return self._celsius * 9/5 + 32

    @property
    def kelvin(self):
        """Get Kelvin - O(1)"""
        return self._celsius + 273.15

# O(1) - convert automatically
temp = Temperature(20)
c = temp.celsius      # 20
f = temp.fahrenheit   # 68.0
k = temp.kelvin       # 293.15

Advanced Usage¶

Property with Side Effects¶

# O(1) - setter with side effects
class Volume:
    def __init__(self, value):
        self._value = value
        self.changed = False

    @property
    def value(self):
        """Get value - O(1)"""
        return self._value

    @value.setter
    def value(self, new_value):
        """Set value and track change - O(1)"""
        if new_value != self._value:
            self.changed = True
            self._value = new_value

# O(1) - track changes
vol = Volume(100)
vol.value = 150  # Sets changed = True
print(vol.changed)  # True

Property Inheritance¶

# O(d) - override property in subclass
class Parent:
    @property
    def name(self):
        return "Parent"

class Child(Parent):
    @property
    def name(self):
        return "Child"

# O(d) - finds correct property
parent = Parent()
child = Child()

print(parent.name)  # "Parent"
print(child.name)   # "Child"

Dynamic Properties¶

# O(1) - add properties dynamically
class Dynamic:
    def __init__(self):
        self._data = {}

    def add_property(self, name, getter, setter=None):
        """Add property dynamically - O(1)"""
        prop = property(getter, setter)
        setattr(self.__class__, name, prop)

# Create instance
obj = Dynamic()

# Add property dynamically
obj.add_property(
    'value',
    getter=lambda self: self._data.get('value', 0)
)

# Use property
obj.value = 42

Practical Examples¶

Account Balance with Limits¶

# O(1) - property with validation and limits
class BankAccount:
    def __init__(self, initial_balance):
        self._balance = initial_balance

    @property
    def balance(self):
        """Get balance - O(1)"""
        return self._balance

    @balance.setter
    def balance(self, amount):
        """Set balance with validation - O(1)"""
        if amount < 0:
            raise ValueError("Balance cannot be negative")
        if amount > 1_000_000:
            raise ValueError("Balance exceeds limit")
        self._balance = amount

    @balance.deleter
    def balance(self):
        """Reset balance - O(1)"""
        self._balance = 0

# O(1) - use account
account = BankAccount(100)
print(account.balance)  # 100

account.balance = 500   # Valid
print(account.balance)  # 500

del account.balance     # Reset to 0
print(account.balance)  # 0

Configuration Management¶

# O(1) - validate configuration
class Config:
    def __init__(self):
        self._debug = False
        self._timeout = 30

    @property
    def debug(self):
        """Get debug flag - O(1)"""
        return self._debug

    @debug.setter
    def debug(self, value):
        """Set debug - O(1)"""
        if not isinstance(value, bool):
            raise TypeError("debug must be boolean")
        self._debug = value

    @property
    def timeout(self):
        """Get timeout - O(1)"""
        return self._timeout

    @timeout.setter
    def timeout(self, seconds):
        """Set timeout with bounds - O(1)"""
        if not (1 <= seconds <= 300):
            raise ValueError("timeout must be 1-300 seconds")
        self._timeout = seconds

# O(1) - validated configuration
config = Config()
config.debug = True
config.timeout = 60

User Authentication¶

# O(1) - property for authenticated user
class User:
    def __init__(self, username, password_hash):
        self.username = username
        self._password_hash = password_hash
        self._authenticated = False

    @property
    def authenticated(self):
        """Check authentication - O(1)"""
        return self._authenticated

    def authenticate(self, password):
        """Authenticate - O(1)"""
        import hashlib
        password_hash = hashlib.sha256(password.encode()).hexdigest()
        if password_hash == self._password_hash:
            self._authenticated = True
        return self._authenticated

    @authenticated.setter
    def authenticated(self, value):
        """Prevent direct setting - O(1)"""
        raise AttributeError("Cannot set authenticated directly")

# O(1) - use authentication
user = User("alice", "hash123")
is_auth = user.authenticated  # False

user.authenticate("mypassword")  # Authenticates if password matches

Edge Cases¶

Property Can't Have Same Name as init Param¶

# O(1) - attribute name matters
class Good:
    def __init__(self, value):
        self._value = value  # Different name

    @property
    def value(self):
        return self._value

# O(1) - works fine
obj = Good(42)
val = obj.value  # 42

Deleter Without Setter¶

# O(1) - can define deleter without setter
class OnlyDelete:
    def __init__(self):
        self.value = 42

    @property
    def value(self):
        return self._value if hasattr(self, '_value') else None

    @value.deleter
    def value(self):
        if hasattr(self, '_value'):
            del self._value

# O(1) - delete works
obj = OnlyDelete()
del obj.value

Property on Module¶

# O(1) - properties in classes, not modules
# (modules don't support property directly in Python 3.6+)

# But can be emulated with __getattr__
import sys

class ModuleProxy:
    def __init__(self, module):
        self.module = module

    @property
    def version(self):
        return "1.0.0"

# Access: ModuleProxy(sys).version

Performance Considerations¶

Avoiding Property Overhead¶

# If performance critical, use direct attribute:
class Fast:
    def __init__(self):
        self.value = 42

# Slightly faster than property (no function call)
obj = Fast()
val = obj.value  # Very fast

# vs property (function call overhead)
class WithProperty:
    def __init__(self):
        self._value = 42

    @property
    def value(self):
        return self._value

# Overhead is minimal (~5%), only matters in extreme cases

Caching Computed Properties¶

# O(1) - cache expensive computation
class Cached:
    def __init__(self):
        self._cached_value = None

    @property
    def value(self):
        """Compute once - O(1) after first access"""
        if self._cached_value is None:
            self._cached_value = expensive_computation()
        return self._cached_value

    def expensive_computation(self):
        return sum(range(1_000_000))

# First access does computation, subsequent accesses are fast

Best Practices¶

✅ Do:

Use for read-only computed attributes
Use to validate input in setters
Use to implement lazy loading
Use to provide clean attribute-like interface
Document properties in docstrings
Make getters fast (no I/O)

❌ Avoid:

Complex logic in property methods (keep simple)
I/O operations in getters (causes performance issues)
Properties with unexpected side effects
Using property when method is clearer
Expensive computations without caching
Hiding errors in property methods

classmethod() - Decorator for class methods
staticmethod() - Decorator for static methods
type() - Get object type
hasattr() - Check attribute existence
getattr() - Get attribute value

Version Notes¶

Python 2.x: @property available, same behavior
Python 3.x: Same behavior, optimized in CPython
All versions: Uses descriptor protocol for efficiency

property() Decorator Complexity¶

Complexity Analysis¶

Basic Usage¶

Simple Getter¶

Getter and Setter¶

Computed Property¶

Getter, Setter, and Deleter¶

Complexity Details¶

Descriptor Protocol¶

MRO Traversal¶

Property Override¶

Performance Patterns¶

Property vs Direct Attribute¶

Property vs Method Call¶

Caching in Properties¶

Common Use Cases¶

Validation on Set¶

Computed Properties¶

Lazy Loading¶

Read-Only Properties¶

Type Conversion¶

Advanced Usage¶

Property with Side Effects¶

Property Inheritance¶

Dynamic Properties¶

Practical Examples¶

Account Balance with Limits¶

Configuration Management¶

User Authentication¶

Edge Cases¶

Property Can't Have Same Name as init Param¶

Deleter Without Setter¶

Property on Module¶

Performance Considerations¶

Avoiding Property Overhead¶

Caching Computed Properties¶

Best Practices¶

Related Functions¶

Version Notes¶