Introduction to Python Libraries: attrs, bidict, Box, dataclasses, and DottedDict

1. attrs

attrs is a library that simplifies Python class definitions by automatically generating __init__, __repr__, __eq__, and other special methods.

Installation:

pip install attrs

Example – basic usage:

import attr
@attr.s
class Person:
    name = attr.ib()
    age = attr.ib()

p = Person(name="Alice", age=30)
print(p)  # Output: Person(name='Alice', age=30)

Example – default values:

@attr.s
class Person:
    name = attr.ib(default="Unknown")
    age = attr.ib(default=0)

p = Person(age=30)
print(p)  # Output: Person(name='Unknown', age=30)

Example – validator:

def positive_number(instance, attribute, value):
    if value < 0:
        raise ValueError("Age must be a positive number")

@attr.s
class Person:
    name = attr.ib()
    age = attr.ib(validator=positive_number)

try:
    p = Person(name="Alice", age=-1)
except ValueError as e:
    print(e)  # Output: Age must be a positive number

Example – converter:

@attr.s
class Person:
    name = attr.ib(converter=str)
    age = attr.ib(converter=int)

p = Person(name=123, age="30")
print(p)  # Output: Person(name='123', age=30)

Example – frozen instance:

@attr.s(frozen=True)
class Person:
    name = attr.ib()
    age = attr.ib()

p = Person(name="Alice", age=30)
try:
    p.age = 31
except attr.exceptions.FrozenInstanceError as e:
    print(e)  # Output: can't set attributes of frozen Person

Use case: Quickly define data classes for configuration objects, DTOs, etc.

2. bidict

bidict provides a bidirectional mapping where keys can be accessed as values and vice versa.

Installation:

pip install bidict

Example – basic usage:

from bidict import bidict
b = bidict({"a": 1, "b": 2})
print(b["a"])      # Output: 1
print(b.inv[1])    # Output: 'a'

Example – add new item:

b = bidict({"a": 1, "b": 2})
b["c"] = 3
print(b)  # Output: bidict({'a': 1, 'b': 2, 'c': 3})

Example – delete item:

b = bidict({"a": 1, "b": 2})
del b["a"]
print(b)  # Output: bidict({'b': 2})

Example – existence check:

b = bidict({"a": 1, "b": 2})
print("a" in b)      # Output: True
print(1 in b.inv)    # Output: True

Example – update multiple items:

b = bidict({"a": 1, "b": 2})
b.update({"c": 3, "d": 4})
print(b)  # Output: bidict({'a': 1, 'b': 2, 'c': 3, 'd': 4})

Use case: Useful for bidirectional look‑ups in NLP, database reverse mapping, etc.

3. Box

Box is a Python dictionary with advanced dot‑notation access, making nested data handling more convenient.

Installation:

pip install python-box

Example – basic usage:

from box import Box

data = {
    "person": {
        "name": "Alice",
        "age": 30,
        "address": {
            "street": "123 Main St",
            "city": "Anytown"
        }
    }
}
box_data = Box(data)
print(box_data.person.name)  # Output: Alice

Modify nested data:

box_data.person.age = 31
print(box_data)  # Shows updated age

Add new field:

box_data.person.email = "[email protected]"
print(box_data)  # Shows email added

Load data from a JSON file:

import json
with open('data.json') as f:
    data = json.load(f)
box_data = Box(data)
print(box_data.person.name)  # Output: Alice

Convert back to a plain dict:

dict_data = box_data.to_dict()
print(dict_data)  # Output: standard dictionary

Use case: Ideal for configuration management and JSON data manipulation.

4. dataclasses

dataclasses is part of the Python standard library that simplifies the creation of data classes by auto‑generating __init__, __repr__, __eq__, and more.

Example – basic usage:

from dataclasses import dataclass

@dataclass
class Person:
    name: str
    age: int

p = Person(name="Alice", age=30)
print(p)  # Output: Person(name='Alice', age=30)

Example – default values:

@dataclass
class Person:
    name: str = "Unknown"
    age: int = 0

p = Person(age=30)
print(p)  # Output: Person(name='Unknown', age=30)

Example – object comparison:

p1 = Person(name="Alice", age=30)
p2 = Person(name="Alice", age=30)
print(p1 == p2)  # Output: True

Example – immutable (frozen) dataclass:

@dataclass(frozen=True)
class Person:
    name: str
    age: int

p = Person(name="Alice", age=30)
try:
    p.age = 31
except FrozenInstanceError as e:
    print(e)  # Output: cannot assign to field 'age'

Example – inheritance:

@dataclass
class Employee(Person):
    employee_id: int

e = Employee(name="Bob", age=35, employee_id=123)
print(e)  # Output: Employee(name='Bob', age=35, employee_id=123)

Use case: Perfect for simple DTOs, configuration objects, and any scenario needing concise data containers.

5. DottedDict

DottedDict offers lightweight dot‑path access to nested dictionaries and lists, similar to Box.

Installation:

pip install dotted-dict

Example – basic usage:

from dotted_dict import DottedDict

data = {
    "person": {
        "name": "Alice",
        "age": 30,
        "address": {
            "street": "123 Main St",
            "city": "Anytown"
        }
    }
}

dotted_data = DottedDict(data)
print(dotted_data.get("person.name"))  # Output: Alice

Modify nested data:

dotted_data.set("person.age", 31)
print(dotted_data.to_dict())  # Shows updated age

Add new field:

dotted_data.set("person.email", "[email protected]")
print(dotted_data.to_dict())  # Shows email added

Load from JSON file:

import json
with open('data.json') as f:
    data = json.load(f)

dotted_data = DottedDict(data)
print(dotted_data.get("person.name"))  # Output: Alice

Convert back to dict:

dict_data = dotted_data.to_dict()
print(dict_data)  # Output: standard dictionary

Use case: Suitable for frequent nested dict/list manipulation in configuration management and JSON processing.