Data serialization is the process of converting data into a format that can be stored, transmitted, and reconstructed later. Python provides multiple libraries for serialization, such as JSON, Pickle, YAML, and MessagePack.

---

1. Why Use Data Serialization?

Store and retrieve structured data
Send data over networks (e.g., APIs, databases)
Save application state
Inter-process communication

---

2. JSON Serialization in Python

JSON (JavaScript Object Notation) is a lightweight, human-readable format used for data exchange between applications.

a) Converting Python Objects to JSON (json.dumps())

import json

data = {"name": "Alice", "age": 25, "city": "New York"}
json_data = json.dumps(data)  # Convert Python dict to JSON string

print(json_data)  # Output: {"name": "Alice", "age": 25, "city": "New York"}

b) Writing JSON to a File (json.dump())

pythonCopyEditwith open("data.json", "w") as file:
    json.dump(data, file)

c) Reading JSON from a File (json.load())

with open("data.json", "r") as file:
    loaded_data = json.load(file)

print(loaded_data)  # Output: {'name': 'Alice', 'age': 25, 'city': 'New York'}

d) Converting JSON to a Python Object (json.loads())

json_string = '{"name": "Alice", "age": 25, "city": "New York"}'
python_obj = json.loads(json_string)

print(python_obj["name"])  # Output: Alice

e) Custom Serialization (Handling Non-Serializable Objects)

from datetime import datetime

class CustomEncoder(json.JSONEncoder):
    def default(self, obj):
        if isinstance(obj, datetime):
            return obj.isoformat()  # Convert datetime to string
        return super().default(obj)

data = {"timestamp": datetime.now()}
json_data = json.dumps(data, cls=CustomEncoder)

print(json_data)

---

3. Pickle Serialization (Python-Specific)

Pickle is used to serialize and deserialize Python objects (including custom classes).

a) Serializing with Pickle (pickle.dumps())

import pickle

data = {"name": "Alice", "age": 25}
serialized_data = pickle.dumps(data)  # Convert object to binary

print(serialized_data)

b) Writing Pickled Data to a File (pickle.dump())

with open("data.pkl", "wb") as file:
    pickle.dump(data, file)

c) Reading Pickled Data from a File (pickle.load())

with open("data.pkl", "rb") as file:
    loaded_data = pickle.load(file)

print(loaded_data)  # Output: {'name': 'Alice', 'age': 25}

Pickle Warning:

• Pickle is not secure for untrusted data (it can execute arbitrary code).
• Use it only for local storage or trusted environments.

---

4. YAML Serialization (Human-Friendly Format)

YAML is more readable than JSON and supports complex data types like dictionaries and lists.

a) Installing PyYAML

pip install pyyaml

b) Serializing Python Data to YAML (yaml.dump())

import yaml

data = {"name": "Alice", "age": 25, "skills": ["Python", "Data Science"]}

yaml_data = yaml.dump(data)
print(yaml_data)

c) Writing YAML to a File

with open("data.yaml", "w") as file:
    yaml.dump(data, file)

d) Loading YAML Data (yaml.load())

with open("data.yaml", "r") as file:
    loaded_data = yaml.safe_load(file)

print(loaded_data)  # Output: {'name': 'Alice', 'age': 25, 'skills': ['Python', 'Data Science']}

---

5. MessagePack Serialization (Efficient Binary Format)

MessagePack is a faster alternative to JSON for compact binary serialization.

a) Installing MessagePack

pip install msgpack

b) Serializing with MessagePack (msgpack.packb())

import msgpack

data = {"name": "Alice", "age": 25}
packed_data = msgpack.packb(data)

print(packed_data)  # Binary data

c) Deserializing MessagePack Data (msgpack.unpackb())

unpacked_data = msgpack.unpackb(packed_data)

print(unpacked_data)  # Output: {'name': 'Alice', 'age': 25}

---

6. CSV Serialization (For Tabular Data)

CSV (Comma-Separated Values) is used for handling structured tabular data.

a) Writing a Dictionary to CSV (csv.DictWriter())

import csv

data = [{"name": "Alice", "age": 25}, {"name": "Bob", "age": 30}]

with open("data.csv", "w", newline="") as file:
    writer = csv.DictWriter(file, fieldnames=["name", "age"])
    writer.writeheader()
    writer.writerows(data)

b) Reading CSV Data (csv.DictReader())

with open("data.csv", "r") as file:
    reader = csv.DictReader(file)
    for row in reader:
        print(row)

---

7. XML Serialization (For Web and Configurations)

XML (Extensible Markup Language) is commonly used in web applications.

a) Creating an XML String (xml.etree.ElementTree)

import xml.etree.ElementTree as ET

root = ET.Element("person")
ET.SubElement(root, "name").text = "Alice"
ET.SubElement(root, "age").text = "25"

tree = ET.ElementTree(root)
tree.write("data.xml")

b) Parsing XML Data

tree = ET.parse("data.xml")
root = tree.getroot()

for child in root:
    print(child.tag, ":", child.text)

---

8. Comparing Serialization Formats

Format	Readable	Speed	Best Use Case
JSON	Yes	Fast	APIs, Web Applications
Pickle	No	Fast	Python Object Storage
YAML	Yes	Slow	Configurations, Readable Data
MessagePack	No	Very Fast	Efficient Data Transfer
CSV	Yes	Fast	Tabular Data, Spreadsheets
XML	Yes	Slow	Web, Configuration Files