Python Functional Programming Guide

A practical guide to functional programming concepts in Python, organized into dedicated modules for easy learning and reference.

TL;DR - Quick Start

Run all demonstrations:

bash run.sh

This will build a Docker container and execute all functional programming demonstrations.

Modules Description

1. immutability.py - Immutable Data Structures

Learn how to work with immutable data in Python:

• Tuples: Immutable sequences
• Frozenset: Immutable sets
• Namedtuple: Immutable objects with named fields
• Frozen Dataclass: Immutable dataclasses (Python 3.7+)
• MappingProxyType: Immutable dictionaries

Key Takeaway: Python isn't purely functional, so immutability requires discipline.

2. map_filter_reduce.py - Map, Filter, Reduce

Functional transformations and reductions:

• Map: Transform each element in a collection
• Filter: Select elements matching a predicate
• Reduce: Accumulate values (fold operation)
• Pythonic alternatives using list comprehensions

Key Takeaway: List comprehensions are often preferred in Python for readability.

3. lambdas.py - Lambda Expressions

Using anonymous functions effectively:

• Basic lambda syntax
• Lambdas in sorting operations
• Chaining lambdas with map/filter
• Best practices and limitations

Key Takeaway: Use lambdas for short, simple expressions; use def for complex logic.

4. currying.py - Currying and Partial Application

Function transformation techniques:

• Manual Currying: Converting multi-argument functions to single-argument chains
• Partial Application: Pre-filling function arguments with functools.partial
• General curry function implementation

Key Takeaway: functools.partial() is the pythonic way to handle partial application.

5. higher_order_functions.py - Higher-Order Functions

Functions as first-class citizens:

• Functions as arguments
• Functions as return values
• Function Composition: Combining functions
• Decorators: Pythonic higher-order functions
• Memoization: Caching function results

Key Takeaway: Very pythonic! Decorators are HOFs and widely used in Python.

6. transducers.py - Transducers

Composable algorithmic transformations:

• Traditional multi-pass vs. single-pass transformations
• Transducer pattern implementation
• Pythonic alternatives with itertools

Key Takeaway: Not idiomatic in Python; use itertools for lazy, composable transformations.

7. monads.py - Monads

Handling context in a composable way:

• Maybe/Option Monad: Handle None/null safely
• Either/Result Monad: Error handling without exceptions
• Function pipelines
• Pythonic alternatives

Key Takeaway: Not idiomatic in Python, but useful patterns for specific use cases.

Running the Project

Run All Demonstrations

bash run.sh

Run Individual Modules

Each module can be run independently:

python -m modules.immutability
python -m modules.map_filter_reduce
python -m modules.lambdas
python -m modules.currying
python -m modules.higher_order_functions
python -m modules.transducers
python -m modules.monads

Run via Main Entry Point

python main.py

Key Takeaways

1. IMMUTABILITY: Use tuples, frozenset, namedtuple, or dataclass(frozen=True)
2. MAP/FILTER/REDUCE: List comprehensions often preferred in Python
3. LAMBDAS: Best for short, simple expressions
4. CURRYING/PARTIAL: functools.partial() is pythonic
5. HIGHER-ORDER FUNCTIONS: Very pythonic! Decorators are HOFs
6. TRANSDUCERS: Not idiomatic; use itertools instead
7. MONADS: Not idiomatic, but useful patterns exist

Recommendation

Use Python's functional features where they improve clarity, but don't force purely functional patterns. Python is multi-paradigm - embrace comprehensions, generators, and itertools for pythonic functional style.

Requirements

• Python 3.11+
• Docker (for containerized execution)
• UV package manager (automatically installed in Docker)

License

MIT