Skip to content

07-exercise-sources.md

Latest commit

 

History

History
206 lines (169 loc) · 12.9 KB

07-exercise-sources.md

File metadata and controls

206 lines (169 loc) · 12.9 KB
Raw

Exercise Source Repositories — Research

Summary

We researched open-source repositories with Python exercises suitable for absolute beginners. The goal: find pre-built, tested exercises with solutions that we can adapt for the Python Tutor skill's exercise bank.

Tier 1 — Best for Our Use Case

0. exercism/python (THE #1 FIND)

  • URL: https://github.com/exercism/python
  • Stars: ~2.4k
  • License: MIT (can freely adapt everything)
  • What it has: 146 exercises across 17 concept areas, with difficulty ratings from 1 to 10. Split into "concept exercises" (structured syllabus) and "practice exercises" (open-ended).
  • Topics: Variables, strings, conditionals, loops, functions, lists, dicts, classes, error handling, comprehensions, generators, and more — covers EVERYTHING in our curriculum.
  • Tests: YES — every single exercise has a full pytest test suite. Students write code to make tests pass.
  • Solutions: Yes (community solutions on exercism.org)
  • Why it's THE best: Purpose-built for education. MIT license. Graded difficulty (1-10). pytest-driven. Actively maintained. The test-driven format is EXACTLY our /learn check model. 146 exercises means we have more than enough for all 8 modules + extras. This is the motherlode.

1. jerry-git/learn-python3

  • URL: https://github.com/jerry-git/learn-python3
  • Stars: ~6.8k
  • License: MIT (can freely adapt)
  • What it has: Jupyter notebooks with exercises for: Strings, Numbers, Conditionals, Lists, Dictionaries, For loops, Functions, File I/O, Classes, Exceptions, Debugging, Modules
  • Tests: YES — includes pytest-based exercises. Dedicated "Testing with pytest" modules
  • Difficulty: Beginner + Intermediate sections
  • Solutions: Yes, notebooks include solutions
  • Why it's #1 for us: Perfect topic alignment with our curriculum. Has pytest tests. MIT license. Exercises are structured as notebook → exercise pairs. We can extract exercise logic and adapt into our .py file format.

2. trekhleb/learn-python

  • URL: https://github.com/trekhleb/learn-python
  • Stars: ~17.8k
  • License: MIT
  • What it has: Python scripts split by topic, full of inline assertions. Covers: operators, data types, strings, lists, dicts, tuples, sets, control flow, functions, classes, modules, errors, files
  • Tests: YES — uses pytest framework. Code is full of assertions that serve as self-documenting tests
  • Difficulty: Not explicitly graded, but organized logically from simple to complex
  • Solutions: Yes, the scripts ARE the solutions with assertions
  • Why it's great: The assertion-based approach is exactly what we need — we can extract assertion patterns and reverse them into test files. Comprehensive topic coverage.

3. Asabeneh/30-Days-Of-Python

  • URL: https://github.com/Asabeneh/30-Days-Of-Python
  • Stars: ~45k+
  • License: Open source (no explicit license file)
  • What it has: 30-day curriculum: Day 1 (Intro) through Day 30 (Conclusions). Topics: Variables, Operators, Strings, Lists, Tuples, Sets, Dictionaries, Conditionals, Loops, Functions, Modules, List Comprehension, Exception Handling, File Handling, Classes, APIs
  • Tests: No automated tests — exercises are checked manually
  • Difficulty: YES — Level 1, Level 2, Level 3 per day
  • Solutions: Partial — examples in the lessons, but exercises don't have provided solutions
  • Why it's great: The graded difficulty system (Level 1-3) is exactly what we want. The curriculum structure is almost identical to ours. We can use their exercise prompts and add our own tests + solutions.

NEW — 4GeeksAcademy/python-beginner-programming-exercises

  • URL: https://github.com/4GeeksAcademy/python-beginner-programming-exercises
  • License: Open source (4Geeks Academy)
  • What it has: Interactive, auto-graded beginner exercises. Topics: print, data types, lists/tuples, functions/dictionaries. Step-by-step increasing difficulty.
  • Tests: YES — each exercise has a test.py file with automated tests. Structure: app.py (student writes) + README.md (instructions) + test.py (auto-grading)
  • Solutions: Partial
  • Why it's critical: The app.py + test.py per-exercise structure is EXACTLY our model. This is the closest existing implementation to what we're building. Has video tutorials too.
  • Note: Tests are "very rigid and strict" — we should make ours more flexible.

NEW — gregmalcolm/python_koans

  • URL: https://github.com/gregmalcolm/python_koans
  • Stars: ~5.1k
  • License: MIT
  • What it has: Port of Ruby Koans — learn Python through TDD. Fix failing tests to learn. Covers: assertions, tuples, strings, lists, logic problems.
  • Tests: YES — the entire teaching method IS tests. Run contemplate_koans.py → see failing test → fix it → learn.
  • Approach: "Red, Green, Refactor" — see failure, make it pass, reflect.
  • Why it's great: The fill-in-the-assert approach (assert __ == 4) is brilliant for our "predict the output" exercises. The TDD-first teaching philosophy aligns perfectly with our test-driven feedback loop. MIT license.

NEW — ELC/python-basics

  • URL: https://github.com/ELC/python-basics
  • License: Open source
  • What it has: Autograding template for Python assignments. Each exercise has function signature + docstring, student implements. Tests in separate tests/ folder.
  • Tests: YES — designed for GitHub Classroom autograding, also works standalone with python exercise/<exercise>.py
  • Topics: IF blocks, logical operators, chained comparisons, recursion, boolean expressions, FOR loops (range, len, enumerate, zip)
  • Why useful: Clean separation of exercise/tests structure we can directly adapt.

Tier 2 — Good Supplementary Sources

4. donnemartin/interactive-coding-challenges

  • URL: https://github.com/donnemartin/interactive-coding-challenges
  • Stars: ~30k+
  • License: Personal open source
  • What it has: 120+ interactive Python challenges in Jupyter notebooks. Algorithm/data structure focus.
  • Tests: YES — every challenge has built-in unit tests
  • Solutions: Yes, solution notebooks provided
  • Why useful: Great test patterns to adapt. The Jupyter Challenge → Solution structure is a model for our exercise → test approach. However, topics skew toward algorithms/interviews rather than absolute beginner basics.

5. zhiwehu/Python-programming-exercises + darkprinx/break-the-ice-with-python

6. karan/Projects

  • URL: https://github.com/karan/Projects
  • Stars: ~47k+
  • License: MIT
  • What it has: ~100+ practical project ideas categorized by: Numbers, Text, Networking, Classes, Files, Databases, Graphics, Security
  • Tests: No
  • Solutions: User-contributed via links
  • Why useful: Excellent for our PROJECT exercises (not individual lessons). Many overlap with our project ideas: Calculator, Mortgage Calculator, Fibonacci, Hangman, Address Book. Can mine for project descriptions and requirements.

Tier 3 — Reference / Inspiration Only

7. TheAlgorithms/Python

  • URL: https://github.com/TheAlgorithms/Python
  • Stars: ~195k+
  • License: MIT
  • What it has: Hundreds of algorithm implementations with doctests
  • Why limited: Too advanced for absolute beginners. Useful as a reference for "what good Python code looks like" in later modules (7-8).

8. geekcomputers/Python

  • URL: https://github.com/geekcomputers/Python
  • Stars: ~33k+
  • License: MIT
  • What it has: 41+ educational Python scripts for beginners (batch renamers, downloaders, etc.)
  • Why limited: Scripts are example programs, not structured exercises. Good for "here's what Python can do" inspiration during onboarding.

9. freeCodeCamp/freeCodeCamp

  • URL: https://github.com/freeCodeCamp/freeCodeCamp
  • Stars: ~410k+
  • License: BSD-3-Clause (software), Copyright (curriculum)
  • What it has: Full Python certification with interactive challenges
  • Why limited: Curriculum content is copyrighted. The platform is web-based, not file-based. However, the curriculum STRUCTURE is useful as a reference for ordering topics.

10. mbucko/openleetcode

  • URL: https://github.com/mbucko/openleetcode
  • Stars: Small
  • License: Not stated
  • What it has: Open-source LeetCode clone with test framework
  • Why limited: Currently C++ only. But the .test file format for auto-verification is an interesting model we could adapt.

Additional Finds (from deep web research)

  • bregman-arie/python-exercises — 231 exercises in Q&A format by topic. No tests, no license. Good for exercise prompts.
  • realpython/python-basics-exercises — Companion to Real Python book. Well-structured by chapter. No license specified.
  • Pierian-Data/Complete-Python-3-Bootcamp (~29.5k stars) — Jupyter exercise/solution notebooks from Udemy bootcamp. No license.
  • quobit/awesome-python-in-education (CC0) — Meta-list of Python education resources. Good for discovering more.

How to Use These Sources

Structural Models to Adopt

Source What to Copy
exercism/python pytest test suites per exercise, difficulty ratings, concept syllabus — THE primary source
4GeeksAcademy app.py + test.py per-exercise folder structure — almost identical to our model
python_koans Fill-in-the-assert pattern for "predict the output" exercises
ELC/python-basics Function signature + docstring as exercise scaffold
trekhleb/learn-python Assertion-heavy scripts as self-documenting test patterns

For Exercise Bank (Modules 1-3)

Our Lesson Primary Source What to Extract
01 Hello World 4GeeksAcademy + exercism "hello-world" Exercise structure, auto-grading pattern
02 Variables exercism concept exercises + 30-Days Day 2 pytest patterns + exercise prompts
03 Strings exercism "two-fer", "bob" + learn-python3 String exercises with pytest tests
04 Numbers exercism "currency-exchange" + learn-python3 Number exercises + tests
05 Input 30-Days Day 2 + 4GeeksAcademy Interactive exercise patterns
06 Booleans python_koans + exercism "ghost-gobble" Assert fill-in-the-blank + predict output
07-10 Conditionals exercism "meltdown-mitigation" + ELC/python-basics Conditional exercises with pytest
11-15 Loops exercism "making-the-grade" + ELC/python-basics + 30-Days Day 10 Loop exercises + enumerate/zip

For Solution Variants

  • darkprinx/break-the-ice-with-python — shows multiple solution approaches per problem
  • trekhleb/learn-python — clean assertion-based solutions as reference
  • learn-python3 — pytest solutions as test patterns

For Project Ideas

  • karan/Projects — practical project descriptions (MIT license)
  • 30-Days-Of-Python — project-level exercises in later days
  • practical-tutorials/project-based-learning — full project tutorials

For Test Patterns

  1. trekhleb/learn-python — inline assertions (adapt to our test files)
  2. donnemartin/interactive-coding-challenges — unit test structure
  3. learn-python3 — pytest exercises (closest to our format)

License Summary

Repo License Can we adapt exercises?
exercism/python MIT YES — freely. Our #1 source.
4GeeksAcademy/python-beginner-exercises Open source YES — adapt structure + test patterns
gregmalcolm/python_koans MIT YES — freely adapt assert-fill patterns
ELC/python-basics Open source YES — adapt exercise/test separation
jerry-git/learn-python3 MIT YES — freely
trekhleb/learn-python MIT YES — freely
Asabeneh/30-Days-Of-Python Unspecified Exercise prompts OK, don't copy verbatim
karan/Projects MIT YES — freely
donnemartin/interactive-coding-challenges Personal OSS Adapt patterns, credit author
zhiwehu/Python-programming-exercises Unspecified Use as inspiration, write our own
TheAlgorithms/Python MIT YES — for later modules
geekcomputers/Python MIT YES

NotebookLM Research Notebook

All sources loaded and queryable at: https://notebooklm.google.com/notebook/a0cb9549-82f9-4f9d-bd78-df359fd4c1aa

Deep web research was also initiated (may still be processing). Results will be imported into this notebook when complete.