Practical Data‑Structure Implementations in Pure Python

I co‑authored datastructpy as part of UBC MDS’s DSCI 524: Reproducible Software course. The goal: ship a Python package that offers clean, interview‑ready implementations of classic data structures while demonstrating software‑engineering best practices—versioning, automated tests, CI/CD, and comprehensive docs—all in under four weeks. 

Highlights

What we built Why it matters Binary Search Tree (BST) with full CRUD (insert, search, delete) plus DFS/BFS traversals. Gives learners a hands-on, dependency-light way to explore tree operations and complexity analysis. Modular, poetry‑managed codebase (src/datastructpy/…). Clear separation of concerns and packaging hygiene mirror production Python projects. Pytest suite + Codecov badge. > 95% statement coverage ensures every method behaves as advertised and prevents regressions on PRs. GitHub Actions workflow (lint ⟶ test ⟶ build ⟶ publish). Every push runs linting, unit tests, and publishes the wheel to TestPyPI for repeatable releases. Sphinx docs on Read‑the‑Docs. Auto‑generated API reference, tutorials, and example notebooks keep the learning curve shallow.

Package Architecture

datastructpy/
├─ src/datastructpy/
│  └─ non_linear/trees/
│     └─ binary_search_tree.py
├─ tests/
│  └─ non-linear/trees/binary_search_tree/
│     ├─ test_insert.py
│     ├─ test_search.py
│     └─ …
├─ docs/               ← Sphinx source
├─ pyproject.toml      ← poetry config + semver
└─ .github/workflows/  ← CI pipeline

Tech Stack

Python, Poetry, Pytest, PyPI, Tox, GitHub Actions, Codecov, Sphinx/Read‑the‑Docs