bigopy 0.1.0

Static AST-based Big-O complexity estimator — no code execution needed

bigopy 🔍

Static Big-O complexity estimation for Python code via AST analysis. No code execution needed — just point it at your source file.

bigopy statically analyzes Python source files using the Abstract Syntax Tree (AST) and estimates the Big-O time complexity of every function — without running your code.

$ bigopy analyze bubble_sort.py

============================================================
📂  bubble_sort.py
============================================================

  Function:      bubble_sort()  (line 1)
  Complexity:    O(n²)
  Confidence:    [█████████████████░░░] 85%
  Reasons:
    • doubly-nested loops (depth=2)

---

Why bigopy is different

	bigopy	Other tools
Approach	Static AST analysis	Runtime measurement
Runs your code?	❌ Never	✅ Must execute
Works on broken code?	✅ Yes	❌ No
Dependencies	Zero	numpy, etc.
Graph algorithms	✅ O(V+E), O((V+E)logV)	❌ No
Safe for CI/CD	✅ Yes	⚠️ Maybe

---

Installation

pip install bigopy

---

Quick Start

CLI

bigopy analyze my_module.py
bigopy analyze my_module.py --verbose
bigopy analyze src/ --format json
bigopy analyze my_module.py --only bubble_sort

Python API

from bigopy import analyze_file, analyze_source

# From a file
result = analyze_file("my_module.py")
for func in result.functions:
    print(f"{func.name}: {func.complexity.label} ({func.confidence:.0%})")

# From a string
source = """
def bubble_sort(arr):
    n = len(arr)
    for i in range(n):
        for j in range(n - i - 1):
            if arr[j] > arr[j + 1]:
                arr[j], arr[j + 1] = arr[j + 1], arr[j]
"""
result = analyze_source(source)
print(result.functions[0].complexity.label)  # O(n²)

---

Detected Complexity Classes

Class	Pattern Detected
O(1)	No loops, no recursion
O(log n)	Halving/doubling while loops
O(n)	Single loop over input
O(n log n)	Divide-and-conquer, sorted()
O(n²)	Doubly-nested loops
O(n³)	Triple loops, range(i*i)
O(2^n)	Branching recursion
O(V+E)	DFS, BFS graph traversal
O((V+E) log V)	Dijkstra, Prim

---

Project Structure

bigopy/
├── bigopy/
│   ├── analyzers/      # estimation engine
│   ├── detectors/      # loop, recursion, graph, builtin
│   └── reporters/      # terminal, json output
├── tests/              # test suite
├── examples/           # demo algorithms
└── README.md

---

License

MIT © 2024 Zul-Qarnain