codebase-stats 0.0.1

Comprehensive codebase analysis library with coverage, metrics, and test duration analysis

Codebase Stats

Production-ready code quality metrics analysis library for Python projects. Comprehensive analysis of coverage, complexity, maintainability, and code structure with automated reporting and quality gates.

Quick Start

Installation

# Using uv (recommended)
uv pip install codebase-stats

# Or with pip
pip install codebase-stats

Basic Usage

from codebase_stats import CodebaseStatsReporter

# Generate comprehensive metrics report
reporter = CodebaseStatsReporter(
    coverage_file='coverage.json',
    report_file='report.json',
    radon_root='src',
    fs_root='src',
    tree_root='src'
)

# Save report to file
reporter.save_report('metrics_report.txt', include_coverage=True, include_complexity=True)

Command Line Interface

# Full analysis with all metrics
python cli.py coverage.json --radon-root src --fs-root src

# Show specific sections
python cli.py coverage.json --show coverage complexity mi

# List low-coverage files
python cli.py coverage.json --show list --threshold 80 --top 20

# Help
python cli.py --help

Features

📊 Coverage Analysis

• Statement & Branch Coverage: Full coverage metrics from pytest-cov
• Coverage Distribution: Histogram visualization with percentiles (Q1, Q2, Q3, p90, p95, p99)
• Low-Coverage Detection: Identify files below thresholds with automatic prioritization
• Pragma Tracking: Track # pragma: no cover usage for documentation

🔍 Code Complexity

• Cyclomatic Complexity (CC): Radon integration for function complexity analysis
  • Grade A: 1-5 (ideal)
  • Grade B: 6-10 (acceptable)
  • Grade C+: 11+ (refactor recommended)
• Maintainability Index (MI): Code readability/maintainability scoring (0-100)
• Halstead Metrics: Bug estimation and code volume analysis
• Comment Ratios: Documentation density analysis

📈 Code Metrics

• File Size Distribution: Line count per file with outlier detection
• Directory Structure Analysis: Module organization and hierarchy
• Test Duration Distribution: Identify slow tests
• Quality Gates: Automated threshold validation

📋 Reporting

• Histogram Visualization: ASCII histograms with customizable bins and scaling
• Blame Sections: Highlight problematic files (Q3 + 1.5×IQR threshold)
• Percentile Analysis: Q1, median, Q3, p90, p95, p99
• Structured Output: Markdown, text, or programmatic JSON

Documentation

• Architecture Documentation - System design, data flows, and component details
• Development Roadmap - Release planning and feature roadmap
• Architecture Decision Records - Design decisions and rationales
• Governance & Quality Gates - Development workflow and CI/CD policies

API Reference

Core Classes

CodebaseStatsReporter

Main interface for generating comprehensive metrics reports.

from codebase_stats import CodebaseStatsReporter

reporter = CodebaseStatsReporter(
    coverage_file: str,           # Path to coverage.json
    report_file: str = None,      # Path to pytest report.json (optional)
    radon_root: str = None,       # Root for CC/MI/comment analysis
    fs_root: str = None,          # Root for file size analysis
    tree_root: str = None         # Root for structure analysis
)

# Methods
reporter.save_report(filename, include_coverage=True, include_complexity=True)
reporter.get_stats() -> dict      # Get raw metrics dictionary
report_text = str(reporter)       # Generate text report

Data Structure

Coverage Stats Dictionary

stats = {
    "coverages_sorted": [float],                    # All file coverage %
    "proj_pct": float,                              # Project-wide coverage %
    "proj_total": int,                              # Total lines
    "proj_covered": int,                            # Covered lines
    "file_stats": [
        {
            "pct": float,                           # File coverage %
            "path": str,                            # File path
            "missing_count": int,                   # Missing line count
            "missing_lines": [int],                 # Missing line numbers
            "cc_avg": float,                        # Avg cyclomatic complexity
            "mi": float,                            # Maintainability index
            "comment_ratio": float,                 # Comment/SLOC ratio
            "hal_bugs": float,                      # Halstead bug estimate
            "size_lines": int,                      # File line count
            ...
        }
    ]
}

Module APIs

coverage.py - Coverage Analysis

from codebase_stats.coverage import load_coverage, precompute_coverage_stats

stats = load_coverage('coverage.json')
stats = precompute_coverage_stats(stats, radon_root='src')

metrics.py - Complexity Metrics

from codebase_stats.metrics import get_cyclomatic_complexity, get_maintainability, get_comments_ratio

cc = get_cyclomatic_complexity('file.py')
mi = get_maintainability('file.py')
ratio = get_comments_ratio('file.py')

radon.py - Radon Integration

from codebase_stats.radon import get_cc_list, get_mi_list, get_metrics

cc_data = get_cc_list('src')
mi_data = get_mi_list('src')
hal_data = get_metrics('src')

reporter.py - Report Generation

from codebase_stats.reporter import CodebaseStatsReporter

reporter = CodebaseStatsReporter(...)
reporter.save_report('output.txt')  # Save formatted report

utils.py - Utilities

from codebase_stats.utils import ascii_histogram, percentile, format_value

hist_str = ascii_histogram(data, bins=10, width=80)
p95 = percentile(data, 0.95)
formatted = format_value(value, decimals=2)

Quality Gates

All code must meet these thresholds:

Metric	Threshold	Rationale
Coverage	100%	All source code must be tested
Cyclomatic Complexity	≤10 average	Grade B maintainability
Maintainability Index	≥50	Grade A minimum
File Size	≤400 lines	Modules remain manageable

Development Workflow

Setup

# Clone repository
git clone https://github.com/brunolnetto/codebase-stats.git
cd codebase-stats

# Install with dev dependencies
uv venv
uv pip install -e ".[dev]"

# Activate environment
source .venv/bin/activate

Testing

# Run all tests
pytest

# With coverage
pytest --cov=codebase_stats --cov-report=term-plus

# Specific test file
pytest tests/test_coverage.py -v

Code Quality

# Linting
ruff check codebase_stats/ tests/

# Format check
ruff format --check codebase_stats/ tests/

# Type checking
mypy codebase_stats/

# All quality checks
make quality

Commits & PRs

This project uses GitFlow workflow with conventional commits:

# Feature branches
git checkout -b feat/feature-name
# Fix branches  
git checkout -b fix/issue-name
# Chore/documentation
git checkout -b chore/update-name

# Commit format: <type>(<scope>): <description>
git commit -m "feat(coverage): add pragma tracking"
git commit -m "fix(radon): handle empty files gracefully"
git commit -m "docs(readme): add API reference"

See Development Policy for full workflow details.

Architecture Highlights

Data Flow:

Raw Input (coverage.json, report.json)
    ↓
load_coverage() → Enrich with Radon metrics
    ↓
precompute_coverage_stats() → Compute distributions
    ↓
Display Functions → Histograms, tables, blame sections
    ↓
Reporter → Formatted text/markdown output

Key Design Patterns:

• Single Responsibility: Each module handles one analysis type
• Composition: Reporter combines multiple analysis modules
• Lazy Evaluation: Radon metrics computed on-demand
• Immutable Data: Stats dicts treated as read-only
• Histogram Abstraction: Consistent visualization across metrics

See Architecture Documentation for detailed system design.

Examples

Generate Full Metrics Report

# After running tests
pytest --cov=src --cov-report=json

# Generate and save report
python cli.py coverage.json \
  --radon-root src \
  --fs-root src \
  --tree-root src \
  --report metrics_report.txt

Analyze Coverage Gaps

python cli.py coverage.json --show coverage gaps

Monitor Complexity Trends

from codebase_stats import CodebaseStatsReporter

def check_complexity_trend():
    reporter = CodebaseStatsReporter('coverage.json', radon_root='src')
    stats = reporter.get_stats()
    
    cc_values = [f['cc_avg'] for f in stats['file_stats']]
    avg_cc = sum(cc_values) / len(cc_values)
    
    if avg_cc > 10:
        print(f"⚠️  Complexity increasing: {avg_cc:.2f}")
    else:
        print(f"✅ Complexity healthy: {avg_cc:.2f}")

Contributing

1. See Development Roadmap for planned work
2. Submit PRs against develop branch (GitFlow)
3. All PRs require passing quality gates and 100% test coverage
4. Follow Governance for workflow details

License

MIT License - see LICENSE file for details

Metrics Definitions

Metric	Range	Interpretation
Coverage	0-100%	Percentage of code lines executed by tests
CC	1-50+	Function branching complexity; A≤5, B≤10, C≤15, D≤20, E/F>20
MI	0-100	Code readability/maintainability; A≥20, B≥10, C≥0
Comment Ratio	0-100%	Percentage of code that is comments/docstrings
Halstead Bugs	0-N	Estimated number of bugs; lower is better
File Size	Lines	Module size; target ≤400 for maintainability

---

Status: Production-ready · Latest Release: 1.0.0 · Python: 3.12+