GitHub Repository PR Analysis Tools

This collection of tools allows you to analyze pull request metrics across all repositories for a GitHub organization or user, including private repositories you have access to.

Note: All commands should be run from the project directory where these scripts are located.

🚀 Quick Start

1. Set up authentication and test access:

./setup_analysis.sh

2. Analyze all repositories for an organization or user:

python3 analyze_repos.py your-company-name    # Organization
python3 analyze_repos.py johndoe              # User

3. Export results to CSV for further analysis:

python3 analyze_repos.py your-account --export-csv results.csv

📊 What You'll Get

Account-Wide Metrics:

• Total repositories analyzed (organization or user)
• Overall PR merge rate across all repositories
• Language distribution (which programming languages are most used)
• PR size patterns (single commit vs multi-commit PRs)
• Most active repositories by PR volume
• Average commits per PR across all repositories

Per-Repository Metrics:

• Total pull requests and merge rate
• Average and median commits per PR
• PR size distribution (single, small, medium, large)
• Repository metadata (language, stars, forks)
• Last push date and activity level

🛠 Tools Included

1. analyze_repos.py - Main Analysis Script

The comprehensive tool that discovers and analyzes all repositories for an organization or user.

Key Features:

• Discovers all repositories you have access to (private and public)
• Analyzes PR patterns for each repository
• Generates organization-wide summary reports
• Exports detailed results to CSV
• Handles rate limiting and timeouts gracefully
• Filters out inactive repositories (optional)

2. setup_analysis.sh - Setup Helper

An interactive script that helps with authentication and testing.

Features:

• Guides you through GitHub CLI authentication
• Tests organization access before running analysis
• Provides usage examples and options
• Can run the analysis directly with proper setup

3. github_pr_metrics.py - Single Repository Analysis

For analyzing individual repositories in detail.

4. git_commit_analysis.py - Local Git History Analysis

For analyzing commit patterns in local repositories.

5. analyze_repo.sh - Single Repository Wrapper

Easy-to-use script for analyzing individual repositories.

📋 Usage Examples

Basic Analysis

# Analyze all repositories for an organization
python3 analyze_repos.py mycompany

# Analyze all repositories for a user
python3 analyze_repos.py johndoe

# Only analyze private repositories
python3 analyze_repos.py mycompany --private-only

# Quick test with first 5 repositories only
python3 analyze_repos.py johndoe --repo-limit 5

Advanced Options

# Analyze with custom PR limit per repo and export to CSV
python3 analyze_repos.py mycompany --limit 200 --export-csv detailed_results.csv

# Just list repositories without analyzing (useful for testing)
python3 analyze_repos.py johndoe --skip-analysis

# Analyze only public repositories for a user
python3 analyze_repos.py johndoe --public-only

# Analyze only private repositories for an organization
python3 analyze_repos.py mycompany --private-only

Using the Setup Helper

# Interactive setup and guidance
./setup_analysis.sh

# Test access to an organization or user
./setup_analysis.sh --test mycompany
./setup_analysis.sh --test johndoe

# Run analysis with setup verification
./setup_analysis.sh mycompany --private-only --export-csv results.csv
./setup_analysis.sh johndoe --export-csv user_results.csv

📄 Sample Output

Organization Summary Report:

================================================================================
ORGANIZATION SUMMARY: mycompany
================================================================================
Repositories analyzed: 25
Repositories with PRs: 23
Total pull requests: 1,247
Total merged PRs: 1,089
Organization merge rate: 87.3%

Top languages:
  JavaScript: 8 repositories
  Python: 6 repositories
  TypeScript: 5 repositories
  Go: 3 repositories
  Java: 2 repositories

PR Size Distribution (organization-wide):
  Single commit: 423 (34.9%)
  Small (2-3 commits): 512 (42.2%)
  Medium (4-10 commits): 201 (16.6%)
  Large (10+ commits): 76 (6.3%)

Average commits per PR (across all repos): 2.8

Most active repositories (by PR count):
  web-frontend: 234 PRs, 198 merged
  api-service: 187 PRs, 165 merged
  data-pipeline: 156 PRs, 142 merged
  mobile-app: 143 PRs, 128 merged
  auth-service: 98 PRs, 87 merged

CSV Export Fields:

The CSV export includes these fields for each repository:

• repo_name, full_name, language, stars, forks
• total_prs, merged_prs, merge_rate
• avg_commits_per_pr, median_commits_per_pr
• single_commit_prs, small_prs, medium_prs, large_prs
• last_push, analysis_date

🔧 Setup Requirements

Prerequisites:

1. GitHub CLI (gh) - Install from https://cli.github.com/
2. Python 3.6+ - Available on most systems
3. GitHub Authentication - Set up with proper permissions

Authentication Setup:

# Authenticate with GitHub CLI (includes org access)
gh auth login --scopes "repo,read:org"

The authentication needs these permissions:

• repo - To access private repositories
• read:org - To list organization repositories

⚡ Performance Notes

• Rate Limiting: The script includes delays and timeouts to respect GitHub API limits
• Sampling: For repositories with many PRs, it samples the first 20 merged PRs for detailed analysis
• Filtering: Only analyzes repositories updated in the last year by default
• Parallel Processing: Each repository is analyzed sequentially to avoid rate limits

🔍 Troubleshooting

Common Issues:

1. "GitHub CLI not authenticated"

   gh auth login --scopes "repo,read:org"

2. "Cannot access organization"

   • Verify the organization name is correct
   • Ensure you're a member of the organization
   • Check that your token has the necessary permissions

3. "Command timed out"

   • This can happen with very large repositories
   • The script will continue with the next repository
   • Consider using --repo-limit to test with fewer repositories first

4. Rate limiting errors

   • The script includes built-in delays
   • If you hit limits, wait and re-run the script
   • Consider reducing --limit to analyze fewer PRs per repository

📈 Use Cases

For Engineering Managers:

• Understand development workflow patterns across teams
• Identify repositories with different PR practices
• Track merge rates and development velocity
• Plan process improvements based on data

For DevOps/Platform Teams:

• Audit development practices across the organization
• Identify repositories that might benefit from different workflows
• Understand language and technology distribution
• Plan tooling and infrastructure improvements

For Engineering Teams:

• Compare your repository's practices with organization norms
• Understand what "typical" PR sizes look like
• Identify opportunities for process improvements
• Track progress on development workflow changes

🎯 Next Steps

1. Run your first analysis: Start with ./setup_analysis.sh
2. Export to CSV: Use the data in spreadsheets for custom analysis
3. Schedule regular analysis: Consider running monthly to track trends
4. Share insights: Use the summary reports to discuss development practices

---

The analysis respects GitHub API limits and handles private repositories securely.