# CGC Bundle Quick Reference
## 🚀 Quick Start
```bash
# Load a pre-indexed bundle
cgc load numpy.cgc
# Export your project
cgc export my-project.cgc --repo /path/to/project
# Import a bundle
cgc bundle import bundle.cgc --clear
```
## 📦 Bundle Commands
### Export
```bash
# Export specific repository
cgc bundle export OUTPUT.cgc --repo /path/to/repo
# Export all indexed repositories
cgc bundle export all-repos.cgc
# Export without statistics (faster)
cgc bundle export quick.cgc --repo /path/to/repo --no-stats
# Shortcut
cgc export OUTPUT.cgc --repo /path/to/repo
```
### Import
```bash
# Import bundle (add to existing graph)
cgc bundle import BUNDLE.cgc
# Import and clear existing data
cgc bundle import BUNDLE.cgc --clear
```
### Load
```bash
# Load bundle (convenience command)
cgc load BUNDLE.cgc
# Load and clear existing data
cgc load BUNDLE.cgc --clear
# Future: Download from registry
cgc load numpy # Will download numpy.cgc
```
## 🌐 Download Pre-indexed Bundles
```bash
# From GitHub Releases
wget https://github.com/Shashankss1205/CodeGraphContext/releases/latest/download/numpy-1.26.4.cgc
# Load it
cgc load numpy-1.26.4.cgc
# Start using
cgc find name linalg
```
## 🛠️ Create Your Own Bundle
### Method 1: From Existing Index
```bash
# Index your project
cgc index /path/to/project
# Export to bundle
cgc export my-project.cgc --repo /path/to/project
```
### Method 2: Using Helper Script
```bash
# Clone, index, and export in one go
./scripts/create-bundle.sh owner/repo [output-name]
# Example
./scripts/create-bundle.sh numpy/numpy
```
### Method 3: Manual Process
```bash
# Clone repository
git clone https://github.com/owner/repo
cd repo
# Index it
cgc index .
# Get version info
COMMIT=$(git rev-parse --short HEAD)
TAG=$(git describe --tags --abbrev=0 2>/dev/null || echo "main")
# Export with version
cgc export "repo-${TAG}-${COMMIT}.cgc" --repo .
```
## 📊 Inspect Bundles
```bash
# List contents
unzip -l bundle.cgc
# View metadata
unzip -p bundle.cgc metadata.json | jq
# View statistics
unzip -p bundle.cgc stats.json | jq
# Read README
unzip -p bundle.cgc README.md
# Extract bundle
unzip bundle.cgc -d extracted/
```
## 🔄 Common Workflows
### Workflow 1: Load Famous Library
```bash
# Download
wget https://github.com/.../numpy.cgc
# Load
cgc load numpy.cgc
# Query
cgc find name array
cgc analyze deps numpy.linalg
```
### Workflow 2: Share Team Project
```bash
# Developer A: Create bundle
cgc index /path/to/company-api
cgc export company-api.cgc --repo /path/to/company-api
# Share file (email, S3, GitHub, etc.)
# Developer B: Load bundle
cgc load company-api.cgc
# Start working
cgc find name authenticate
```
### Workflow 3: CI/CD Analysis
```bash
# Load pre-indexed dependencies
cgc load fastapi.cgc
cgc load sqlalchemy.cgc
# Index your code
cgc index ./my-api
# Analyze
cgc analyze deps my_api
cgc analyze complexity --threshold 10
```
### Workflow 4: Educational Exploration
```bash
# Load famous codebase
cgc load django.cgc
# Explore
cgc find name authenticate
cgc analyze callers authenticate
cgc analyze chain login authenticate --depth 10
```
## 🎯 Bundle Naming Convention
```
<repo-name>-<version>-<commit>.cgc
```
Examples:
- `numpy-1.26.4-a1b2c3d.cgc`
- `pandas-2.1.0-xyz789.cgc`
- `my-project-v1.0.0-abc123.cgc`
## 📈 Bundle Statistics
After export, check stats:
```bash
unzip -p bundle.cgc stats.json | jq
```
Output:
```json
{
"total_nodes": 15234,
"total_edges": 42156,
"files": 1342,
"nodes_by_type": {
"Function": 8211,
"Class": 942,
"File": 1342,
"Module": 156
},
"edges_by_type": {
"CALLS": 25432,
"INHERITS": 1234,
"IMPORTS": 5678
}
}
```
## 🔐 Security Best Practices
```bash
# Always verify bundle source
unzip -p bundle.cgc metadata.json | jq .repo
# Check commit hash matches official repo
unzip -p bundle.cgc metadata.json | jq .commit
# Use --clear cautiously (deletes existing data)
cgc load bundle.cgc --clear # Only if you're sure!
# Keep backups before importing
cgc export backup.cgc # Backup current state
cgc load new-bundle.cgc # Then load new bundle
```
## 🐛 Troubleshooting
### Bundle Import Fails
```bash
# Check bundle integrity
unzip -t bundle.cgc
# Verify format version
unzip -p bundle.cgc metadata.json | jq .cgc_version
# Try with --clear flag
cgc load bundle.cgc --clear
```
### Large Bundle Performance
```bash
# For very large bundles, be patient
# Import happens in batches of 1000 nodes
# Monitor progress in logs
cgc load large-bundle.cgc 2>&1 | tee import.log
```
### Version Mismatch
```bash
# Check your CGC version
cgc --version
# Check bundle version
unzip -p bundle.cgc metadata.json | jq .cgc_version
# Update if needed
pip install --upgrade codegraphcontext
```
## 📚 Available Pre-indexed Bundles
| Repository | Description | Typical Size |
|------------|-------------|--------------|
| **numpy** | Scientific computing | ~50MB |
| **pandas** | Data analysis | ~80MB |
| **fastapi** | Web framework | ~15MB |
| **requests** | HTTP library | ~10MB |
| **flask** | Web framework | ~12MB |
Download from: [GitHub Releases](https://github.com/Shashankss1205/CodeGraphContext/releases)
## 🔗 Related Commands
```bash
# Index management
cgc index . # Index current directory
cgc list # List indexed repos
cgc delete /path/to/repo # Delete indexed repo
cgc stats # Show statistics
# Database management
cgc config db neo4j # Switch to Neo4j
cgc config db falkordb # Switch to FalkorDB
cgc clean # Clean orphaned nodes
# Analysis
cgc find name <name> # Find by name
cgc analyze callers <fn> # Find callers
cgc analyze deps <mod> # Analyze dependencies
```
## 💡 Pro Tips
1. **Use bundles for onboarding** - New team members get instant context
2. **Create bundles before major refactors** - Easy rollback if needed
3. **Share bundles instead of indexing instructions** - Consistent results
4. **Use version tags in bundle names** - Track what version you're analyzing
5. **Combine multiple bundles** - Load numpy, pandas, scikit-learn together
## 📖 Documentation
- **Full Guide:** `docs/BUNDLES.md`
- **Architecture:** `docs/BUNDLE_ARCHITECTURE.md`
- **Implementation:** `docs/BUNDLE_IMPLEMENTATION.md`
- **CLI Reference:** `CLI_Commands.md`
---
**Questions?** Open an issue on [GitHub](https://github.com/Shashankss1205/CodeGraphContext/issues)
