SCIP Setup Scripts

This directory contains scripts for setting up and managing SCIP (Source Code Intelligence Protocol) integration.

How scip_pb2.py was Generated

The scip_pb2.py file is generated from the official SCIP Protocol Buffer definition. Here's how it was created:

1. Download the Protocol Definition

# Download the official scip.proto file from the SCIP repository
curl -o scip.proto https://raw.githubusercontent.com/sourcegraph/scip/main/scip.proto

2. Generate Python Bindings

# Install protobuf compiler (if not already installed)
# macOS:
brew install protobuf

# Ubuntu/Debian:
sudo apt-get install protobuf-compiler

# Generate Python bindings
protoc --python_out=. scip.proto

This creates scip_pb2.py with all the necessary Python classes to read/write SCIP protocol buffer files.

3. Current Version Info

The current scip_pb2.py was generated with:

• Protocol Buffer compiler: version varies
• Protobuf Python version: 5.29.3 (as seen in the file header)
• Source: https://github.com/sourcegraph/scip/blob/main/scip.proto

Using the Initialization Script

Quick Start

# Initialize SCIP for current directory
python scripts/initialize_scip.py

# Initialize SCIP for a specific project
python scripts/initialize_scip.py /path/to/project --project-name "my-project"

# Verbose output
python scripts/initialize_scip.py --verbose

What the Script Does

1. Checks Prerequisites: Verifies scip-python and protoc are installed
2. Downloads scip.proto: Gets the latest protocol definition
3. Generates Bindings: Creates fresh scip_pb2.py (if protoc available)
4. Tests Bindings: Verifies the protobuf bindings work
5. Creates Index: Generates the SCIP index for your project

Prerequisites

# Install scip-python (Node.js tool)
npm install -g @sourcegraph/scip-python

# Install protobuf compiler (optional, for regenerating bindings)
# macOS:
brew install protobuf

# Ubuntu/Debian:
sudo apt-get install protobuf-compiler

Manual SCIP Setup

If you prefer to set up SCIP manually:

1. Install scip-python

npm install -g @sourcegraph/scip-python

2. Generate SCIP Index

# In your project directory
scip-python index --project-name "your-project" --output index.scip

3. Use in Python

from blarify.code_references.scip_helper import ScipReferenceResolver

# Initialize resolver
resolver = ScipReferenceResolver("/path/to/project")

# Get references for a node
references = resolver.get_references_for_node(node)

Regenerating Protobuf Bindings

If you need to update the protobuf bindings (e.g., when SCIP protocol is updated):

# Download latest scip.proto
curl -o scip.proto https://raw.githubusercontent.com/sourcegraph/scip/main/scip.proto

# Regenerate Python bindings
protoc --python_out=. scip.proto

# Move to project root
mv scip_pb2.py ../

Troubleshooting

"scip-python not found"

# Install via npm
npm install -g @sourcegraph/scip-python

# Verify installation
scip-python --version

"protoc not found"

# macOS
brew install protobuf

# Ubuntu/Debian  
sudo apt-get install protobuf-compiler

# Verify installation
protoc --version

"Import scip_pb2 failed"

• Ensure scip_pb2.py is in your project root or Python path
• Try regenerating with the initialization script
• Check that protobuf Python package is installed: pip install protobuf

Version Mismatch Warnings

The warning about protobuf gencode version is expected when the runtime protobuf version is newer than the version used to generate scip_pb2.py. This doesn't affect functionality but can be resolved by regenerating the bindings.

Performance Benefits

Once properly set up, SCIP provides:

• 330x faster reference resolution compared to LSP
• Identical accuracy to LSP results
• Automatic index management (regenerates when files change)
• No runtime dependencies on language servers