Skip to content

Add Codegen embedding provider with OpenAI and DeepSeek support #120

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 116 commits into
base: develop
Choose a base branch
from

Conversation

codegen-sh[bot]
Copy link

@codegen-sh codegen-sh bot commented May 17, 2025

This PR adds a flexible embedding provider system that can be used with Codegen's FileIndex for semantic code search.

Features

  • Seamless integration with Codegen's FileIndex class
  • Support for multiple embedding providers (currently OpenAI and DeepSeek)
  • Easy to extend with additional providers
  • Simple patching mechanism that doesn't require modifying Codegen's source code

Files Added

  • codegen_embedding_provider.py: The main implementation with provider classes and patching mechanism
  • example_codegen_usage.py: Example script showing how to use the tool with Codegen
  • README_CODEGEN.md: Documentation on usage and extension

How It Works

The tool uses a simple patching mechanism to replace Codegen's _get_embeddings method in the FileIndex class with our own implementation that routes requests through the specified provider. This approach allows you to switch providers without modifying Codegen's source code.

Usage Example

from codegen.extensions.index.file_index import FileIndex
from codegen.sdk.core.codebase import Codebase
from codegen_embedding_provider import CodegenEmbeddingManager, patch_codegen_file_index

# Initialize your codebase
codebase = Codebase("/path/to/repo")

# Create a FileIndex instance
file_index = FileIndex(codebase)

# Initialize the embedding manager with your preferred provider
embedding_manager = CodegenEmbeddingManager(
    provider="deepseek",  # or "openai"
    api_key="your_api_key_here"
)

# Patch the FileIndex to use your provider
restore_fn = patch_codegen_file_index(file_index, embedding_manager)

# Now use the FileIndex as normal
file_index.create()  # This will use your specified provider for embeddings
results = file_index.similarity_search("find user authentication code", k=5)

# If needed, restore the original embedding method
restore_fn()

Notes

  • DeepSeek provider currently falls back to OpenAI since DeepSeek doesn't offer a direct embeddings API
  • The patch is temporary and only affects the specific FileIndex instance you patch
  • You'll need to re-apply the patch if you create a new FileIndex instance

💻 View my workAbout Codegen

Summary by Sourcery

Add a flexible embedding provider system to plug alternative backends into Codegen’s FileIndex for semantic code search, including OpenAI and DeepSeek support, a non-invasive patching mechanism, example usage script, and accompanying documentation.

New Features:

  • Introduce CodegenEmbeddingManager with support for OpenAI and DeepSeek embedding providers
  • Add patch_codegen_file_index function to non-invasively override FileIndex’s embedding method
  • Include example script for integrating custom embedding providers with Codegen

Enhancements:

  • Design the provider framework for easy extension to additional backends without modifying Codegen’s source
  • Implement DeepSeek provider with automatic fallback to OpenAI embeddings

Documentation:

  • Add README_CODEGEN.md detailing installation, usage, and extension of the embedding provider system

clee-codegen and others added 30 commits February 26, 2025 23:54
# Motivation

The **Codegen on OSS** package provides a pipeline that:

- **Collects repository URLs** from different sources (e.g., CSV files
or GitHub searches).
- **Parses repositories** using the codegen tool.
- **Profiles performance** and logs metrics for each parsing run.
- **Logs errors** to help pinpoint parsing failures or performance
bottlenecks.

<!-- Why is this change necessary? -->

# Content

<!-- Please include a summary of the change -->
see
[codegen-on-oss/README.md](https://github.com/codegen-sh/codegen-sdk/blob/acfe3dc07b65670af33b977fa1e7bc8627fd714e/codegen-on-oss/README.md)

# Testing

<!-- How was the change tested? -->
`uv run modal run modal_run.py`
No unit tests yet 😿 

# Please check the following before marking your PR as ready for review

- [ ] I have added tests for my changes
- [x] I have updated the documentation or added new documentation as
needed
Original commit by Tawsif Kamal: Revert "Revert "Adding Schema for Tool Outputs"" (codegen-sh#894)

Reverts codegen-sh#892

---------

Co-authored-by: Rushil Patel <rpatel@codegen.com>
Co-authored-by: rushilpatel0 <171610820+rushilpatel0@users.noreply.github.com>
Original commit by Ellen Agarwal: fix: Workaround for relace not adding newlines (codegen-sh#907)
Copy link

sourcery-ai bot commented May 17, 2025

Reviewer's Guide

This PR adds a pluggable embedding provider system to Codegen’s FileIndex by defining a provider abstraction (OpenAI and DeepSeek), a manager to select providers, and a patching mechanism to override the built-in embedding method, along with documentation and an example script.

Class Diagram: New Embedding Provider System

classDiagram
    class EmbeddingProvider {
        +api_key: str
        +__init__(api_key: str)
        +get_embeddings(texts: List[str], model: str) List~List~float~~*
    }
    note for EmbeddingProvider "Base class for embedding providers"

    class OpenAIEmbeddingProvider {
        +base_url: str
        +__init__(api_key: str, base_url: str)
        +get_embeddings(texts: List[str], model: str) List~List~float~~
    }

    class DeepSeekEmbeddingProvider {
        +base_url: str
        -openai_fallback: OpenAIEmbeddingProvider
        +__init__(api_key: str, base_url: str)
        +get_embeddings(texts: List[str], model: str) List~List~float~~
    }
    note for DeepSeekEmbeddingProvider "Uses OpenAIEmbeddingProvider as fallback"

    class CodegenEmbeddingManager {
        -provider: EmbeddingProvider
        +PROVIDERS: dict$
        +__init__(provider: str, api_key: str, base_url: Optional[str])
        +get_embeddings(texts: List[str], model: Optional[str]) List~List~float~~
    }
    note for CodegenEmbeddingManager "Manages and uses an EmbeddingProvider instance"

    EmbeddingProvider <|-- OpenAIEmbeddingProvider : Inheritance
    EmbeddingProvider <|-- DeepSeekEmbeddingProvider : Inheritance
    DeepSeekEmbeddingProvider *-- "1" OpenAIEmbeddingProvider : openai_fallback
    CodegenEmbeddingManager *-- "1" EmbeddingProvider : provider
Loading

File-Level Changes

Change Details Files
Introduce modular embedding provider framework
  • Define EmbeddingProvider base class and OpenAI/DeepSeek implementations
  • Implement DeepSeek fallback to OpenAI for embeddings
  • Add CodegenEmbeddingManager to select and delegate to providers
codegen_embedding_provider.py
Add patching mechanism for Codegen FileIndex
  • Override _get_embeddings with provider-backed method
  • Return restoration function to revert the patch
codegen_embedding_provider.py
Provide documentation for embedding provider
  • Document installation, usage, and extension in README
  • Explain patching approach and DeepSeek fallback
README_CODEGEN.md
Add example usage script
  • CLI to index and search code repositories
  • Demonstrate initializing and patching FileIndex with providers
example_codegen_usage.py

Tips and commands

Interacting with Sourcery

  • Trigger a new review: Comment @sourcery-ai review on the pull request.
  • Continue discussions: Reply directly to Sourcery's review comments.
  • Generate a GitHub issue from a review comment: Ask Sourcery to create an
    issue from a review comment by replying to it. You can also reply to a
    review comment with @sourcery-ai issue to create an issue from it.
  • Generate a pull request title: Write @sourcery-ai anywhere in the pull
    request title to generate a title at any time. You can also comment
    @sourcery-ai title on the pull request to (re-)generate the title at any time.
  • Generate a pull request summary: Write @sourcery-ai summary anywhere in
    the pull request body to generate a PR summary at any time exactly where you
    want it. You can also comment @sourcery-ai summary on the pull request to
    (re-)generate the summary at any time.
  • Generate reviewer's guide: Comment @sourcery-ai guide on the pull
    request to (re-)generate the reviewer's guide at any time.
  • Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
    pull request to resolve all Sourcery comments. Useful if you've already
    addressed all the comments and don't want to see them anymore.
  • Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
    request to dismiss all existing Sourcery reviews. Especially useful if you
    want to start fresh with a new review - don't forget to comment
    @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

  • Enable or disable review features such as the Sourcery-generated pull request
    summary, the reviewer's guide, and others.
  • Change the review language.
  • Add, remove or edit custom review instructions.
  • Adjust other review settings.

Getting Help

Copy link

korbit-ai bot commented May 17, 2025

By default, I don't review pull requests opened by bots. If you would like me to review this pull request anyway, you can request a review via the /korbit-review command in a comment.

Copy link

coderabbitai bot commented May 17, 2025

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.


🪧 Tips

Chat

There are 3 ways to chat with CodeRabbit:

  • Review comments: Directly reply to a review comment made by CodeRabbit. Example:
    • I pushed a fix in commit <commit_id>, please review it.
    • Explain this complex logic.
    • Open a follow-up GitHub issue for this discussion.
  • Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query. Examples:
    • @coderabbitai explain this code block.
    • @coderabbitai modularize this function.
  • PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
    • @coderabbitai gather interesting stats about this repository and render them as a table. Additionally, render a pie chart showing the language distribution in the codebase.
    • @coderabbitai read src/utils.ts and explain its main purpose.
    • @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.
    • @coderabbitai help me debug CodeRabbit configuration file.

Support

Need help? Join our Discord community for assistance with any issues or questions.

Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments.

CodeRabbit Commands (Invoked using PR comments)

  • @coderabbitai pause to pause the reviews on a PR.
  • @coderabbitai resume to resume the paused reviews.
  • @coderabbitai review to trigger an incremental review. This is useful when automatic reviews are disabled for the repository.
  • @coderabbitai full review to do a full review from scratch and review all the files again.
  • @coderabbitai summary to regenerate the summary of the PR.
  • @coderabbitai generate sequence diagram to generate a sequence diagram of the changes in this PR.
  • @coderabbitai resolve resolve all the CodeRabbit review comments.
  • @coderabbitai configuration to show the current CodeRabbit configuration for the repository.
  • @coderabbitai help to get help.

Other keywords and placeholders

  • Add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.
  • Add @coderabbitai summary to generate the high-level summary at a specific location in the PR description.
  • Add @coderabbitai anywhere in the PR title to generate the title automatically.

CodeRabbit Configuration File (.coderabbit.yaml)

  • You can programmatically configure CodeRabbit by adding a .coderabbit.yaml file to the root of your repository.
  • Please see the configuration documentation for more information.
  • If your editor has YAML language server enabled, you can add the path at the top of this file to enable auto-completion and validation: # yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json

Documentation and Community

  • Visit our Documentation for detailed information on how to use CodeRabbit.
  • Join our Discord Community to get help, request features, and share feedback.
  • Follow us on X/Twitter for updates and announcements.

Copy link
Author

codegen-sh bot commented May 17, 2025

I see a check failed - I'm on it! 🫡

💻 View my work

Copy link
Author

codegen-sh bot commented May 17, 2025

✅ Fixed the failing checks in PR #120 by addressing linting and type checking issues in the embedding provider files. The main changes include:

  1. Fixed type annotations to use List instead of list for better compatibility
  2. Added proper error handling and message formatting
  3. Removed unused imports
  4. Fixed whitespace and formatting issues

The changes have been pushed to the original PR branch. You can view the fix in this commit.

💻 View my work • React 👍 or 👎

Copy link
Author

codegen-sh bot commented May 17, 2025

I see a check failed - I'm on it! 🫡

💻 View my work

Copy link
Author

codegen-sh bot commented May 17, 2025

✅ Fixed the failing checks in PR #120 by addressing formatting issues in the README_CODEGEN.md file. The main changes include:

  1. Fixed the numbered list formatting in the README to use proper sequential numbers (1, 2, 3) instead of (1, 1, 3)
  2. Ensured consistent formatting throughout the document

The changes have been pushed to the original PR branch. You can view the fix in this commit.

💻 View my work • React 👍 or 👎

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants