potpie Implement semantic caching for code parsing

Summary

Implemented content-based hashing to optimize LLM usage by reusing docstrings for identical code
Added semantic caching system that works across branches and repositories
Improved docstring generation efficiency with cache hit rate metrics

Implementation Details

Computes SHA-256 hashes based on node name + code content during parsing
Stores hashes in Neo4j graph with proper indices for fast retrieval
Creates content hash-based lookups before LLM inference
Reuses existing docstrings, tags, and embeddings for identical code
Preserves hashes when duplicating graphs between repositories
Adds detailed logging and metrics for cache hit rate performance

Performance Improvement

Achieves 65-70% cache hit rate when parsing similar branches
Significantly reduces LLM API calls and processing time
Uses existing Neo4j infrastructure without additional dependencies

Testing

Tested on multiple branches of the same repository with minor code changes. Cache hit rate increased with each additional branch parsed as expected.

🤖 Generated with Claude Code

Summary by CodeRabbit

Documentation
- Introduced a new set of development guidelines to standardize best practices across the project.
New Features
- Enhanced graph processing with semantic caching for improved performance and data consistency.
- Updated node duplication and inference logic to leverage unique content identifiers, reducing redundant computations and streamlining data handling.

Feb 26 '25 13:02 dhirenmathur

Walkthrough

This pull request introduces a new guidelines document, CLAUDE.md, that details development practices and conventions for the Momentum Server project. In addition, several modules now incorporate a content hashing mechanism for nodes. The CodeGraphService generates a SHA-256 hash for node data, and the ParsingService propagates this hash during graph duplication. Updates to the knowledge graph include new attributes in the schema and enhancements in the caching flow of the inference service, which now leverages the content hash to reduce redundant language model calls.

Changes

File(s)	Change Summary
`CLAUDE.md`	New guidelines document outlining setup, build commands, coding style, error handling, and dependency requirements for the project.
`app/modules/parsing/graph_construction/code_graph_service.py` `app/modules/parsing/graph_construction/parsing_service.py`	Introduced content hashing: added `generate_content_hash` static method and updated graph creation/storage and duplication queries to include a `content_hash` field.
`app/modules/parsing/knowledge_graph/inference_schema.py` `app/modules/parsing/knowledge_graph/inference_service.py`	Expanded the `DocstringRequest` schema with `content_hash` and `name` attributes and enhanced caching in inference functions by adding a new `find_cached_nodes` method and updating `generate_docstrings` and `batch_nodes` to manage semantic caching.

Sequence Diagram(s)

sequenceDiagram
    participant Client as API Client
    participant IS as InferenceService
    participant DB as Database/Cache
    participant LLM as Language Model

    Client->>IS: Request generate_docstrings(repo_id)
    IS->>DB: Call find_cached_nodes(content_hashes)
    alt Cache Hit
         DB-->>IS: Return cached node data
         IS->>Client: Return cached docstrings
    else Cache Miss
         IS->>DB: Query nodes missing content_hash
         Note right of IS: Compute SHA-256 for each node
         IS->>LLM: Request docstring generation
         LLM-->>IS: Return generated docstrings
         IS->>DB: Update cache with new content_hashes
         IS->>Client: Return new docstrings
    end

sequenceDiagram
    participant Node as Node Data
    participant CS as CodeGraphService
    participant DB as Graph Database

    Node->>CS: Provide node name and text
    CS->>CS: Normalize inputs & compute SHA-256 hash
    alt Text is empty
         CS-->>Node: Return None as content_hash
    else
         CS-->>Node: Return computed content_hash
    end
    CS->>DB: Store node with content_hash

Possibly related PRs

potpie-ai/potpie#210: Introduces similar modifications in the graph creation process, especially around using a content hash for improved node handling in the CodeGraphService.

Poem

I'm a bunny hopping through fields of code,
Where hashes hide like treasures in a burrow mode.
New guidelines light our path so clear,
Caching magic now brings our purpose near.
With each bound and byte, I celebrate our code's delight! 🐇🥕

✨ Finishing Touches

[ ] 📝 Generate Docstrings

Thank you for using CodeRabbit. We offer it for free to the OSS community and would appreciate your support in helping us grow. If you find it useful, would you consider giving us a shout-out on your favorite social media?

❤️ Share

🪧 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.
- Generate unit testing code for this file.
- 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 generate unit testing code for this file.
- @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 generate unit testing code.
- @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.

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 docstrings to generate docstrings for 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