Feat: neo4j community

[CaralHsi]

Description

Summary:
feat: support Neo4j Community Edition with Qdrant integration

• Introduced Neo4jCommunityGraphDB and Neo4jCommunityGraphDBConfig for Community Edition compatibility
  • Disabled enterprise-only features: multi-database, vector index, DB creation/drop
  • Enforced per-user isolation via mandatory user_name
• Integrated external vector DB (Qdrant) for embedding storage and retrieval
  • Implemented get_by_ids and ensure_payload_indexes in base VecDB interface
• Updated factories (GraphDBConfigFactory, GraphStoreFactory) to support neo4j-community backend
• Added tree_config_community.json as an example config using Neo4j Community + Qdrant
• Added quickstart script simple_openapi_memos_neo4j_community.py to demonstrate core functionality:
  • Memory add/get/search
  • OpenAI chat integration
  • Embedding filtering in display
• Refactored node parsing for subclass extensibility
• Cleaned up legacy code: removed unused comments and fixed edge deletion/property bugs

This change enables full MemOS support for open-source Neo4j deployments, expanding accessibility and ease of adoption.

Fix: #95

Reviewer: @(reviewer)

Checklist:

• I have performed a self-review of my own code | 我已自行检查了自己的代码
• I have commented my code in hard-to-understand areas | 我已在难以理解的地方对代码进行了注释
• I have added tests that prove my fix is effective or that my feature works | 我已添加测试以证明我的修复有效或功能正常
• I have added necessary documentation (if applicable) | 我已添加必要的文档（如果适用）
• I have linked the issue to this PR (if applicable) | 我已将 issue 链接到此 PR（如果适用）
• I have mentioned the person who will review this PR | 我已提及将审查此 PR 的人

[fridayL]

Code Review & Summary for PR #121 – Neo4j Community Edition Support

1. Neo4j Community Edition Support

• New Classes & Configs

  • Neo4jCommunityGraphDB and Neo4jCommunityGraphDBConfig are introduced.
  • These provide Neo4j Community Edition compatibility by disabling multi-database and vector index features (which are enterprise-only).

• User Isolation & Vector Database

  • Enforces per-user isolation via a mandatory user_name field.
  • Embeddings are handled by an external vector database (Qdrant), not Neo4j itself.

• Indexing

  • Only creates basic B-tree property indexes on Neo4j.
  • Vector payload indexes are managed in Qdrant.

• Unsupported Operations

  • Multi-DB, DB creation, and destructive DB drop operations are explicitly disallowed or raise clear errors.

2. Factory and Configuration Updates

• Factory Methods
  • Both GraphDBConfigFactory and GraphStoreFactory are updated to support the "neo4j-community" backend.
• Example Configuration
  • A sample config file (tree_config_community.json) is added showing how to use Neo4j Community with Qdrant for vectors.

3. Example Scripts

• Quickstart Example
  • simple_openapi_memos_neo4j_community.py: Shows how to set up MOS with OpenAI API, Neo4j Community, and Qdrant.
  • Demonstrates memory add/get/search, conversation, and filtering out embeddings from display results.

4. Vector DB Abstraction & Qdrant

• Interface Expansion
  • get_by_ids and ensure_payload_indexes are now abstract methods in the base vector DB interface.
  • Qdrant backend implements these, allowing multi-id fetches and idempotent payload index creation.

5. Code Quality and Minor Changes

• Refactoring
  • Node parsing logic is moved to instance methods for subclass extensibility.
  • Unused debug and Chinese comments are removed.
• Bug Fixes
  • Fixes around node parsing, edge deletion, and property handling.

---

Summary

This PR enables MemOS to use Neo4j Community Edition as a graph backend by:

• Separating enterprise-specific logic,
• Integrating Qdrant for vector search,
• Enforcing per-user logical isolation,
• Providing clear example usage and configuration.

Impact:
Greatly broadens deployability—users can now run MemOS with open-source Neo4j and Qdrant, no enterprise license needed.

Code Quality:
Code is modular and maintainable, with good extensibility for future data backends.

Overall:
A significant improvement for open-source and multi-tenant deployments. Well-structured, well-documented, and easy to adopt.