Roadmap
Maturity matrix
| Package | Language | Status | Notes |
|---|---|---|---|
| pycubrid | Python | Stable | On PyPI; 16 releases; Python 3.10–3.14 × CUBRID 10.2–11.4 |
| sqlalchemy-cubrid | Python | Stable | On PyPI; 19 releases; official SQLAlchemy test suite integrated |
| cubrid-mcp-server | Python | Stable | On PyPI; read-only whitelist; domain knowledge packs |
| cubrid-cookbook-python | Python | Stable | 75 examples; nightly golden runs on live servers |
| cubrid-rs | Rust | In progress | Workspace: protocol, sync client, tokio, pool |
| sea-orm-cubrid | Rust | In progress | SeaORM dialect |
| cubrid-go | Go | In progress | database/sql driver |
| gorm-cubrid | Go | In progress | GORM dialect |
| cubrid-client | TypeScript | In progress | Node.js client |
| drizzle-cubrid | TypeScript | In progress | Drizzle ORM dialect |
Stable means: published to a package registry, documented, CI-gated on live databases, and covered by the release process (changelog, SBOM, semantic versioning). In-progress repositories follow the same playbook but have not reached that bar yet — they are not advertised as production-ready.
Direction
- Deepen Python — performance work on the driver hot path, broader SQLAlchemy/Alembic coverage, more cookbook golden tests.
- Replicate the playbook — the process validated on Python (protocol-level tests, live-DB CI matrix, translation governance, registry publishing) is being applied to Rust, Go, and TypeScript.
- AI-ready tooling — expand MCP server capabilities and domain knowledge packs as the ecosystem grows.
Performance focus
Current verified optimizations (documented, reproducible via cubrid-benchmark):
| Optimization | Result |
|---|---|
| Native ping (CHECK_CAS) | +280% throughput |
SQLAlchemy pool_pre_ping |
+588% throughput |
| Bulk insert (1,000 rows) | 12.3% faster |
| Query select-all | 19.9% faster |
Further driver-level optimization is tracked in each repository's issue tracker and milestones (e.g. pycubrid milestones).