imbajin commented on issue #364: URL: https://github.com/apache/hugegraph-ai/issues/364#issuecomment-4761998798
Thanks for the detailed proposal. The gradual, non-blocking adoption path sounds reasonable to me. A few thoughts: 1. I think we can avoid pinning `ty` to an exact version for now and use a lower bound instead, for example `ty>=0.0.51`, with a short inline comment explaining the rationale. Since `ty` has not reached GA yet, allowing newer versions helps us discover breaking changes or behavior changes early. Exact pinning may also be easy to forget later and could leave us stuck on an outdated pre-stable version. 2. I agree with starting from `hugegraph-python-client` and `hugegraph-llm` only. That keeps the initial scope focused and avoids too much noise from heavier optional dependencies. We can also add a TODO comment to cover `graph-mcp` and `text2gremlin` later once the first two modules are under control. 3. The incremental migration approach makes sense. We can first run `ty` in non-blocking mode, fix issues gradually, and eventually make it part of the required CI checks together with `ruff`. I would prefer adding it to the existing lint/type-check workflow rather than creating a separate CI workflow, so the developer feedback path stays simple and consolidated. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
