WindsorWZZ opened a new pull request, #140: URL: https://github.com/apache/spark-connect-rust/pull/140
### What changes were proposed in this pull request? Fill missing UserContext user IDs from ChannelBuilder on typed ExecutePlan, AnalyzePlan, reattachable execution, and artifact-upload requests. Preserve caller-provided IDs, user names, and extensions, and leave raw serialized requests unchanged. Draft proposal: the public Apache SPARK issue key is pending. ### Why are the changes needed? The configured user ID reaches gRPC metadata, but these typed request bodies omit it. This makes request identity inconsistent across RPCs and can cause servers to reject the requests. Artifact chunks also need the UserContext message to be present when no user ID is configured. ### Does this PR introduce _any_ user-facing change? Yes. The affected typed RPCs now send the configured identity when the caller has not supplied one. Requests with an explicit identity retain it. Raw passthrough requests remain byte-exact. ### How was this patch tested? - Five regression tests inspect protobuf requests received by an in-memory gRPC endpoint. They cover missing/empty/explicit contexts, preserved names and extensions, reattach/release context, all chunks of both upload paths, and raw requests containing unknown fields. - Before the fix, four of these tests fail; after the fix, all 100 core unit tests pass. - Checked-in-lockfile validation: 100 core unit tests and 392 client unit tests pass. All client integration targets compile with and without `wasm-udf`. The three existing core live-server tests remain ignored. - `cargo fmt --all --check` passes. The existing Rust CI workflow runs these tests through `cargo test -p apache-spark-connect -p apache-spark-connect-core --release`. ### Was this patch authored or co-authored using generative AI tooling? Generated-by: OpenAI Codex (GPT-6) This pull request and its description were written by Isaac. -- 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]
