Fokko commented on PR #18899: URL: https://github.com/apache/iceberg/pull/18899#issuecomment-6846331881
Verified locally by building the image from this branch and running PyIceberg's full integration suite against it: - `./gradlew :iceberg-open-api:stageRESTFixtureRuntime` on this branch stages **47 JARs (197 MB)**, matching the PR description: single shadow bundle JARs (`iceberg-aws/azure/gcp-bundle`), no JUnit, no `iceberg-bigquery`, no unshaded SDK transitives - `docker build -t apache/iceberg-rest-fixture:latest -f docker/iceberg-rest-fixture/Dockerfile .` from the branch root - From `apache/iceberg-python` (main): `docker compose -f dev/docker-compose-integration.yml up -d --build --wait` — the `rest` service ran the locally built image (confirmed via image ID) with `S3FileIO` against RustFS; all services healthy and `GET /v1/config` responds - `python dev/provision.py` (Spark creates databases/tables through the REST catalog, landing in `s3://warehouse/...`) succeeded - Full suite `pytest tests/ -m integration`: **1362 passed, 71 skipped** in ~6 min - Server logs across the whole run (~117k lines): no `NoClassDefFoundError` / `ClassNotFoundException` / `LinkageError`; the only ERRORs are expected 404s from negative test cases; S3 metadata commits are visible in the logs So the pruned 47-JAR image runs the REST server (JdbcCatalog/SQLite + S3FileIO) end-to-end under PyIceberg's integration tests with no class-loading errors. -- 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]
