james-willis opened a new pull request, #3138: URL: https://github.com/apache/sedona/pull/3138
## Did you read the Contributor Guide? - Yes ## Is this PR related to a ticket? - Yes, closes #3137. ## What changes were proposed in this PR? Bump `jt-jiffle.version` 1.1.24 → 1.1.31 and `janino-version` 3.1.9 → 3.1.12 (the version jt-jiffle 1.1.31 is built against; janino is already shaded so Spark is unaffected). Add a `MapAlgebraTest` case for `atan2` (added to Jiffle in 1.1.28 — the test guards against regressing below that) and document the new loop-iteration cap in the map algebra docs. ## How was this patch tested? - `mvn -pl common test -Dtest=MapAlgebraTest` — 27/27 pass (JDK 17), including the new `atan2` test which fails on 1.1.24. - `mvn -pl common package` — verified the shaded jar still relocates antlr/janino/jiffle correctly, with `it.geosolutions.jaiext.jiffle.runtime.*` left unshaded per the existing shade rules (antlr stays 4.7.1 upstream, so relocations are unchanged). ## Why bump? 1. **Executor protection:** on 1.1.24 a Jiffle script with a runaway loop hangs the Spark executor forever — jt-jiffle 1.1.24 generates unguarded loops. 1.1.31 adds a per-pixel iteration guard (default 200), so such scripts throw instead. Configurable via `it.geosolutions.jaiext.jiffle.maxIterations` (negative disables); now documented in Raster-map-algebra.md. 2. **`atan2`** and other math functions added in 1.1.28. 3. NoData support in JiffleOp (1.1.29) and assorted fixes. ## Compatibility note User scripts with legitimately heavy per-pixel loops (>200 iterations) will now fail with a clear error unless the system property is raised — suggest a release-note entry. -- 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]
