3.0.0-palantir.40

Upgrade Hadoop to 2.9.2-palantir.10.

3.0.0-palantir.39

Re-releases 3.0.0-palantir.38 for publishing again. Functionally identical to 3.0.0-palantir.37 and 3.0.0-palantir.38.

3.0.0-palantir.38

Identical to 3.0.0-palantir.37, but re-publishing to get around that release's publish failure.

3.0.0-palantir.37

• Upgrade Kubernetes Client to 4.4.2 (#603)

3.0.0-palantir.36

• kubernetes local should work with in process launcher (#601)

3.0.0-palantir.35

• Allow K8s to deploy in local mode (#600)
• [SPARK-28776] SparkML Writer gets hadoop conf from session state (#599)

3.0.0-palantir.34

• log conda output when verbosity > 0 (#588)

3.0.0-palantir.33

• revert API break in StageInfo from SPARK-26090 (#587)
• bump hadoop to 2.9.2-palantir.9 (#586)

3.0.0-palantir.32

• Upgrade Hadoop to 2.9.2-palantir.8 (#585)

3.0.0-palantir.31

Improvements

• [SPARK-27744][SQL] Preserve spark properties on async subquery tasks (#579)
• Remove inaccessible fields in SizeEstimator (#580)
• Add support for StructType return in Scalar Pandas UDF (#582) ([SPARK-26887][SQL][PYTHON][NS], [SPARK-23836][PYTHON], [SPARK-27163][PYTHON], [SPARK-27240][PYTHON])
• [SPARK-26762][SQL][R] Arrow optimization for conversion from Spark DataFrame to R DataFrame (#578)
• SafeLogging in MapOutputTracker (#581)