0.4.0-incubating

Release Notes for release-0.4.0-incubating

Apache XTable™ (Incubating) is a cross-table converter for table formats that facilitates omni-directional interoperability across data processing systems and query engines. Currently, Apache XTable™ supports widely adopted open-source table formats such as Apache Hudi, Apache Iceberg, Delta Lake, and (as a source) Apache Paimon.

✨ What's Changed

🔄 Core Functionality Enhancements

• New formats:
  • Added a Delta Kernel conversion target (#801, #729).
  • Added Apache Paimon source support, with incremental sync and column stats (#742, #780, #767).
  • Added Apache Parquet source support, with partition extraction and incremental sync (#728, #669, #768, #805).
• Sync improvements:
  • Delta source: extract nested field statistics (#769).
  • Nested-column partition support in HudiPartitionValuesExtractor with hive-style partitioning (#738).
  • Fall back to parquet footers when metadata-table column stats are missing (#818).
  • Enable Delta checkpoint CRC exception throwing by default (#817).
  • Add additional properties when converting for Hudi (#714).
• Bug fixes:
  • Delta: null partition value for composite generated-column partitions (#828), map-key path handling (#826), NPE for binary in map/array schemas (#795), column-mapping-enabled source tables (#766).
  • Hudi: batch INSERT_OVERWRITE replacecommits dropping adds (#816), empty EarliestCommitToRetain (#732).
  • Iceberg: nested comments with qualified name (#797), schema sync (#749), empty source table (#756).
  • Parquet: snapshot sync on multiple commits with partitions (#806), schema-conversion bugs (#805).

📦 Packaging & Distribution

• Added xtable-spark-runtime, a thin drop-in Spark bundle for Spark 3.4 and 3.5 (#843).
• Made the bundled xtable-utilities jar runnable again (#841, #736).
• Fixed the scope of spark-catalyst in the utilities bundle (#790).

🧪 Testing & Quality

• Added an initial integration test for the XTable REST service (#716).
• Improved TestHudiInstantUtils with parameterization and more cases (#703).
• Removed wildcard imports and enforced with Spotless (#809).
• Removed redundant getSnapshotAt calls per commit (#791).

📘 Documentation & Community

• Added a 0.3.0 release page (#725) and updated the community sync page (#734).

🧰 Infrastructure & Maintenance

• Upgraded Apache Iceberg to 1.9.2 and Avro to 1.12.0 (#784).
• Added a Maven wrapper for consistent builds (#788).
• Automated license file generation and validation (#822); removed a duplicate javassist LICENSE entry (#745).
• Restored the delombok execution so release builds generate javadoc jars (#853).
• Added a Source Build Check workflow that builds the source release without any installed XTable artifacts (#868).
• Added .sdkmanrc to .gitignore (#793); bumped GitHub Actions (checkout, setup-node, setup-java).

Detailed Change Log

• [553] Parquet Schema and Column Stats Converters by @sapienza88 in #669
• Add additional properties when doing conversion for hudi by @rahil-c in #714
• Improved all tests in TestHudiInstantUtils by parameterising and adding more test cases by @Monilnarang in #703
• Fix typo in InternalTable by @vinishjail97 in #720
• [DOCS] Add a new 0.3.0 release page by @sagarlakshmipathy in #725
• Add intial IT for XTable REST Service by @rahil-c in #716
• Update community sync page by @pratyakshsharma in #734
• Bump actions/setup-java from 4 to 5 by @dependabot[bot] in #740
• [739] Nested column partition support in HudiPartitionValuesExtractor when hive style partitioning is enabled by @kroushan-nit in #738
• Bump actions/setup-node from 4 to 5 by @dependabot[bot] in #741
• Bump actions/checkout from 4 to 5 by @dependabot[bot] in #737
• [553] parquet source with partition extraction by @sapienza88 in #728
• [LICENSE] Remove duplicate entry for org.javassist:javassist:3.25.0-G… by @piyushdubey in #745
• [750] Fix schema sync for Iceberg tables by @the-other-tim-brown in #749
• Paimon Source Support by @mikedias in #742
• Handle empty Iceberg table source in conversion by @brishi19791 in #756
• Bump actions/checkout from 5 to 6 by @dependabot[bot] in #761
• Delta Kernel Draft PR by @vaibhavk1992 in #729
• Delta source: extract nested field's stats by @emilie-wang in #769
• Fix Delta to Iceberg not working on column mapping enabled Delta source table by @xr-chen in #766
• Implement Paimon Source Incremental Sync by @mikedias in #780
• [787] Add maven wrapper for consistent builds by @the-other-tim-brown in #788
• [755] Support column stats for paimon by @mao-liu in #767
• [789] Fix scope of spark catalyst in utilities bundle by @the-other-tim-brown in #790
• Upgrade Iceberg to 1.9.2 and Avro to 1.12.0 by @kevinjqliu in #784
• fix(delta): avoid NPE for binary in map/array schemas by @parisni in #795
• Remove redundant getSnapshotAt calls per commit by @brishi19791 in #791
• fix(iceberg): nested comments with qualified name by @parisni in #797
• add .sdkmanrc to .gitignore by @kevinjqliu in #793
• Remove wildcard imports and enforce with spotless by @the-other-tim-brown in #809
• Parquet source: Expand column stats support, fix bugs in schema conversion by @the-other-tim-brown in #805
• [bug] Parquet Source: snapshot sync fails on multiple commits with partitions by @the-other-tim-brown in #806
• Enable Delta checkpoint CRC exception throwing by default by @vinishjail97 in #817
• Bump actions/setup-node from 5 to 6 by @dependabot[bot] in #753
• Handle empty EarliestCommitToRetain by @alexr17 in #732
• [bug] Fix batch INSERT_OVERWRITE replacecommits dropping adds in HudiDataFileExtractor by @vinishjail97 in #816
• Fallback to parquet footers when metadata table col stats are missing by @vinishjail97 in #818
• Automate license file generation and validation by @the-other-tim-brown in #822
• Delta kernel conversion target by @vaibhavk1992 in #801
• Fix map key path handling in Delta schema extractors by @brishi19791 in #826
• Parquet Incremental Sync by @sapienza88 in #768
• [MINOR] Preserve original cause in ReadException on close failure by @alealandreev in #827
• fix(delta): null partition value for composite generated-column partitions by @alealandreev in #828
• [736] Make the bundled xtable-utilities jar runnable again (Hudi 0.x) + CI on branch-0.4 by @vinishjail97 in #841
• [836] xtable-spark-runtime: thin drop-in Spark bundle for 0.4.x (Spark 3.4 + 3.5) by @vinishjail97 in #843
• fix(build): restore delombok execution so release builds generate javadoc jars by @vinishjail97 in #853
• [864] Make the source release build with mvn package by @vinishjail97 in #868
• [872] Cover 2026 in the NOTICE copyright range and check it at release time by @vinishjail97 in #874

New Contributors

• @sapienza88 made their first contribution in #669
• @Monilnarang made their first contribution in #703
• @pratyakshsharma made their first contribution in #734
• @mikedias made their first contribution in #742
• @xr-chen made their first contribution in #766
• @mao-liu made their first contribution in #767
• @kevinjqliu made their first contribution in #784
• @parisni made their first contribution in #795
• @alexr17 made their first contribution in #732
• @alealandreev made their first contribution in #827

Full Changelog: 0.3.0-incubating...0.4.0-incubating