commits  

Messages by Thread

• (terraform-provider-paimon) branch dependabot/github_actions/github/codeql-action/init-4.38.0 created (now de36006) github-bot
• (paimon-rust) branch main updated: feat(python): deserialize cross-language splits (#889) lzljs3620320
• (paimon-mosaic) branch main updated: perf(core): speed up schema BPE construction (#89) lzljs3620320
• (paimon-rust) branch main updated: perf: optimize scalar index reads and support BTree V2 (#888) lzljs3620320
• (terraform-provider-paimon) branch dependabot/github_actions/github/codeql-action/init-4.37.9 deleted (was 174fe4b) github-bot
• (paimon) branch master updated: [python] Expand native read coverage for nested and streaming reads (#10009) lzljs3620320
• (terraform-provider-paimon) branch dependabot/github_actions/github/codeql-action/autobuild-4.37.9 deleted (was 0cc65f4) github-bot
• (terraform-provider-paimon) branch dependabot/github_actions/github/codeql-action/analyze-4.38.0 created (now 7f47ed5) github-bot
• (terraform-provider-paimon) branch dependabot/github_actions/github/codeql-action/analyze-4.37.9 deleted (was 9fe1b21) github-bot
• svn commit: r87705 - in dev/paimon: paimon-2.1.0-rc3 pypaimon-2.1.0-rc3 hope
• (paimon-rust) branch main updated: fix: make Python record batch reader cancellable (#885) lzljs3620320
• (paimon) branch master updated: [python] Improve parallel native read scheduling (#10003) lzljs3620320
• (paimon) branch master updated: [python] Close full-text readers when search raises synchronously (#9997) lzljs3620320
• (paimon) branch master updated: [python] Reduce memory usage for overlapping temporal windows (#10000) lzljs3620320
• (paimon) branch master updated: [core] Preserve reader errors for compacted changelog files (#9995) lzljs3620320
• (paimon) branch master updated: [python] Verify candidate-only scalar filters before vector top-k (#9999) lzljs3620320
• svn commit: r87689 - dev/paimon/paimon-rust-0.4.0-rc1 hope
• (paimon-rust) branch main updated: fix: Map precision-zero timestamps to Arrow seconds (#884) lzljs3620320
• (paimon) branch master updated: [core] Avoid redundant row-range sorting for global-index results (#9957) lzljs3620320
• (paimon-cpp) branch gh-pages updated (044d5703 -> 34185600) github-bot
• (paimon-rust) annotated tag v0.4.0-rc1 updated (0139f7a0 -> 606fac36) hope
• (paimon) annotated tag release-2.1.0-rc3 updated (3d8243d610 -> eb5a764337) hope
• (paimon-rust) branch main updated: [rust] Bound sparse BlobDescriptor range reads (#883) lzljs3620320
• (paimon) branch master updated: [flink] Preserve query authorization when narrowing splits in the Flink source (#9949) lzljs3620320
• (paimon) branch master updated: [python] Distribute batch vector refinement across Ray workers (#9988) lzljs3620320
• (paimon-vector-index) branch main updated: docs: publish vector index 0.5.0 release notes (#109) jerryjing
• (paimon) branch master updated: Fix: partition index bucket overflow (#9977) lzljs3620320
• (paimon) branch master updated: [core] Preserve merge correctness when pushing filters into chain scans (#9979) lzljs3620320
• (paimon) branch master updated: [spark][core] Preserve SQL semantics for CAST predicate pushdown (#9980) lzljs3620320
• (paimon) branch master updated: [python] Stream vector index candidates during global merging (#9989) lzljs3620320
• (paimon) branch master updated: [common] Reject a serialized row shorter than its fixed-length part (#9991) lzljs3620320
• (paimon) branch master updated: [python] Enable Parquet page indexes when supported (#9976) lzljs3620320
• (paimon) branch master updated: [flink] Support negated predicates in PredicateConverter (#9427) lzljs3620320
• (paimon-rust) branch main updated: Upgrade paimon-vindex-core to 0.5.0 (#879) lzljs3620320
• (paimon-rust) branch main updated: [rust] Coalesce inline BLOB reads with bounded amplification (#880) lzljs3620320
• (paimon) branch master updated: [common] Release the BTree index file on every path that abandons the writer (#9474) lzljs3620320
• (paimon) branch master updated: [common] Keep the selected variant schema when its evidence is gone (#9532) lzljs3620320
• (paimon) branch release-2.1 updated (b6d8a70499 -> eaf7c8744b) hope
  • (paimon) 19/20: [core] Reuse sidecar block cache for unfiltered scans (#9924) hope
  • (paimon) 11/20: [core] Fall back to raw vector search when scalar index cannot evaluate filter (#9909) hope
  • (paimon) 01/20: [core] Skip BTree index reads for all-matching predicates (#9834) hope
  • (paimon) 07/20: [python] Clean up normal files when composite writer metadata creation fails (#9917) hope
  • (paimon) 14/20: [common] Reject out-of-range integral casts when extracting a variant scalar (#9809) hope
  • (paimon) 03/20: [format] Drop filters on unprojected columns before Parquet conversion (#9858) hope
  • (paimon) 16/20: [python] Build global index shards concurrently (#9798) hope
  • (paimon) 02/20: [python] Add manifest sidecar codec and block-pruned reads (#9908) hope
  • (paimon) 12/20: [common] Keep blank variant object keys unshredded during schema inference (#9813) hope
  • (paimon) 15/20: [flink] Scope scan.parallelism to source operators (#9803) hope
  • (paimon) 18/20: [core] Preserve query authorization when wrapping data evolution index splits (#9925) hope
  • (paimon) 04/20: [python] Keep local search and lookup on one snapshot (#9843) hope
  • (paimon) 10/20: [python] Support allow_exact_matches in temporal as-of joins (#9915) hope
  • (paimon) 06/20: [python] Preserve manifest sidecar references on commit retries (#9921) hope
  • (paimon) 08/20: [format] Support file.block-size for Avro data file (#9910) hope
  • (paimon) 13/20: [spark] Support remove_orphan_blobs procedure. (#9812) hope
  • (paimon) 05/20: [core][flink][spark] Add API for top partitions (#9914) hope
  • (paimon) 20/20: [core][python] Add optional Bloom filters to BTree indexes (#9946) hope
  • (paimon) 09/20: [core] Add option to disable overflow exception in sum/product agg (#9874) hope
  • (paimon) 17/20: [api] Add an opt-in DLFOpenApiV4Signer for DlfNext endpoints (#9918) hope
• (paimon) branch release-2.1 updated (6bfd8418dc -> 8ce380d29b) hope
  • (paimon) 01/13: [e2e] Strip any trailing Spark error log from spark-sql output before comparison (#9971) hope
  • (paimon) 05/13: [python] Support batch vector search with Ray (#9950) hope
  • (paimon) 11/13: [python] Fix MAP and MULTISET type JSON serialization (#9682) hope
  • (paimon) 06/13: [python] Consolidate composite writer lifecycle and file output (#9948) hope
  • (paimon) 10/13: [python] Avoid redundant blob stream opens on index cache hits (#9974) hope
  • (paimon) 08/13: [core] Add adaptive BTree posting list encoding (#9981) hope
  • (paimon) 13/13: [release] Align Spark 4.0 UT parent version with 2.1 hope
  • (paimon) 04/13: [flink] Fix GlobalFullCompactionSinkWrite re-reading snapshots on every checkpoint after a full compaction (#9959) hope
  • (paimon) 09/13: [python] Fix flaky Daft-on-Ray CI failures (#9975) hope
  • (paimon) 02/13: [eslib] Do not require build-time es-index options when reading persisted indexes (#9969) hope
  • (paimon) 07/13: [python] Reuse Parquet fragment metadata for single-file reads (#9881) hope
  • (paimon) 03/13: [format] Validate the avro block size before creating the writer (#9961) hope
  • (paimon) 12/13: [python] Reduce LeRobot DataLoader worker startup time (#9888) hope
• (paimon) branch release-2.1 updated (eaf7c8744b -> 6bfd8418dc) hope
  • (paimon) 20/20: [core] Optimize tag reads during snapshot expiration (#9967) hope
  • (paimon) 19/20: [python] Add opt-in Rust native data reads (#9954) hope
  • (paimon) 16/20: [core] Request only top-k candidates from data-evolution full-text indexes (#9955) hope
  • (paimon) 11/20: [core] Apply deletion vectors during data evolution file index pushdown (#9922) hope
  • (paimon) 05/20: [common] Infer a variant shredding schema regardless of the writer's key order (#9940) hope
  • (paimon) 09/20: [core] Preserve query authorization when narrowing chain table splits (#9929) hope
  • (paimon) 17/20: [core] Fix OOM error handling in MergeTreeWriter and improve resource cleanup (#9933) hope
  • (paimon) 03/20: [flink] Fix rollback_to procedure throwing NullPointerException when neither snapshot_id nor tag is set (#9944) hope
  • (paimon) 18/20: [python] Distribute vector search refinement with Ray (#9951) hope
  • (paimon) 07/20: [python] Share final lookups across batch vector queries (#9935) hope
  • (paimon) 02/20: [core] Use a single BTree scan for bounded range predicates (#9844) hope
  • (paimon) 13/20: [api] Sign x-acs-action in DLFOpenApiV4Signer (#9937) hope
  • (paimon) 10/20: [python] Support distributed vector search with Ray (#9930) hope
  • (paimon) 14/20: [format] Simplify buffered Avro encoder selection hope
  • (paimon) 08/20: [python] Keep batch vector scoring vectorized for null rows (#9934) hope
  • (paimon) 06/20: [common] Convert variant timestamps with the requested zone instead of the JVM default (#9939) hope
  • (paimon) 04/20: [core] Reject a non-parquet per-level or changelog format for variant columns at DDL time (#9942) hope
  • (paimon) 01/20: [core][spark][docs] Support row filters in full-text search (#9855) hope
  • (paimon) 15/20: [core] Separate commit history bounds from strict mode (#9960) hope
  • (paimon) 12/20: [format] Use buffered encoder for Avro writes (#9928) hope
• (paimon) branch release-2.1 updated (1fc0f17ad2 -> b6d8a70499) hope
  • (paimon) 03/20: [core] Fix FileStore tag manager ignoring the table branch (#9868) hope
  • (paimon) 16/20: [python] Batch TorchCodec frame decoding for LeRobot datasets (#9836) hope
  • (paimon) 10/20: [format] Reject multi-character csv delimiter, quote and escape (#9904) hope
  • (paimon) 04/20: [common] Stop answering value predicates from a truncated bitmap index (#9866) hope
  • (paimon) 13/20: [common] Read VECTOR fields from NestedRow (#9864) hope
  • (paimon) 11/20: [test] Wait for a commit past the aborted savepoint after region failover (#9907) hope
  • (paimon) 17/20: [core] Preserve quoted empty and null keys/values in string-to-MAP cast (#9849) hope
  • (paimon) 19/20: [codegen] Fix equaliser treating null array elements and map values as unequal (#9860) hope
  • (paimon) 20/20: [core] Fix the failing DataEvolutionUtilsTest (#9911) hope
  • (paimon) 06/20: [cdc] Support per-table configuration for MySQL database CDC (#9871) hope
  • (paimon) 18/20: [core] Add compression extension to text format data file names (#9853) hope
  • (paimon) 01/20: [core] Introduce a vector read planner for data evolution (#9877) hope
  • (paimon) 12/20: [core] Use full-range anchor as row representative in DE read-type pruning (#9899) hope
  • (paimon) 15/20: [common] Fix snapshot and tag reads failing when the file is deleted concurrently (#9851) hope
  • (paimon) 02/20: [common] Delegate listFilesIterative through the remaining FileIO wrappers (#9865) hope
  • (paimon) 09/20: [iceberg] Retry transient REST DNS failures (#9903) hope
  • (paimon) 07/20: [flink][cdc][iceberg] Include table identifier in schema-change and CDC-commit exception messages (#9892) hope
  • (paimon) 05/20: [core] Fix remaining byte[] identity comparisons in collect and merge_map aggregators (#9870) hope
  • (paimon) 08/20: [format] Close the codec probe stream in HadoopCompressionUtils (#9902) hope
  • (paimon) 14/20: [python] Search vector index shards concurrently (#9805) hope
• (paimon) branch release-2.1 updated (a5f3dd4b99 -> 1fc0f17ad2) hope
  • (paimon) 11/20: [python] Reduce memory usage for Parquet row-id updates (#9840) hope
  • (paimon) 05/20: [python] Speed up LeRobot video window reads (#9852) hope
  • (paimon) 12/20: [spark][core] Support more expressions in V2 filter pushdown (#9891) hope
  • (paimon) 20/20: [python] Extend ContiguousWindowDataset for frame-level training (#9875) hope
  • (paimon) 10/20: [core] Generate sidecars for newly written manifests (#9889) hope
  • (paimon) 04/20: [core] Execute complete filters with unprojected fields (#9872) hope
  • (paimon) 02/20: [python] Speed up nested VARIANT reads by vectorizing offsets (#9847) hope
  • (paimon) 15/20: [python] Resolve value stats columns using file schema (#9906) hope
  • (paimon) 01/20: [python] Cover native dynamic PK events and compacted first-row scans (#9854) hope
  • (paimon) 03/20: [python] Expand native planning across resolved tables and scan modes (#9873) hope
  • (paimon) 09/20: [python] Use conditional OSS puts for atomic metadata writes (#9715) hope
  • (paimon) 06/20: [python] Support writing shared-shredding MAPs in Parquet (#9837) hope
  • (paimon) 08/20: [core] Manage manifest extra file lifecycle (#9886) hope
  • (paimon) 19/20: [python] Honor snapshot.ignore-empty-commit for empty appends (#9878) hope
  • (paimon) 13/20: [core] Prune manifest blocks with sidecars during scans (#9897) hope
  • (paimon) 16/20: [python] Make BLOB I/O merge parameters configurable (#9890) hope
  • (paimon) 17/20: [format] Support file.block-size for Avro (#9901) hope
  • (paimon) 18/20: Revert "[format] Support file.block-size for Avro (#9901)" hope
  • (paimon) 14/20: [spark][core] Support more scalar transforms in V2 filter pushdown (#9905) hope
• (paimon) branch release-2.1 updated (eb43b419ce -> a5f3dd4b99) hope
  • (paimon) 10/20: [python] Group LeRobot video frame decoding by payload (#9799) hope
  • (paimon) 05/20: [common] Fix inferred variant shredding failing on decimals with trailing zeros (#9811) hope
  • (paimon) 04/20: [vortex] Do not push nanosecond timestamp literals that overflow int64 (#9810) hope
  • (paimon) 15/20: [python] Align batch scan semantics with Java and extend native planning (#9839) hope
  • (paimon) 02/20: [python] Reuse gathered window values across aggregations (#9804) hope
  • (paimon) 20/20: [rest] Add semantic view definition management to REST Catalog (#9848) hope
  • (paimon) 16/20: [rest] Add generic entity labels to REST Catalog (#9838) hope
  • (paimon) 14/20: [core] Expose Avro header and physical block locations (#9835) hope
  • (paimon) 09/20: [format] Validate ORC stripe metadata before reading (#9828) hope
  • (paimon) 01/20: [python] Preserve declared primary-key order in merge reads (#9801) hope
  • (paimon) 13/20: [python] Add block-based manifest Avro reader (#9826) hope
  • (paimon) 03/20: [python] Stream batch vector refinement candidates (#9806) hope
  • (paimon) 06/20: [filesystem][s3] Serve s3a:// paths with the S3 plugin (#9814) hope
  • (paimon) 19/20: [python] Expand native planning and align incremental reads with Java (#9846) hope
  • (paimon) 08/20: [python] Skip empty batches when advancing merge readers (#9820) hope
  • (paimon) 11/20: [python] Support selected-key projection for shared-shredding MAP (#9789) hope
  • (paimon) 07/20: [common] Read a shredded variant decimal as the same string as an unshredded one (#9818) hope
  • (paimon) 18/20: [core] Support input streams in Avro block readers (#9842) hope
  • (paimon) 17/20: [python] Decode LeRobot cameras in parallel (#9832) hope
• (paimon) branch release-2.1 updated (2a5cbe906c -> eb43b419ce) hope
  • (paimon) 04/20: [common] Reject out-of-range range-bitmap chunk-size option (#9765) hope
  • (paimon) 14/20: [python] Align manifest metadata and prune manifest files by bucket (#9807) hope
  • (paimon) 13/20: [spark] Compile the shared Spark test sources against 4.0.3 for paimon-spark-4.0 (#9794) hope
  • (paimon) 18/20: [python] Run native planning CI against paimon-rust main (#9817) hope
  • (paimon) 07/20: [common] Support VECTOR elements in InternalArray accessors (#9773) hope
  • (paimon) 20/20: [python] Execute hybrid search routes concurrently (#9797) hope
  • (paimon) 03/20: [core] Allow custom primary-key compaction rewriters (#9762) hope
  • (paimon) 08/20: [common] Widen hilbert index bytes beyond 8 dimensions (#9777) hope
  • (paimon) 09/20: [fs] Delegate skipFully to hadoop's IOUtils (#9779) hope
  • (paimon) 06/20: [common] Make quoting mean literal in array and row string cast rules (#9769) hope
  • (paimon) 16/20: [core] Make compact_manifest perform a full manifest sort (#9802) hope
  • (paimon) 10/20: [python] Support custom dataset readers in PaimonLeRobotDataset (#9785) hope
  • (paimon) 02/20: [python] Stream training samples into native vector index trainers (#9758) hope
  • (paimon) 19/20: [python] Vectorize batch raw vector scoring (#9796) hope
  • (paimon) 17/20: [core] Fix nullable file records in the manifest Avro reader (#9816) hope
  • (paimon) 15/20: [core] Support manifest sorting for non-partitioned bucketed tables (#9808) hope
  • (paimon) 12/20: [python] Fix bucket registration for dynamic bucket overflow (#9788) hope
  • (paimon) 05/20: [api] Require field ids for schema fields, auto-assign them for parameter lists (#9767) hope
  • (paimon) 11/20: [common] Bound the file-size memo in both local cache managers (#9775) hope
  • (paimon) 01/20: [common] Forward newTwoPhaseOutputStream through ResolvingFileIO and resolve it at commit (#9655) hope
• (paimon) branch master updated: [format] Follow Parquet list backward-compatibility rules in schema clipping (#9550) lzljs3620320
• (paimon) branch release-2.1 updated (08f3fa8bdf -> 2a5cbe906c) hope
  • (paimon) 19/20: [core] Fix bucket-first manifest sorting for postpone tables hope
  • (paimon) 05/20: [python] Batch vector writes when building global indexes (#9752) hope
  • (paimon) 16/20: [python] Stream batches when building generic global indexes (#9751) hope
  • (paimon) 08/20: [python] Use persisted vector metrics for raw search and refinement (#9754) hope
  • (paimon) 02/20: [core] Improve the error when listing a Format Table custom partition location fails (#9748) hope
  • (paimon) 17/20: [lumina] Release the vector file when FileBackedDataset fails to construct (#9481) hope
  • (paimon) 01/20: [python] Support linear interpolation in temporal alignment (#9747) hope
  • (paimon) 06/20: [python] Share raw vector scans across batch queries (#9753) hope
  • (paimon) 04/20: [test] Switch minio test image to quay.io (#9761) hope
  • (paimon) 10/20: [python] Support window joins in temporal alignment (#9760) hope
  • (paimon) 20/20: [common] Fix JSON round-trip of temporal and decimal predicate literals (#9640) hope
  • (paimon) 12/20: [ci] Run affected test suites through reusable workflows (#9787) hope
  • (paimon) 18/20: [core] Support bucket-first sorting for manifest files (#9792) hope
  • (paimon) 07/20: [python] Read native vector index ranges with bounded concurrency (#9756) hope
  • (paimon) 03/20: [python] Close vector readers when search raises synchronously (#9755) hope
  • (paimon) 13/20: [core] Support bucket pruning for manifest files (#9791) hope
  • (paimon) 15/20: [python] Vectorize raw vector scoring and refinement in bounded blocks (#9759) hope
  • (paimon) 14/20: [core] Simplify ManifestFileMeta constructors and field order hope
  • (paimon) 09/20: [python] Read LeRobot video datasets for training (#9757) hope
  • (paimon) 11/20: [python] Support reading shared-shredding maps (#9763) hope
• (paimon) branch release-2.1 updated (7152705145 -> 08f3fa8bdf) hope
  • (paimon) 18/20: [api] Fall back to the HTTP status when the error body omits code (#9661) hope
  • (paimon) 07/20: [spark] Support source table fileio for blob descriptors on spark (#9598) hope

• Earlier messages
• Later messages