Google Git Google Git
Sign in
apache/datafusion-comet/HEAD
  1. 785fdbc chore: refresh the native dependencies with cargo update (#7169) by Andy Grove · 9 hours ago main
  2. 6a6d71c fix: refuse codegen dispatch for a TRY cast that can put a null key in a map (#6178) by Han-Yin Chang · 12 hours ago
  3. c56f6a0 fix: fall back when an outer hash join's streamed order cannot be checked (#7126) by dustin · 12 hours ago
  4. eefa4dd test: cover speculation and executor loss in multi-task native Iceberg writes (#6601) by Yu Hsun Hung · 12 hours ago
  5. 886e0a9 fix: scope fallback reasons to the current conversion (#7095) by Yu Hsun Hung · 13 hours ago
  6. 6cc7b86 refactor: share native filter and projection metrics wrapper (#7056) by KUAN-HAO HUANG · 13 hours ago
  7. 897f257 fix: preserve native sliding integer sum overflow semantics (#6069) by KUAN-HAO HUANG · 13 hours ago
  8. 333bae8 fix: report output rows and time of the native window operator (#7140) by Oleks V · 13 hours ago
  9. 653c795 feat: support AtLeastNNonNulls natively (#6180) by KUAN-HAO HUANG · 14 hours ago
  10. a81a591 fix: log native panic backtraces instead of attaching them to the exception (#7042) by Liang-Chi Hsieh · 14 hours ago
  11. 06cff60 feat: execute scalar Arrow Python UDFs in native Comet (#6130) by Liang-Chi Hsieh · 14 hours ago
  12. 50ab6b0 perf: share one decoded copy of a broadcast among the tasks of an executor (#7016) by Andy Grove · 14 hours ago
  13. 58978d0 refactor: share Parquet field-id and name resolution between root and nested fields (#6990) by Andy Grove · 16 hours ago
  14. 5a50bd6 test: cover positional struct reconciliation in native CASE WHEN and COALESCE (#7149) by KAZUYUKI TANIMURA · 24 hours ago
  15. 78a0e96 fix: preserve Spark AVG buffer and overflow semantics (#5420) by Chao Sun · 26 hours ago
  16. da2ff08 fix: extract bundled native libraries into private directories (#6987) by Manu Zhang · 2 days ago
  17. ab884e2 fix: fall back to Spark for negative-scale decimal casts and arithmetic (#5050) by ChenChen Lai · 2 days ago
  18. 38d3929 perf: default the native Iceberg scan to two concurrent data files (#7132) by Andy Grove · 2 days ago
  19. 74392a9 test: add window operator benchmarks (#7139) by Oleks V · 3 days ago
  20. 9a3d2a2 perf: load the next Parquet file's footer and page index while the current file is read (#7094) by Andy Grove · 3 days ago
  21. bf49f1e chore: limit open PRs for users without write access to 3 (#7135) by Andy Grove · 3 days ago
  22. 666f86f perf: sort JVM shuffle spill pointers in the sorter's scratch space (#7125) by Liang-Chi Hsieh · 3 days ago
  23. 0b04b38 fix: evaluate the array_append item on rows whose array is NULL (#7025) by KAZUYUKI TANIMURA · 3 days ago
  24. 39380cc feat: opt in to join filter pushdown through deterministic residuals (#6431) by Chao Sun · 4 days ago
  25. e4777a6 fix: fall back for duplicate from_json fields (#6651) by Samarth Mukhija · 4 days ago
  26. a48c82d fix: fall back when an Iceberg equality delete is keyed on a nested field (#7122) by Yunchi Pang · 4 days ago
  27. 5a8c5c5 feat: enable native replace for safe inputs (#6680) by Yu Hsun Hung · 4 days ago
  28. d834665 test: verify native timestamp output timezone labels (#6930) by LinSimon-901101 · 4 days ago
  29. 348d38f fix: preserve Spark nested null ordering in native sort (#6968) by Alex Huang · 4 days ago
  30. 99c4873 perf: vectorize int, decimal, float and boolean casts to timestamp (#6969) by Bill · 4 days ago
  31. 177fef1 refactor: deduplicate split list construction (#7004) by Armin Azimi · 4 days ago
  32. 137c894 feat: support native wide-decimal hashing (#6548) by KUAN-HAO HUANG · 4 days ago
  33. cae599a feat: add config to disable or restrict native UDF library loading (#6775) by Andy Grove · 4 days ago
  34. 3565f2f fix: treat -0.0 and 0.0 as one value in arrays_overlap on Spark 4.0.5, 4.1.4 and 4.2.1 (#6989) by Andy Grove · 4 days ago
  35. 53d8867 perf: create a native plan's spill directories on its first spill (#7066) by Andy Grove · 4 days ago
  36. 47730e3 chore: fix stddev accumulator size and resolve session libhdfs schemes once per scan rule (#7120) by Liang-Chi Hsieh · 4 days ago
  37. 3355602 fix: keep a streamed side sorted outside the native plan in order through outer hash joins (#6835) by Andy Grove · 4 days ago
  38. 0c7c5d6 fix: serialize large-offset string and binary vectors with 32-bit offsets (#6371) by Andy Grove · 4 days ago
  39. 2a2aa05 perf: write every partition of a JVM shuffle spill in one native call (#7065) by Andy Grove · 4 days ago
  40. 4ab1130 ci: give the Delta contrib Maven cache its own key prefix (#7111) by Andy Grove · 4 days ago
  41. 115058d fix: preserve Parquet root field IDs after nested pruning (#7081) by LinSimon-901101 · 4 days ago
  42. e3503ef perf: start bypass shuffle partition writers on small pages that grow (#7059) by Andy Grove · 4 days ago
  43. 80b5181 refactor: share window frame bound offset serialization between lower and upper bounds (#7107) by ChenChen Lai · 4 days ago
  44. 85b9e39 build(deps): bump actions/setup-python from 6 to 7 (#7046) by dependabot[bot] · 4 days ago
  45. aacb2e7 fix: count grouped integer SUM state in the aggregate's memory reservation (#6370) by Andy Grove · 5 days ago
  46. 4989561 fix: build release jars from the same commit as the native libraries (#6377) by Andy Grove · 5 days ago
  47. 4a8a949 fix: fall back to Spark when Iceberg reports an output ordering (#7043) by Parth Chandra · 5 days ago
  48. d0f367d test: write ancient dates with CORRECTED rebase mode in the DATE RANGE window test (#7103) by Andy Grove · 5 days ago
  49. 0d0a1c6 fix: size the native runtime for standalone executors without spark.executor.cores (#6368) by Andy Grove · 5 days ago
  50. c9f5617 docs: consolidate Parquet and Iceberg documentation into one page each (#7090) by Andy Grove · 5 days ago
  51. 4ef1036 fix: handle NullType struct fields in JVM columnar shuffle (#7099) by Jaden Fix · 5 days ago
  52. d60cc1d perf: optimize JVM columnar-to-row conversion (#5496) by Peter Lee · 6 days ago
  53. 642660c ci: label Delta pull requests with area:Delta (#6965) by dustin · 6 days ago
  54. 14748af fix: return errors instead of panicking in native planner (#7087) by ChenChen Lai · 6 days ago
  55. 27f437a refactor: move MergeRows, Explode, aggregate, join and window planning into dedicated modules (#7014) by Andy Grove · 6 days ago
  56. a0cbb0f feat: bump iceberg-rust and write float and double identity partitions natively (#6851) by Andy Grove · 7 days ago
  57. 79ab485 docs: list known Parquet and Iceberg issues and add a data source audit log (#7052) by Andy Grove · 7 days ago
  58. f180d65 perf: reuse validity buffers and use unary kernels in native expressions (#6953) by Oleks V · 7 days ago
  59. d443662 feat: support PivotFirst aggregate for the optimized PIVOT fast path (#4798) by Andy Grove · 7 days ago
  60. 46b9c6a fix: stop building maps natively when the key type holds a FLOAT or DOUBLE (#6992) by Andy Grove · 7 days ago
  61. c4ca20e fix: fall back to Spark for NOT IN over an ordered streamed side (#6810) by dustin · 7 days ago
  62. e242944 fix: keep unconverted DPP filters in a scan's canonical form (#6270) by dustin · 7 days ago
  63. 1bad7c9 perf: make the native Iceberg writer's dictionary pre-pass 1.8x to 3.9x faster (#6937) by Andy Grove · 7 days ago
  64. 3a4a159 fix: read fs.comet.libhdfs.schemes for the scan rule from the Hadoop conf (#6838) by dustin · 7 days ago
  65. 6a7e56c fix: honor subexpression elimination off in the native null short-circuit (#6986) by dustin · 7 days ago
  66. 86d2d78 feat: native RANGE window frames with explicit offset on DATE ORDER BY (#4974) by ChenChen Lai · 7 days ago
  67. c3316fb refactor: simplify shuffle support predicates in CometShuffleExchangeExec (#7045) by ChenChen Lai · 7 days ago
  68. 758ef1c perf: read only the nested fields Spark keeps in the native Iceberg scan (#6725) by Oleks V · 7 days ago
  69. 0f2ae44 fix: give each nondeterministic dispatched expression its own kernel (#6712) by dustin · 7 days ago
  70. 53d398b fix: include try_add suggestion in ANSI SUM overflow error (#6732) by DeviousCardi · 7 days ago
  71. c230bf4 feat: add nested benchmark derived from TPC-H (#7044) by Oleks V · 8 days ago
  72. a45445c fix: compare a Remainder divisor with zero in Spark's ordering without ANSI (#6896) by Andy Grove · 8 days ago
  73. 5e7d918 fix: read each native scan partition through the object store its files live in (#6753) by dustin · 8 days ago
  74. 01709c7 fix: do not fail planning on a cast of a literal that fails to evaluate (#6826) by Liang-Chi Hsieh · 8 days ago
  75. 2a0f1cf fix: fall back to Spark for decimal AVG cases that return wrong results (#7017) by Andy Grove · 8 days ago
  76. 5ef55a7 fix: report row contents to native write statistics trackers (#6383) by LinSimon-901101 · 8 days ago
  77. d64915e fix: raise DIVIDE_BY_ZERO from native corr on a zero denominator in ANSI mode (#6830) by KAZUYUKI TANIMURA · 8 days ago
  78. dbefab7 fix: compact oversized nested list_extract buffers (#6498) by KUAN-HAO HUANG · 8 days ago
  79. c61a79c docs: document casts involving arrays, structs, and maps (#6320) by Andy Grove · 8 days ago
  80. 88241e8 build: use a single jni version across the native workspace (#6998) by Andy Grove · 8 days ago
  81. f2ed550 docs: correct native Iceberg write docs that drifted from the code (#7009) by Andy Grove · 8 days ago
  82. 72b9045 feat: expose per-scan S3 HTTP coverage and retries (#6429) by Chao Sun · 8 days ago
  83. 7a42a35 feat: run broadcast joins natively when an unsupported scan is on the build side (#6786) by Parth Chandra · 8 days ago
  84. 66c63bd fix: keep the values of sliced batches when coalescing a broadcast (#6900) by Oleks V · 9 days ago
  85. a230ae5 docs: add proposal to establish Apache Comet as a top-level project (#6817) by Andy Grove · 9 days ago
  86. e555dba feat: make native Iceberg writes opt-in for 1.2.0 (#6962) by Andy Grove · 9 days ago
  87. b513921 test: tag the Azure ABFS suites with AzureTest (#6942) by Oleks V · 9 days ago
  88. e4babc4 perf: store shuffle offsets as lengths with zstd (#6943) by Andy Grove · 9 days ago
  89. 62ba2ab perf: generate the in-memory cache reader's class once per executor (#6860) by Andy Grove · 9 days ago
  90. 9e9eb44 fix: make Comet's in-memory cache opt-in again (#6963) by Andy Grove · 9 days ago
  91. 7d8d9b1 feat: normalize floats in native array_distinct and array_union (#6563) by Andy Grove · 10 days ago
  92. 33f4f88 feat: run global aggregates with no grouping keys or functions natively (#6899) by Andy Grove · 10 days ago
  93. 8a4029c build: drop the compiled maven-wrapper.jar by switching to the script-only wrapper (#6678) by Manu Zhang · 10 days ago
  94. 5d78898 feat: run min and max over string and binary columns natively (#6927) by Oleks V · 10 days ago
  95. a1230c1 ci: retry the scalafix lint when Maven cannot resolve the plugin (#6061) by Andy Grove · 10 days ago
  96. 55ba50d test: add an end-to-end Iceberg read, repartition, and write benchmark (#6910) by Andy Grove · 10 days ago
  97. 5c7b3ea fix: match Spark statistical aggregate updates and merges (#6076) by KUAN-HAO HUANG · 10 days ago
  98. bb67101 feat: enable native Iceberg writes by default (#6664) by Andy Grove · 10 days ago
  99. 474b5ec perf: avoid duplicate Hadoop configs in Iceberg write detection (#6925) by Andy Grove · 10 days ago
  100. 987ee73 perf: read native-written Parquet footers without parsing Hadoop's configuration (#6923) by Andy Grove · 10 days ago