Issues

• IN treats a column-dependent CASE WHEN / IF candidate as the constant NULL

  bugrequires-triage
  apache/datafusion-comet#6762 · Filter by author viirya opened on Oct 7, 2026

• least orders a null inside a struct or array last, where Spark orders it first

  bugrequires-triage
  apache/datafusion-comet#6759 · Filter by author viirya opened on Oct 7, 2026

• greatest/least compare struct arguments by field name instead of position when names differ only in case

  bugrequires-triage
  apache/datafusion-comet#6758 · Filter by author viirya opened on Oct 7, 2026

• perf: converting unsafe arrays of maps to Arrow is up to 1.7x slower when C2 inlines MapWriter.writeEntries

  performancerequires-triage
  apache/datafusion-comet#6757 · Filter by author andygrove opened on Oct 7, 2026

• spark.comet.expression.<name>.enabled has no effect on expressions inside a codegen-dispatched tree

  area:udfbugrequires-triage
  apache/datafusion-comet#6756 · Filter by author andygrove opened on Oct 7, 2026

• Codegen dispatcher: run more Scala UDF null-guard shapes inside the kernel

  area:udfenhancementperformancerequires-triage
  apache/datafusion-comet#6755 · Filter by author andygrove opened on Oct 7, 2026

• Stop ending Spark-to-Arrow batches at input file boundaries

  area:scanenhancementperformancerequires-triage
  apache/datafusion-comet#6752 · Filter by author andygrove opened on Oct 7, 2026
  

• Native Parquet writes on Spark 3.4/3.5 ignore maxRecordsPerFile

  area:writerbugrequires-triage
  apache/datafusion-comet#6751 · Filter by author andygrove opened on Oct 7, 2026

• Add an end-to-end HDFS write test for both native Parquet writers

  area:writerenhancementrequires-triage
  apache/datafusion-comet#6750 · Filter by author andygrove opened on Oct 7, 2026

• Spark SQL Writer Tests (manual) never runs Comet's native writer

  area:ciarea:writerbugrequires-triage
  apache/datafusion-comet#6749 · Filter by author andygrove opened on Oct 7, 2026

• Extend per-scan HTTP attempt and retry metrics to native Azure and GCS

  requires-triage
  apache/datafusion-comet#6748 · Filter by author sunchao opened on Oct 7, 2026

• Native CSV scan reads every partition through the first partition's object store

  requires-triage
  apache/datafusion-comet#6747 · Filter by author dwsmith1983 opened on Oct 7, 2026
  1

  

• Native Parquet scan reads files from another bucket when one partition spans buckets

  requires-triage
  apache/datafusion-comet#6746 · Filter by author dwsmith1983 opened on Oct 7, 2026
  1

  

• Nightly CI failed on 2026-10-07

  ci-nightly-failure
  apache/datafusion-comet#6744 · Filter by author github-actions opened on Oct 7, 2026

• arrays_overlap returns wrong answers or panics with a constant array beside a column

  requires-triage
  apache/datafusion-comet#6742 · Filter by author grorge123 opened on Oct 7, 2026

• [Bug] Native make_date mishandles Date32 overflow and skipped evaluation

  requires-triage
  apache/datafusion-comet#6741 · Filter by author pingzh opened on Oct 6, 2026
  1

  

• date_trunc SECOND/MILLISECOND wraps on Spark 4.2, where Spark raises long overflow

  area:expressionsbugrequires-triage
  apache/datafusion-comet#6739 · Filter by author andygrove opened on Oct 6, 2026

• Native mode treats all NaNs as one value on Spark 3.4, where Spark counts each NaN separately

  area:aggregationbugrequires-triage
  apache/datafusion-comet#6738 · Filter by author andygrove opened on Oct 6, 2026

• Design question: long-term, should native Iceberg scans run through DataFusion's ParquetSource instead of iceberg-rust's ArrowReader?

  requires-triage
  apache/datafusion-comet#6737 · Filter by author unikdahal opened on Oct 6, 2026
  4

• Explore Kubernetes in-place pod memory resize (Spark 4.2 ExecutorResizePlugin) for Comet's untracked native memory

  area:memoryenhancementrequires-triage
  apache/datafusion-comet#6727 · Filter by author andygrove opened on Oct 6, 2026

• Native Iceberg scan reads every field of a nested column even when Spark prunes it

  area:Icebergenhancementperformancerequires-triage
  apache/datafusion-comet#6724 · Filter by author comphead opened on Oct 6, 2026
  

• Row-input conversionTime metric includes the time the child spends producing rows

  area:scanenhancementrequires-triage
  apache/datafusion-comet#6723 · Filter by author andygrove opened on Oct 6, 2026
  

• Admit all array and map types in CometSparkToColumnarExec

  area:scanenhancementperformancerequires-triage
  apache/datafusion-comet#6722 · Filter by author andygrove opened on Oct 6, 2026
  

• Write strings and nested values in bulk when converting Spark rows to Arrow

  area:scanenhancementperformancerequires-triage
  apache/datafusion-comet#6721 · Filter by author andygrove opened on Oct 6, 2026
  

• Reuse Arrow buffers across batches in Spark-to-Arrow conversion instead of reallocating and zeroing them

  area:scanenhancementperformancerequires-triage
  apache/datafusion-comet#6720 · Filter by author andygrove opened on Oct 6, 2026