github  

Messages by Thread

• • Re: [PR] Fix array_except type mismatch issue #3646 [datafusion-comet] via GitHub
  • Re: [PR] Fix array_except type mismatch issue #3646 [datafusion-comet] via GitHub
  • Re: [PR] Fix array_except type mismatch issue #3646 [datafusion-comet] via GitHub
• [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
  • Re: [PR] perf: route final aggregate buckets by column [datafusion] via GitHub
• [PR] feat: support first_value/last_value aggregates over sliding window frames [datafusion] via GitHub
  • Re: [PR] feat: support first_value/last_value aggregates over sliding window frames [datafusion] via GitHub
  • Re: [PR] feat: support first_value/last_value aggregates over sliding window frames [datafusion] via GitHub
• [I] PiecewiseMergeJoin returns wrong results for List keys with NULL elements and fails on Struct keys [datafusion] via GitHub
  • Re: [I] PiecewiseMergeJoin returns wrong results for List keys with NULL elements and fails on Struct keys [datafusion] via GitHub
• Re: [I] Use Exact Hash-Table Allocation Accounting in `GroupValuesRows` [datafusion] via GitHub
• Re: [PR] fix(physical-plan): replace approximate `map_size` accounting with exact `HashTable::allocation_size()` in `GroupValuesRows` [datafusion] via GitHub
  • Re: [PR] fix(physical-plan): replace approximate `map_size` accounting with exact `HashTable::allocation_size()` in `GroupValuesRows` [datafusion] via GitHub
• [PR] chore(ci): bump taiki-e/install-action from 2.87.21 to 2.87.22 [datafusion-ballista] via GitHub
  • Re: [PR] chore(ci): bump taiki-e/install-action from 2.87.21 to 2.87.22 [datafusion-ballista] via GitHub
• Re: [I] `arrays_except` type mismatch [datafusion-comet] via GitHub
• Re: [I] [EPIC] Native Variant Support [datafusion-comet] via GitHub
• Re: [PR] feat: account JVM UDF Arrow allocations in Spark task memory [datafusion-comet] via GitHub
  • Re: [PR] feat: account JVM UDF Arrow allocations in Spark task memory [datafusion-comet] via GitHub
  • Re: [PR] feat: account JVM UDF Arrow allocations in Spark task memory [datafusion-comet] via GitHub
• Re: [I] Recursive CTE silently casts recursive term to the static term's type, producing wrong results or a non-terminating query [datafusion] via GitHub
• [PR] fix: validate recursive CTE column coercion [datafusion] via GitHub
  • Re: [PR] fix: validate recursive CTE column coercion [datafusion] via GitHub
  • Re: [PR] fix: validate recursive CTE column coercion [datafusion] via GitHub
• Re: [I] Add ExistenceJoin support to Comet native execution [datafusion-comet] via GitHub
• Re: [I] Native IF fails with "column types must match schema types" when its branches differ in nested nullability [datafusion-comet] via GitHub
• Re: [I] Support first_value/last_value aggregates as sliding window aggregates (blocks FILTER + windowed first/last value) [datafusion] via GitHub
• Re: [PR] feat: decouple partition location from executor metadata [datafusion-ballista] via GitHub
• Re: [PR] feat(scheduler): call try_acquire_job on active cache miss in update_task_statuses [datafusion-ballista] via GitHub
• Re: [PR] Scheduler client reuse [datafusion-ballista] via GitHub
• Re: [PR] WIP: feat(executor): allocator-backed OOM guard behind an opt-in oom-guard feature [datafusion-ballista] via GitHub
• Re: [PR] WIP: feat: treat ResourcesExhausted as a retriable task failure [datafusion-ballista] via GitHub
• Re: [PR] feat(core): eagerly reclaim job data when the client finishes consuming results [datafusion-ballista] via GitHub
• Re: [PR] feat: spilling hash join operator (opt-in, inner joins) [datafusion-ballista] via GitHub
• Re: [PR] fix(scheduler): reclaim intermediate shuffle data under AQE [datafusion-ballista] via GitHub
• Re: [PR] fix: coalesce AQE shuffle partitions per alignment group [datafusion-ballista] via GitHub
• Re: [PR] perf: passthrough last stream is `SortPreservingMergeStream` [datafusion] via GitHub
• Re: [PR] feat(spark): add levenshtein with optional threshold support [datafusion] via GitHub
• Re: [PR] feat: add native distinct-combined collect_list aggregate support [datafusion-comet] via GitHub
  • Re: [PR] feat: add native distinct-combined collect_list aggregate support [datafusion-comet] via GitHub
  • Re: [PR] feat: add native distinct-combined collect_list aggregate support [datafusion-comet] via GitHub
• [PR] MSSQL: Add typed TRY_PARSE expressions [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] MSSQL: Add typed TRY_PARSE expressions [datafusion-sqlparser-rs] via GitHub
• [PR] MSSQL: Preserve nonreserved keyword column expressions [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] MSSQL: Preserve nonreserved keyword column expressions [datafusion-sqlparser-rs] via GitHub
• Re: [I] Native Iceberg years/months/days/hours differ from Iceberg for some pre-1970 timestamps [datafusion-comet] via GitHub
• [I] MSSQL: Clarify Synapse final-column trailing-comma support [datafusion-sqlparser-rs] via GitHub
• [PR] MSSQL: Support deferred ON clauses in right-nested joins [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] MSSQL: Support deferred ON clauses in right-nested joins [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] MSSQL: Support deferred ON clauses in right-nested joins [datafusion-sqlparser-rs] via GitHub
• Re: [I] Observed metrics (Dataset.observe) on a cached DataFrame are lost with Comet's native cache scan [datafusion-comet] via GitHub
• Re: [I] Comet's in-memory cache reports its compressed size to the planner, which changes join planning [datafusion-comet] via GitHub
• [I] Native CASE WHEN and COALESCE reconcile struct fields by name instead of position [datafusion-comet] via GitHub
• [PR] fix: preserve decimal precision and scale in Spark negative [datafusion] via GitHub
• [PR] fix: fall back for Iceberg writes with float or double identity partitions [datafusion-comet] via GitHub
  • Re: [PR] fix: fall back for Iceberg writes with float or double identity partitions [datafusion-comet] via GitHub
  • Re: [PR] fix: fall back for Iceberg writes with float or double identity partitions [datafusion-comet] via GitHub
  • Re: [PR] fix: fall back for Iceberg writes with float or double identity partitions [datafusion-comet] via GitHub
  • Re: [PR] fix: fall back for Iceberg writes with float or double identity partitions [datafusion-comet] via GitHub
  • Re: [PR] fix: fall back for Iceberg writes with float or double identity partitions [datafusion-comet] via GitHub
• [I] AQE skew join split makes the join fall back to Spark with no fallback reason [datafusion-comet] via GitHub
  • Re: [I] AQE skew join split makes the join fall back to Spark with no fallback reason [datafusion-comet] via GitHub
• [PR] refactor: move write planning into dedicated module [datafusion-comet] via GitHub
  • Re: [PR] refactor: move write planning into dedicated module [datafusion-comet] via GitHub
• [I] perf: Comet hash joins use more task time than Spark (skewed SHJ with nested payload, BHJ) [datafusion-comet] via GitHub
  • Re: [I] perf: Comet hash joins use more task time than Spark (skewed SHJ with nested payload, BHJ) [datafusion-comet] via GitHub
• [PR] fix: [branch-1.1] build with Rust 1.99, which deprecates the legacy f64 constants and fetch_update (#6507) [datafusion-comet] via GitHub
  • Re: [PR] fix: [branch-1.1] build with Rust 1.99, which deprecates the legacy f64 constants and fetch_update (#6507) [datafusion-comet] via GitHub
• [PR] test: [branch-1.1] fix flaky mixed field id directory test on Spark 3.4 and 3.5 (#6312) [datafusion-comet] via GitHub
  • Re: [PR] test: [branch-1.1] fix flaky mixed field id directory test on Spark 3.4 and 3.5 (#6312) [datafusion-comet] via GitHub
• Re: [PR] feat: Sort-aware Iceberg reads in Comet via a per-partition streaming merge [datafusion-comet] via GitHub
  • Re: [PR] feat: Sort-aware Iceberg reads in Comet via a per-partition streaming merge [datafusion-comet] via GitHub
  • Re: [PR] feat: Sort-aware Iceberg reads in Comet via a per-partition streaming merge [datafusion-comet] via GitHub
• [PR] docs: add examples for walking and rewriting logical plans [datafusion] via GitHub
• Re: [PR] fix: avoid divide-by-zero in `avg` for negative-scale decimals [datafusion] via GitHub
• [PR] fix: preserve list element field in array_repeat [datafusion] via GitHub
  • Re: [PR] fix: preserve list element field in array_repeat [datafusion] via GitHub
• [PR] bench: introduce create_hash benchmarks [datafusion] via GitHub
  • Re: [PR] bench: introduce create_hash benchmarks [datafusion] via GitHub
  • Re: [PR] bench: introduce create_hash benchmarks [datafusion] via GitHub
  • Re: [PR] bench: introduce create_hash benchmarks [datafusion] via GitHub
• Re: [I] Reconcile the two config namespaces for native writes [datafusion-comet] via GitHub
• [I] Reduce retained buffer allocation when map lookups select short nested values [datafusion-comet] via GitHub
  • Re: [I] Reduce retained buffer allocation when map lookups select short nested values [datafusion-comet] via GitHub
• Re: [PR] fix: make restricted_column run linearly [datafusion] via GitHub
• [I] Native Iceberg sorted-merge scan rebuilds the delete-file cache per file (shared deletes re-read) [datafusion-comet] via GitHub
• [PR] perf(sort-shuffle): only gc view columns that are sparse [datafusion-ballista] via GitHub
  • Re: [PR] perf(sort-shuffle): only gc view columns that are sparse [datafusion-ballista] via GitHub
  • Re: [PR] perf(sort-shuffle): only gc view columns that are sparse [datafusion-ballista] via GitHub
• [I] Sort-shuffle writer compacts view columns even when they're already dense [datafusion-ballista] via GitHub
• Re: [I] Allow DataSouceExec to have child nodes that supply files at runtime [datafusion] via GitHub
• Re: [I] Unexpected failure parsing SQLite CREATE TABLE statement [datafusion-sqlparser-rs] via GitHub
• [I] Performance review: shuffle, executor, scheduler and client findings (tracking) [datafusion-ballista] via GitHub
• [PR] chore(ci): fix clippy errors from Rust 1.99 [datafusion-ballista] via GitHub
  • Re: [PR] chore(ci): fix clippy errors from Rust 1.99 [datafusion-ballista] via GitHub
• [PR] feat: Support Struct type in approx_distinct - #23663 [datafusion] via GitHub
  • Re: [PR] feat: Support RunEndEncoded type in approx_distinct [datafusion] via GitHub
  • Re: [PR] feat: Support RunEndEncoded type in approx_distinct [datafusion] via GitHub
  • Re: [PR] feat: Support RunEndEncoded type in approx_distinct [datafusion] via GitHub
  • Re: [PR] feat: Support RunEndEncoded type in approx_distinct [datafusion] via GitHub
• Re: [PR] Hive: Preserve numeric-prefix field access round trips [datafusion-sqlparser-rs] via GitHub
• [I] Let a Dialect override how a table name in FROM is parsed [datafusion-sqlparser-rs] via GitHub
  • Re: [I] Let a Dialect override how a table name in FROM is parsed [datafusion-sqlparser-rs] via GitHub
• [I] Clarify how to run native Comet -> Celeborn shuffle end-to-end [datafusion-comet] via GitHub
  • Re: [I] Clarify how to run native Comet -> Celeborn shuffle end-to-end [datafusion-comet] via GitHub
  • Re: [I] Clarify how to run native Comet -> Celeborn shuffle end-to-end [datafusion-comet] via GitHub
  • Re: [I] Clarify how to run native Comet -> Celeborn shuffle end-to-end [datafusion-comet] via GitHub
  • Re: [I] Clarify how to run native Comet -> Celeborn shuffle end-to-end [datafusion-comet] via GitHub
  • Re: [I] Clarify how to run native Comet -> Celeborn shuffle end-to-end [datafusion-comet] via GitHub
  • Re: [I] Clarify how to run native Comet -> Celeborn shuffle end-to-end [datafusion-comet] via GitHub
  • Re: [I] Clarify how to run native Comet -> Celeborn shuffle end-to-end [datafusion-comet] via GitHub
• [I] `array_repeat` fails on lists with non-nullable elements or a non-default element field name [datafusion] via GitHub
  • Re: [I] `array_repeat` fails on lists with non-nullable elements or a non-default element field name [datafusion] via GitHub
• Re: [PR] SQLite: Support identifier characters above U+FFFF [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] SQLite: Support identifier characters above U+FFFF [datafusion-sqlparser-rs] via GitHub
• Re: [PR] perf: optimizing take_n for DictionaryGroupValuesColumn [datafusion] via GitHub
• [I] signum(-0.0) returns 0.0 instead of -0.0 [datafusion-comet] via GitHub
  • Re: [I] signum(-0.0) returns 0.0 instead of -0.0 [datafusion-comet] via GitHub
  • Re: [I] signum(-0.0) returns 0.0 instead of -0.0 [datafusion-comet] via GitHub
• [PR] test: sweep Spark's -0.0 and NaN semantics across operators and float expressions [datafusion-comet] via GitHub
  • Re: [PR] test: sweep Spark's -0.0 and NaN semantics across operators and float expressions [datafusion-comet] via GitHub
• Re: [PR] Generic: Add spacing around dot in compound field access adjacent to numbers [datafusion-sqlparser-rs] via GitHub
• [I] Run array_contains on float elements natively with Spark's equality instead of through the codegen dispatcher [datafusion-comet] via GitHub
  • Re: [I] Run array_contains on float elements natively with Spark's equality instead of through the codegen dispatcher [datafusion-comet] via GitHub
• [I] percentile_approx returns different percentiles from Spark when its input holds a NaN with the sign bit set [datafusion-comet] via GitHub
  • Re: [I] percentile_approx returns different percentiles from Spark when its input holds a NaN with the sign bit set [datafusion-comet] via GitHub
• [PR] perf(shuffle-writer): coalesce small batches before writing passthrough shuffle files [datafusion-ballista] via GitHub
  • Re: [PR] perf(shuffle-writer): coalesce small batches before writing passthrough shuffle files [datafusion-ballista] via GitHub
  • Re: [PR] perf(shuffle-writer): coalesce small batches before writing passthrough shuffle files [datafusion-ballista] via GitHub
  • Re: [PR] perf(shuffle-writer): coalesce small batches before writing passthrough shuffle files [datafusion-ballista] via GitHub
  • Re: [PR] perf(shuffle-writer): coalesce small batches before writing passthrough shuffle files [datafusion-ballista] via GitHub
  • Re: [PR] perf(shuffle-writer): coalesce small batches before writing passthrough shuffle files [datafusion-ballista] via GitHub
  • Re: [PR] perf(shuffle-writer): coalesce small batches before writing passthrough shuffle files [datafusion-ballista] via GitHub
  • Re: [PR] perf(shuffle-writer): coalesce small batches before writing passthrough shuffle files [datafusion-ballista] via GitHub
• [I] Coalesce small batches in the passthrough shuffle writer [datafusion-ballista] via GitHub
• Re: [PR] Add support for PostgreSQL `CREATE RULE` [datafusion-sqlparser-rs] via GitHub
• Re: [PR] SQLite: Support multi-word type names [datafusion-sqlparser-rs] via GitHub
• Re: [PR] MsSql: Disallow unquoted `FROM` keyword as alias assignment expression [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] MsSql: Disallow unquoted `FROM` keyword as alias assignment expression [datafusion-sqlparser-rs] via GitHub
• Re: [PR] SQLite: Support signed and decimal numbers in type modifiers [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] SQLite: Support signed and decimal numbers in type modifiers [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] SQLite: Support signed and decimal numbers in type modifiers [datafusion-sqlparser-rs] via GitHub
• Re: [PR] SQLite: Support `CAST(expr AS)` without a type name [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] SQLite: Support `CAST(expr AS)` without a type name [datafusion-sqlparser-rs] via GitHub
• Re: [PR] Parse SQLite's `IS` and `IS NOT` operators [datafusion-sqlparser-rs] via GitHub
• Re: [PR] SQLite: support == as assignment operator in UPDATE SET [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] SQLite: support == as assignment operator in UPDATE SET [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] SQLite: support == as assignment operator in UPDATE SET [datafusion-sqlparser-rs] via GitHub
• Re: [PR] SQLite: Support string literals as column names in `CREATE TABLE` [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] SQLite: Support string literals as column names in `CREATE TABLE` [datafusion-sqlparser-rs] via GitHub
• Re: [PR] Check prefix operator spacing by rendered first char [datafusion-sqlparser-rs] via GitHub
• Re: [PR] DuckDB: Support nested block comments [datafusion-sqlparser-rs] via GitHub
• Re: [PR] Gate fuzz oracle targets behind features [datafusion-sqlparser-rs] via GitHub
• Re: [PR] Generic: Render prefix `!` operator with space before operator characters [datafusion-sqlparser-rs] via GitHub
  • Re: [PR] Generic: Render prefix `!` operator with space before operator characters [datafusion-sqlparser-rs] via GitHub
• Re: [I] Add a builder for `CreateExternalCatalog` and `DropCatalog` [datafusion] via GitHub
• [PR] build(deps): bump virtualenv from 21.5.0 to 21.7.13 [datafusion-python] via GitHub
• [PR] build(deps): bump urllib3 from 2.7.0 to 2.8.0 [datafusion-python] via GitHub
• [PR] build(deps): bump tornado from 6.5.8 to 6.5.9 [datafusion-python] via GitHub
• [PR] fix: match Spark's float semantics in array_remove and sort_array [datafusion-comet] via GitHub
  • Re: [PR] fix: match Spark's float semantics in array_remove and sort_array [datafusion-comet] via GitHub
  • Re: [PR] fix: match Spark's float semantics in array_remove and sort_array [datafusion-comet] via GitHub
• Re: [I] make_array returns wrong row count for null-array input [datafusion] via GitHub
• Re: [PR] feat: preserve bitmap-backed Parquet row selections [datafusion] via GitHub
  • Re: [PR] feat: preserve bitmap-backed Parquet row selections [datafusion] via GitHub
  • Re: [PR] feat: preserve bitmap-backed Parquet row selections [datafusion] via GitHub
  • Re: [PR] feat: preserve bitmap-backed Parquet row selections [datafusion] via GitHub
  • Re: [PR] feat: preserve bitmap-backed Parquet row selections [datafusion] via GitHub
  • Re: [PR] feat: preserve bitmap-backed Parquet row selections [datafusion] via GitHub
  • Re: [PR] feat: preserve bitmap-backed Parquet row selections [datafusion] via GitHub

• Earlier messages
• Later messages