Messages by Thread
-
Re: [I] `percentile_cont` truncates Float64 interpolation weights to six decimal places [datafusion]
mesubash (via GitHub) via github
-
[PR] test(scheduler): share plan, graph and server fixtures in the scheduler tests [datafusion-ballista]
villebro (via GitHub) via github
-
[I] PostgreSQL: support `CREATE TABLE ... OF <type>` [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
Re: [I] Performance review: shuffle, executor, scheduler and client findings (tracking) [datafusion-ballista]
andygrove (via GitHub) via github
-
Re: [I] PostgreSQL DO statement [datafusion-sqlparser-rs]
LucaCappelletti94 (via GitHub) via github
-
Re: [I] Wrong results: Correlated subquery confuses same-named columns of two of its relations [datafusion]
Awesome06 (via GitHub) via github
-
[I] DELETE fails to resolve columns qualified by the target table alias [datafusion]
QuakeWang (via GitHub) via github
-
[PR] fix(optimizer): error on correlated subqueries with ambiguous duplicate columns across relations [datafusion]
Awesome06 (via GitHub) via github
-
[PR] [branch-56] fix: size aggregate spill batches by bytes as well as rows (backport #25853) [datafusion]
andygrove (via GitHub) via github
-
Re: [I] Hash aggregate with a few large groups fails after spilling: spill batches are sized by rows, not bytes [datafusion]
andygrove (via GitHub) via github
-
Re: [PR] perf: combine small columnar batches before caching them [datafusion-comet]
andygrove (via GitHub) via github
-
[I] Rewrite escaped-regex error checks in sqllogictest files as plain text [datafusion]
jayzhan211 (via GitHub) via github
-
[I] Flight SQL can report an intermediate stage's schema for adaptive jobs [datafusion-ballista]
andygrove (via GitHub) via github
-
[PR] feat: run a job's final root on the scheduler [datafusion-ballista]
NoahKusaba (via GitHub) via github
-
[I] Wrapped `file_row_index()`/`input_file_name()`/`get_field()` lose leaf pushdown past `RepartitionExec` (`Projection` not in extraction allowlist) [datafusion]
cool9850311 (via GitHub) via github
-
[PR] docs: Clarify `EquivalenceClass`, `ConstExpr`, and `OrderingEquivalenceClass` docs [datafusion]
alamb (via GitHub) via github
-
[I] array_concat of a 2D and a 1D list returns `[NULL]` or `NULL` for an all-NULL row depending on the other rows in the batch [datafusion]
raghavyadav01 (via GitHub) via github
-
Re: [I] Break start_executor_process_inner into startup phases [datafusion-ballista]
juliomax (via GitHub) via github
-
Re: [I] [dbolympics] Correlated scalar subqueries with `ORDER BY ... LIMIT 1` fail with "Physical plan does not support logical expression ScalarSubquery" [datafusion]
princejain756 (via GitHub) via github
-
[PR] fix: keep URL passwords and query strings out of rejected path reasons [datafusion-comet]
dwsmith1983 (via GitHub) via github
-
[PR] fix(substrait): keep same-input join predicates as filters [datafusion]
bvolpato (via GitHub) via github
-
[I] PostgreSQL: support locale options on `CREATE DATABASE` [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
Re: [PR] test: cover push scheduling in client integration tests [datafusion-ballista]
andygrove (via GitHub) via github
-
Re: [PR] fix: stop gating columnar shuffle on native-serde checks it never uses [datafusion-comet]
Visorgood (via GitHub) via github
-
[PR] PostgreSQL: Support `ALTER TYPE ... RENAME ATTRIBUTE` [datafusion-sqlparser-rs]
LucaCappelletti94 (via GitHub) via github
-
[PR] Convert escaped-regex error checks to plain text in 7 sqllogictest files [datafusion]
Goldloli (via GitHub) via github
-
[PR] fix: sum projected column byte sizes in project_statistics [datafusion]
joseph-isaacs (via GitHub) via github
-
[I] PostgreSQL: support `ALTER TABLE ... ALTER COLUMN ... SET STORAGE` [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: support `CREATE TABLESPACE` [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[PR] fix(client): report an unreadable first block of a shuffle fetch as a fetch failure [datafusion-ballista]
andygrove (via GitHub) via github
-
[PR] build(deps): bump parquet from 59.3.1 to 59.3.2 in /native/native-executors [datafusion-comet]
dependabot[bot] (via GitHub) via github
-
[PR] PostgreSQL: Support `ALTER TYPE ... OWNER TO` [datafusion-sqlparser-rs]
LucaCappelletti94 (via GitHub) via github
-
[PR] fix: keep integer and decimal precision in array_sum [datafusion]
s5dsn-eqee (via GitHub) via github
-
Re: [PR] perf: write a cached batch under the chunk size into one chunk of its exact size [datafusion-comet]
andygrove (via GitHub) via github
-
[PR] chore(deps): bump runs-on/action from 2.5.0 to 2.6.0 [datafusion]
dependabot[bot] (via GitHub) via github
-
[I] PostgreSQL: support operator classes in `PARTITION BY HASH (...)` key lists [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: accept an empty target list without FROM (`SELECT;`) [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[PR] docs: Improve documentation for `SortKeyWalk` [datafusion]
alamb (via GitHub) via github
-
Re: [I] AQE: immediate intermediate-shuffle reclaim on job success is a no-op under the adaptive planner [datafusion-ballista]
andygrove (via GitHub) via github
-
[PR] docs: Add example and cross-links to `GroupingEquivalenceClass` [datafusion]
alamb (via GitHub) via github
-
Re: [PR] feat: aggregate a large final hash table as hash buckets (experimental, off by default) [datafusion]
rluvaton (via GitHub) via github
-
[PR] feat: add single flag to hash and nested loop joins for scalar subqueries [datafusion]
shehabgamin (via GitHub) via github
-
[PR] fix: reject zero parquet record batch buffer capacity [datafusion]
WaterWhisperer (via GitHub) via github
-
[PR] Fix duplicated word in bytes_view test string [datafusion]
aarushkandukoori (via GitHub) via github
-
Re: [PR] docs: add AGENTS.md and a pr-review agent skill [datafusion-ballista]
andygrove (via GitHub) via github
-
Re: [PR] docs: refresh the README and the architecture diagram [datafusion-ballista]
andygrove (via GitHub) via github
-
[I] Aggregate spill: List<Dictionary> state whose values repeat within a group still fails under a memory limit [datafusion]
andygrove (via GitHub) via github
-
[PR] fix(scheduler): set TCP keepalive on scheduler gRPC connections [datafusion-ballista]
ump45nose (via GitHub) via github
-
[PR] perf: deduplicate executor metadata in partition locations [datafusion-ballista]
andygrove (via GitHub) via github
-
[I] Grouped `count(DISTINCT x)` on numbers takes a slower path [datafusion]
mkleen (via GitHub) via github
-
[I] Research: evaluate one spill file per spill event for sort-based shuffle [datafusion-ballista]
andygrove (via GitHub) via github
-
[I] Flaky chaos test: executor loss mid-fetch surfaces as h2 broken pipe instead of a fetch failure [datafusion-ballista]
andygrove (via GitHub) via github
-
[I] UPDATE / DELETE fail with "No field named __common_expr_1" when SET or WHERE repeats an expression [datafusion]
v28k (via GitHub) via github
-
Re: [I] DataFrame sort() order is lost by a later select() of a computed column [datafusion]
lakshaydevoo-arch (via GitHub) via github
-
[PR] fix: do not limit shared CTE body in top-k aggregation (#27219) [datafusion]
Hrushikesh-ramilla (via GitHub) via github
-
[PR] PostgreSQL: Support `ALTER TYPE ... SET SCHEMA` [datafusion-sqlparser-rs]
LucaCappelletti94 (via GitHub) via github
-
Re: [PR] feat: enable datafusion.optimizer.enable_window_topn by default [datafusion]
adriangbot (via GitHub) via github
-
[PR] docs: refresh the README and generate benchmark results from JSON [datafusion-ballista]
andygrove (via GitHub) via github
-
Re: [PR] fix: ignore ordering in approx_median [datafusion]
alamb (via GitHub) via github
-
[PR] fix(sql): resolve dotted field access on lambda parameters (#27088) [datafusion]
Hrushikesh-ramilla (via GitHub) via github
-
[PR] PostgreSQL: Support `ALTER TYPE ... ALTER ATTRIBUTE` [datafusion-sqlparser-rs]
LucaCappelletti94 (via GitHub) via github
-
[PR] PostgreSQL: Support `ALTER TYPE ... DROP ATTRIBUTE` [datafusion-sqlparser-rs]
LucaCappelletti94 (via GitHub) via github
-
[I] Better join support for correlated subqueries: single join, null-aware `RightMark`, partitioned null-aware joins [datafusion]
shehabgamin (via GitHub) via github
-
[PR] feat: support null-aware RightMark and partitioned null-aware joins [datafusion]
shehabgamin (via GitHub) via github
-
[I] PostgreSQL: support `CREATE RULE` [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[PR] PostgreSQL: Support `ALTER TYPE ... ADD ATTRIBUTE` [datafusion-sqlparser-rs]
LucaCappelletti94 (via GitHub) via github
-
[I] PostgreSQL: support `REINDEX` [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[PR] fix: fix: keep null-aware hash joins out of symmetric hash join conversion [datafusion]
shehabgamin (via GitHub) via github
-
[I] PostgreSQL: support `CREATE ACCESS METHOD ... TYPE INDEX HANDLER ...` [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: concatenate adjacent string literals (`'a' 'b'`) [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: tokenize multi-character operators starting with `=` [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: support 2PC (`PREPARE TRANSACTION` / `COMMIT PREPARED` / `ROLLBACK PREPARED`) [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: support the `TABLE tbl` query short form [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: support `ALTER SEQUENCE` [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: support the `WITH OPTIONS` column prefix in CREATE TABLE [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: support column-level `COMPRESSION` option in CREATE TABLE [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[I] PostgreSQL: support `argname OUT/INOUT type` argument order in CREATE FUNCTION [datafusion-sqlparser-rs]
putao520 (via GitHub) via github
-
[PR] fix: EnsureRequirements decides the mode of a hash join still in Auto [datafusion]
jayzhan211 (via GitHub) via github
-
Re: [PR] perf: flush partial batches with fixed group capacity (experiment) [datafusion]
jayzhan211 (via GitHub) via github
-
[PR] fix: do not extract '=' predicate into 'IS NOT DISTINCT FROM' join (#27215) [datafusion]
galaxytemple (via GitHub) via github
-
[I] Wrong rows when a partitioned join reorders the keys of an aggregate whose group columns share a name (`GROUP BY t1.a, t2.a`) [datafusion]
jayzhan211 (via GitHub) via github
-
[I] `count(*) FILTER (WHERE ...)` over an aggregate subquery fails with an internal error from `aggregate_statistics` [datafusion]
jayzhan211 (via GitHub) via github
-
Re: [PR] perf(aggregate): reuse dictionary hashes and group mappings [datafusion]
adriangbot (via GitHub) via github
-
[I] `WITH ... AS MATERIALIZED` returns wrong rows under `ORDER BY ... LIMIT`: top-k aggregation limits the shared CTE body [datafusion]
jayzhan211 (via GitHub) via github
-
Re: [I] [dbolympics] General framework to decorrelate the subqueries [datafusion]
jayzhan211 (via GitHub) via github
-
[I] Wrong results: `NOT IN` over two unbounded inputs loses its NULL handling (`SymmetricHashJoinExec` drops `null_aware`) [datafusion]
jayzhan211 (via GitHub) via github
-
[PR] refactor: factor out map-entry hashing loops into `hash_map_entries!` [datafusion-comet]
jinjin1919 (via GitHub) via github
-
[PR] [branch-56] fix: count the largest capacity seen for a shared buffer address (backport #27097) [datafusion]
alamb (via GitHub) via github
-
Re: [PR] test: add correlated subquery cases for general decorrelation [datafusion]
jayzhan211 (via GitHub) via github
-
[I] Make sqllogictest `--complete` write expected errors as plain text [datafusion]
jayzhan211 (via GitHub) via github
-
[PR] docs: show how to test error messages as plain text in sqllogictest [datafusion]
jayzhan211 (via GitHub) via github
-
[PR] feat: add DataSourceExec instrumentation hooks [datafusion]
alexanderbianchi (via GitHub) via github
-
[PR] fix: account for retained sort-key capacity in group clustering [datafusion]
mohitgurav20 (via GitHub) via github
-
Re: [PR] fix: resolve conflicting Parquet field claims consistently [datafusion-comet]
sam-1112 (via GitHub) via github
-
Re: [PR] perf: nested struct statistics pruning [datafusion-comet]
github-actions[bot] (via GitHub) via github
-
[PR] feat: warn about unrecognized spark.comet configuration keys [datafusion-comet]
tlfua (via GitHub) via github
-
[I] Wrong results: `=` predicate pushed into an `IS NOT DISTINCT FROM` join matches NULLs [datafusion]
jayzhan211 (via GitHub) via github
-
Re: [PR] fix: disable conflicting IfBraceChecker [datafusion-comet]
github-actions[bot] (via GitHub) via github
-
Re: [PR] fix: preserve window inputs during TopK aggregation [datafusion]
github-actions[bot] (via GitHub) via github
-
[PR] fix: warn about the deprecated memoryPool.fraction when it is set on the session [datafusion-comet]
tlfua (via GitHub) via github
-
Re: [I] Null-named attribute fails Arrow export with "field name cannot be null" [datafusion-comet]
tlfua (via GitHub) via github
-
[I] Recursive CTEs over Utf8View columns overcount shared buffers and exhaust the memory pool [datafusion]
akupchinskiy (via GitHub) via github
-
[PR] PostgreSQL: Accept `VIRTUAL` and omitted mode on generated columns [datafusion-sqlparser-rs]
LucaCappelletti94 (via GitHub) via github
-
[I] Poor parallel scaling for window queries over timestamps with time zones [datafusion]
akupchinskiy (via GitHub) via github
-
[PR] ci: label pull requests from workflow_run instead of pull_request_target [datafusion-comet]
andygrove (via GitHub) via github
-
Re: [I] Write blog post for 1.2.0 release [datafusion-comet]
andygrove (via GitHub) via github
-
[PR] fix: share the grouping key type checks between doConvert and canAggregateBeConverted [datafusion-comet]
tlfua (via GitHub) via github
-
[PR] test(core): share execution plan test fixtures [datafusion-ballista]
villebro (via GitHub) via github
-
Re: [PR] feat: prune unused columns from the body of a materialized CTE [datafusion]
kosiew (via GitHub) via github
-
Re: [PR] test: share collation coverage across Spark 4.x [datafusion-comet]
comphead (via GitHub) via github
-
[PR] chore(ci): bump the github-codeql group with 2 updates [datafusion-ballista]
dependabot[bot] (via GitHub) via github
-
[I] Add Spark-compatible AtLeastNNonNulls physical expression [datafusion]
rich7420 (via GitHub) via github
-
Re: [I] SanityCheckPlan error for two window functions with `target_partitions = 1` (`WindowAggExec` loses `can_repartition` when its children are replaced) [datafusion]
1fanwang (via GitHub) via github
-
[PR] fix: use 64-bit row offsets in native columnar-to-row conversion [datafusion-comet]
andygrove (via GitHub) via github