Skip to content

[datafusion] Fix "wrong results" 🙄 in DataFusion due toto_char dialect differences - #7

Merged
toschmidt merged 2 commits into
dbolympics:mainfrom
alamb:alamb/fix_date_formatting
Oct 23, 2026
Merged

toschmidt merged 2 commits into
dbolympics:mainfrom
alamb:alamb/fix_date_formatting

Conversation

@alamb

@alamb alamb commented Oct 23, 2026 •

Copy link
Copy Markdown
Contributor

DataFusion's to_char format string is different than used in this benchmark which results in "wrong results"

DuckDB has an override here for this. DataFusion should have one too.

This brings the errors from "34 wrong results" to "12 wrong results" for each config

Details

Before this fix:

alamb@aal-dev:~/dbolympics$ ./target/release/dbo-bench run     --benchmark bench/benchmark_sf1.toml     --system datafusion     --config sf1-c8a.2xlarge     --keep
building image localhost/dbo-base
building image localhost/dbo-datafusion
datafusion / sf1-c8a.2xlarge: sf 1, 8 threads (cores 0-7), 16GB memory, keys none
datafusion / sf1-c8a.2xlarge: run 1: loaded 2024-and-earlier in 0.1 s
datafusion / sf1-c8a.2xlarge: run 1: 333 queries ran (6.6 s in total), 0 timeouts, 76 errors
datafusion / sf1-c8a.2xlarge: run 2: loaded 2025-Q1 in 0.1 s
datafusion / sf1-c8a.2xlarge: run 2: 332 queries ran (7.5 s in total), 0 timeouts, 77 errors
datafusion / sf1-c8a.2xlarge: run 3: loaded 2025-Q2 in 0.1 s
datafusion / sf1-c8a.2xlarge: run 3: 332 queries ran (7.9 s in total), 0 timeouts, 77 errors
datafusion / sf1-c8a.2xlarge: run 4: loaded 2025-Q3 in 0.1 s
datafusion / sf1-c8a.2xlarge: run 4: 332 queries ran (8.2 s in total), 0 timeouts, 77 errors
datafusion / sf1-c8a.2xlarge: run 5: loaded 2025-Q4 in 0.1 s
datafusion / sf1-c8a.2xlarge: run 5: 331 queries ran (9.0 s in total), 0 timeouts, 78 errors, 34 wrong results, 0 unchecked

After this fix:

alamb@aal-dev:~/dbolympics$ ./target/release/dbo-bench run     --benchmark bench/benchmark_sf1.toml     --system datafusion     --config sf1-c8a.2xlarge     --keep
building image localhost/dbo-base
building image localhost/dbo-datafusion
^[[B^[[B^[[B^[[Bdatafusion / sf1-c8a.2xlarge: sf 1, 8 threads (cores 0-7), 16GB memory, keys none
datafusion / sf1-c8a.2xlarge: run 1: loaded 2024-and-earlier in 0.1 s
datafusion / sf1-c8a.2xlarge: run 1: 333 queries ran (6.7 s in total), 0 timeouts, 76 errors
datafusion / sf1-c8a.2xlarge: run 2: loaded 2025-Q1 in 0.1 s
datafusion / sf1-c8a.2xlarge: run 2: 332 queries ran (7.5 s in total), 0 timeouts, 77 errors
datafusion / sf1-c8a.2xlarge: run 3: loaded 2025-Q2 in 0.1 s
datafusion / sf1-c8a.2xlarge: run 3: 332 queries ran (7.9 s in total), 0 timeouts, 77 errors
datafusion / sf1-c8a.2xlarge: run 4: loaded 2025-Q3 in 0.1 s
datafusion / sf1-c8a.2xlarge: run 4: 332 queries ran (8.2 s in total), 0 timeouts, 77 errors
datafusion / sf1-c8a.2xlarge: run 5: loaded 2025-Q4 in 0.1 s
datafusion / sf1-c8a.2xlarge: run 5: 331 queries ran (9.0 s in total), 0 timeouts, 78 errors, 12 wrong results, 0 unchecked
datafusion / sf1-c8a.2xlarge: results in /home/alamb/dbolympics/systems/datafusion/results/2026-10-23/sf1-c8a.2xlarge.csv

Background

The benchmark currently uses a postgres style formatting string YY-MMMM which returns a constant in DataFusion rather than formatting the fields of a timestamp.

> select to_char('2025-12-23'::timestamp, 'YY-MMM');
+--------------------------------------------+
| to_char(Utf8("2025-12-23"),Utf8("YY-MMM")) |
+--------------------------------------------+
| YY-MMM                                     |
+--------------------------------------------+
1 row(s) fetched.
Elapsed 0.001 seconds.

The correct syntax, per the DataFusion to_char documentation is:

> select to_char('2025-12-23'::timestamp, '%Y-%m');
+-------------------------------------------+
| to_char(Utf8("2025-12-23"),Utf8("%Y-%m")) |
+-------------------------------------------+
| 2025-12                                   |
+-------------------------------------------+
1 row(s) fetched.
Elapsed 0.001 seconds.

This accounts for a substantial number of the 'wrong results' for DataFusion (xx of the 34)

Standard to_char

In case anyone was curious, to_char is not part of the SQL Standard (at least not in my copy of "NCITS/ISO/IEC 9075-2:2011 [2012] -- (ISO/IEC 9075-2:2011, IDT)", which I will fully admit is a nightmare to read). I think to_char was introduced by postgres (or maybe Oracle back in the day)

Therefore this is not a DataFusion violation of the SQL standard, but a dialect incompatibility.

@alamb alamb changed the title Fix DataFusion "wrong results" 🙄 in DataFusion due toto_char dialect differences [datafusion] Fix "wrong results" 🙄 in DataFusion due toto_char dialect differences Oct 23, 2026
@toschmidt
toschmidt force-pushed the alamb/fix_date_formatting branch from 86d016d to b9601b1 Compare October 23, 2026 19:03
@toschmidt

Copy link
Copy Markdown
Contributor

/benchmark

@db-olympics-bot

db-olympics-bot Bot commented Oct 23, 2026 •

Copy link
Copy Markdown
Contributor

/benchmark

@toschmidt datafusion on sf1, sf10, sf100, sf1000 on commit b9601b1a25.

The results of each system: datafusion.

The results are the pull request's last commit, a95a44ad8b.

@db-olympics-bot

Copy link
Copy Markdown
Contributor

datafusion is done, system 1 of 1 of the run on commit b9601b1a25.

System Config Loads OK Wrong Error Timeout Time (OK) Geomean (OK) On main: OK, time
datafusion sf1-c8a.2xlarge 164 ms 319 12 78 0 4.49 s 10.7 ms 297, 4.2 s
datafusion sf1-c8a.large 164 ms 319 12 78 0 5.59 s 12.3 ms 297, 5.14 s
datafusion sf1-c8a.medium 228 ms 318 13 78 0 6.82 s 13.6 ms 296, 6.19 s
datafusion sf1-c8a.xlarge 170 ms 319 12 78 0 4.79 s 11.1 ms 297, 4.44 s
datafusion sf10-c8a.2xlarge 174 ms 315 13 79 2 13 s 17.9 ms 292, 12.4 s
datafusion sf10-c8a.4xlarge 179 ms 315 13 78 3 8.45 s 18.4 ms 292, 7.83 s
datafusion sf10-c8a.8xlarge 218 ms 315 13 78 3 11.7 s 27.3 ms 292, 10.8 s
datafusion sf10-c8a.xlarge 173 ms 315 13 79 2 14.1 s 20.3 ms 292, 13 s
datafusion sf100-c8a.12xlarge 296 ms 314 14 79 2 27.4 s 53.2 ms 292, 25.7 s
datafusion sf100-c8a.16xlarge 305 ms 314 14 79 2 31.4 s 63.8 ms 292, 29.5 s
datafusion sf100-c8a.24xlarge 319 ms 314 14 79 2 41 s 87.4 ms 292, 38.6 s
datafusion sf100-c8a.4xlarge 227 ms 315 14 79 1 30.8 s 43.4 ms 293, 28.5 s
datafusion sf100-c8a.8xlarge 273 ms 315 14 79 1 26.6 s 47.3 ms 293, 24.7 s
datafusion sf1000-c8a.12xlarge 1.94 s 312 13 80 4 172 s 160 ms 285, 162 s
datafusion sf1000-c8a.16xlarge 1.96 s 312 13 80 4 184 s 169 ms 285, 143 s
datafusion sf1000-c8a.24xlarge 2.07 s 312 13 80 4 190 s 187 ms 289, 178 s
Not ok (1501)
- datafusion / sf1-c8a.2xlarge: `q040` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q057` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q064` error: Error during planning: Correlated scalar subquery can only be used in Projection, Filter, Aggregate plan nodes
- datafusion / sf1-c8a.2xlarge: `q077` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q084` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q085` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q086` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q091` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q098` error: Error during planning: Correlated scalar subquery must be aggregated to return at most one row
- datafusion / sf1-c8a.2xlarge: `q103` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q104` error: Schema error: No field named ip_counts.tenant_id. Did you mean 'ip_counts.ip_count'?
- datafusion / sf1-c8a.2xlarge: `q113` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q121` error: Error during planning: Plan: ["BoundedWindowAggExec: wdw=[row_number() PARTITION BY [dsr.id] ORDER BY [abs(coalesce(CAST(lh.placed_at AS Date32) - CAST(dsr.received_at AS Date32), Int64(999999))) ASC 
- datafusion / sf1-c8a.2xlarge: `q124` wrong_result: row 1: col 7: "4.7745469785823559408465068842427332993371" vs "4.5621935874"; col 8: "12.9233499999999921215021458920091390609741" vs "12.0"
- datafusion / sf1-c8a.2xlarge: `q127` wrong_result: row 1: col 5: "2.5623722891566265060240963855421686746988" vs "2.2581259036"; col 6: "4.7629999999999999005240169935859739780426" vs "4.0"; col 8: "7.1755843373493975903614457831325301204819" vs "6.96
- datafusion / sf1-c8a.2xlarge: `q128` wrong_result: row 1: col 7: "1.68725" vs "1.5838000000"
- datafusion / sf1-c8a.2xlarge: `q131` error: This feature is not implemented: Physical plan does not support logical expression InSubquery(InSubquery { expr: Column(Column { relation: Some(Bare { table: "t" }), name: "tenant_id" }), subquery: <s
- datafusion / sf1-c8a.2xlarge: `q135` error: This feature is not implemented: Physical plan does not support logical expression InSubquery(InSubquery { expr: Column(Column { relation: Some(Bare { table: "wd" }), name: "tenant_id" }), subquery: <
- datafusion / sf1-c8a.2xlarge: `q136` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q147` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q149` error: Arrow error: Invalid argument error: Invalid comparison operation: Decimal128(38, 6) == Decimal128(28, 6)
- datafusion / sf1-c8a.2xlarge: `q151` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q154` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: true })
- datafusion / sf1-c8a.2xlarge: `q156` error: This feature is not implemented: Physical plan does not support logical expression OuterReferenceColumn(Field { name: "id", data_type: Int64 }, Column { relation: Some(Bare { table: "d" }), name: "id"
- datafusion / sf1-c8a.2xlarge: `q161` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q162` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q163` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q164` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q167` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q170` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q174` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q175` error: Error during planning: Invalid function 'json_value'.
- datafusion / sf1-c8a.2xlarge: `q176` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q178` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q179` error: Error during planning: Correlated scalar subquery must be aggregated to return at most one row
- datafusion / sf1-c8a.2xlarge: `q181` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q182` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q183` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q194` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q197` wrong_result: row 30: col 7: "128.2189101083333333333333333333333333333333" vs "128.20416666666665"
- datafusion / sf1-c8a.2xlarge: `q207` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q209` error: Arrow error: Invalid argument error: Invalid comparison operation: Decimal128(38, 6) == Decimal128(28, 6)
- datafusion / sf1-c8a.2xlarge: `q233` wrong_result: row 1: col 6: "66.6666666666666666666666666666666666666667" vs "100.0"
- datafusion / sf1-c8a.2xlarge: `q237` wrong_result: row 1: col 6: "73.8111647139903514817367332873880082701585" vs "100.0"
- datafusion / sf1-c8a.2xlarge: `q250` error: Error during planning: Cannot get result type for temporal operation Int32 * Interval(MonthDayNano): Invalid argument error: Invalid interval arithmetic operation: Interval(MonthDayNano) * Interval(Mo
- datafusion / sf1-c8a.2xlarge: `q253` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q259` error: Arrow error: Invalid argument error: Invalid comparison operation: Decimal128(38, 6) == Decimal128(28, 6)
- datafusion / sf1-c8a.2xlarge: `q260` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q261` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q263` error: Error during planning: Correlated scalar subquery must be aggregated to return at most one row
- datafusion / sf1-c8a.2xlarge: `q273` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q274` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q276` error: Schema error: No field named ip_counts.tenant_id. Did you mean 'ip_counts.cnt'?
- datafusion / sf1-c8a.2xlarge: `q287` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q288` wrong_result: row 1: col 5: "68.4789248625534514355528405620036652412951" vs "69.08979841172878"
- datafusion / sf1-c8a.2xlarge: `q289` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q292` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q293` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q304` wrong_result: row 1: col 4: "60.1663764751402592377635906364867479202941" vs "100.0"; col 5: "61.0949893596440317276068872122267363126330" vs "100.0"
- datafusion / sf1-c8a.2xlarge: `q312` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q315` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q341` error: Error during planning: Cannot get result type for temporal operation Int32 * Interval(MonthDayNano): Invalid argument error: Invalid interval arithmetic operation: Interval(MonthDayNano) * Interval(Mo
- datafusion / sf1-c8a.2xlarge: `q349` error: This feature is not implemented: Physical plan does not support logical expression InSubquery(InSubquery { expr: Column(Column { relation: Some(Bare { table: "c" }), name: "product_id" }), subquery: <
- datafusion / sf1-c8a.2xlarge: `q350` wrong_result: row 1: col 7: "87.5" vs "100.0000000000"
- datafusion / sf1-c8a.2xlarge: `q353` error: Schema error: No field named p.id.
- datafusion / sf1-c8a.2xlarge: `q358` error: Schema error: No field named p.id.
- datafusion / sf1-c8a.2xlarge: `q362` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: true })
- datafusion / sf1-c8a.2xlarge: `q363` error: Error during planning: Correlated scalar subquery must be aggregated to return at most one row
- datafusion / sf1-c8a.2xlarge: `q364` error: Schema error: No field named latest.tenant_id. Did you mean 'latest.user_id'?
- datafusion / sf1-c8a.2xlarge: `q365` error: This feature is not implemented: Physical plan does not support logical expression OuterReferenceColumn(Field { name: "user_id", data_type: Int64 }, Column { relation: Some(Bare { table: "o" }), name:
- datafusion / sf1-c8a.2xlarge: `q367` error: Error during planning: Correlated scalar subquery must be aggregated to return at most one row
- datafusion / sf1-c8a.2xlarge: `q368` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q370` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q373` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q376` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q377` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q379` error: Error during planning: Cannot coerce arithmetic expression Duration(µs) * Float64 to valid types
- datafusion / sf1-c8a.2xlarge: `q380` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q383` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q386` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q388` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q393` error: Schema error: No field named w.id.
- datafusion / sf1-c8a.2xlarge: `q396` error: Error during planning: Too many columns! The subquery should only return one column: column1, column2
- datafusion / sf1-c8a.2xlarge: `q399` error: Error during planning: Invalid function 'json_value'.
- datafusion / sf1-c8a.2xlarge: `q400` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.2xlarge: `q403` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.2xlarge: `q404` wrong_result: row 1: col 6: "80.3339517625231910946196660482374768089054" vs "100.0"
- datafusion / sf1-c8a.2xlarge: `q405` wrong_result: row 28: col 4: "82.7454416333333333333333333333333333333333" vs "82.73333333330000"
- datafusion / sf1-c8a.2xlarge: `q408` wrong_result: row 1: col 5: "78.2608695652173913043478260869565217391304" vs "100.0"
- datafusion / sf1-c8a.2xlarge: `q409` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.large: `q040` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.large: `q057` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.large: `q064` error: Error during planning: Correlated scalar subquery can only be used in Projection, Filter, Aggregate plan nodes
- datafusion / sf1-c8a.large: `q077` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.large: `q084` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.large: `q085` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.large: `q086` error: This feature is not implemented: Physical plan does not support logical expression ScalarSubquery(<subquery>)
- datafusion / sf1-c8a.large: `q091` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- datafusion / sf1-c8a.large: `q098` error: Error during planning: Correlated scalar subquery must be aggregated to return at most one row
- datafusion / sf1-c8a.large: `q103` error: This feature is not implemented: Physical plan does not support logical expression Exists(Exists { subquery: <subquery>, negated: false })
- ...

Run by the benchmark bot for pull request dbolympics#7, approved by @toschmidt: dbolympics#7 (comment)

DB-Olympics-Results: datafusion
Benchmarked-System: datafusion d5b86d5a7f52801be1d0560eda5737dfbe75d275
Approved-By: toschmidt
@hbirler

hbirler commented Oct 23, 2026

Copy link
Copy Markdown
Contributor

Thanks for looking into this! We will also review all wrong results again before the official announcement.

@toschmidt
toschmidt merged commit fc7cd40 into dbolympics:main Oct 23, 2026
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants