R-SET NOT_EQUAL
student_club · mini_dev_postgresql from https://bird-bench.oss-cn-beijing.aliyuncs.com/minidev.zip (sha256 cc48ba16838204e4e214512030cb572eeb5f7bcdd999bae4b9b6ff12ec13b92f, downloaded 2026-09-07), member minidev/MINIDEV/mini_dev_postgresql.json
Among the members with t-shirt size of medium, what is the percentage of the amount 50 received by the Student_Club?
the hint the set supplies: t_shirt_size = 'Medium' where position = 'Member'; percentage = DIVIDE(COUNT(amount = 50), COUNT(member_id)) * 100
NOT_EQUAL states that these two statements disagree on this data under this rule. It does not state which of them is wrong.
read by hand, 2026-09-04: C the tool's rule, not the question's: the same number as int8 against numeric or float8 against numeric
float8 against numeric or int8, one row, same value
a maintainer's reading of this question, out of classification.json, copied from plans/reports/prediction-mode-260904-real-predictions/classification.json. It is not a verdict and nothing above it was computed from it.
The two statements name their result columns differently. A projection is compared by position and declared type, so these names did not decide the verdict. They are part of the canonical rendering each result is hashed under, so the two result hashes differ with them.
SELECT CAST(SUM(CASE WHEN T2.amount = 50 THEN 1.0 ELSE 0 END) AS REAL) * 100 / NULLIF(COUNT(T2.income_id), 0) FROM member AS T1 INNER JOIN income AS T2 ON T1.member_id = T2.link_to_member WHERE T1.position = 'Member' AND T1.t_shirt_size = 'Medium'this statement states no ordering of its own
sha256:9b0af7126766a0df3e2480fa31270bd5146caccf67e4c6d8c1ef7653ff38ce31
SELECT
(COUNT(*) FILTER (WHERE i.amount = 50) * 100.0) / COUNT(m.member_id) AS percentage
FROM
member m
LEFT JOIN
income i ON m.member_id = i.link_to_member
WHERE
m.t_shirt_size = 'Medium'
AND m.position = 'Member'this statement states no ordering of its own
sha256:73d671f4ab1f112381e2ef7cb3f7d5f21a638d7943934fe75a3ca94c7673a00e
The marked tokens are where the two texts differ. Two statements that differ everywhere can return the same rows, and two that differ in one token can return other rows; the verdict above is read off the results.
from counterexample.json, 0 rows, up to 25 shown per side
| side | ?column?float8 |
|---|---|
| no rows |
from counterexample.json, 0 rows, up to 25 shown per side
| side | percentagenumeric |
|---|---|
| no rows |
set(second_rows) == set(gold_rows), float4/float8 cells as Python float as psycopg2 returns them, numeric as Decimal
https://github.com/bird-bench/mini_dev/blob/main/evaluation/evaluation_ex.py
result_eq: equal row counts and equal column counts, each row unordered as a quick rejection, then the two equal as a list when the gold text holds ORDER BY and as a multiset otherwise, under some permutation of the columns; DISTINCT is not stripped and re-executed, and the cells are PostgreSQL's as psycopg2 returns them
ruiqi-zhong/test-suite-sql-eval, exec_eval.py, result_eq, at commit 48cb78ec: https://github.com/ruiqi-zhong/test-suite-sql-eval/blob/48cb78ecf7f610620206283846c76751b18a1326/exec_eval.py
The class states what makes these two results unequal under this rule, read off the two results and nothing else. It does not state which of the two statements is wrong.
from counterexample.json, 1 row
| ?column?float8 |
|---|
| 100 |
from counterexample.json, 1 row
| percentagenumeric |
|---|
| 100.0000000000000000 |
A smell is a mechanical reason to read this gold statement again. It is a heuristic: it does not state that the statement is wrong, and a maintainer decides.
this statement orders by a text column holding only numbers, and ordering it as a number gives a different answer, so the gold may be sorting 9.5 above 10
the statement states no top level ORDER BY
{
"heuristic": true,
"reason": "the statement states no top level ORDER BY"
}this statement cuts its result at a LIMIT that does not decide which rows come back, so a different but equally correct statement can return other rows and score zero
the statement states no LIMIT
{
"heuristic": true,
"reason": "the statement states no LIMIT"
}rerun over the same rows in another physical order this statement gives another answer, so its result depends on how the rows are stored and not only on the data
{
"heuristic": true,
"rule": "R-SET",
"baseline_result_hash": "sha256:9b0af7126766a0df3e2480fa31270bd5146caccf67e4c6d8c1ef7653ff38ce31",
"baseline_result": {
"columns": [
{
"name": "?column?",
"declared_type": "float8"
}
],
"row_count": 1,
"truncated": false,
"rows_shown": 1,
"rows": [
[
{
"type": "dec",
"value": "100"
}
]
],
"result_hash": "sha256:9b0af7126766a0df3e2480fa31270bd5146caccf67e4c6d8c1ef7653ff38ce31"
},
"planner_statistics": {
"member": {
"last_analyze": null,
"last_autoanalyze": null,
"n_mod_since_analyze": 33
},
"income": {
"last_analyze": null,
"last_autoanalyze": null,
"n_mod_since_analyze": 36
}
},
"shuffle": {
"seed": "1",
"row_limit": 300000,
"tables": [
"member",
"income"
],
"tables_not_shuffled": [],
"tables_skipped_for_size": {
"laptimes": 400524,
"legalities": 427907,
"posthistory": 303155,
"trans": 1056320,
"yearmonth": 383282
},
"tables_not_reached_by_a_copy": {}
},
"shuffled_copies": {
"run": true,
"verdict": "equal",
"differs": false,
"result_hash": "sha256:9b0af7126766a0df3e2480fa31270bd5146caccf67e4c6d8c1ef7653ff38ce31",
"result": {
"columns": [
{
"name": "?column?",
"declared_type": "float8"
}
],
"row_count": 1,
"truncated": false,
"rows_shown": 1,
"rows": [
[
{
"type": "dec",
"value": "100"
}
]
],
"result_hash": "sha256:9b0af7126766a0df3e2480fa31270bd5146caccf67e4c6d8c1ef7653ff38ce31"
}
},
"plan_variant": {
"run": false,
"reason": "the plan variant was not asked for"
}
}SELECT CAST(SUM(CASE WHEN T2.amount = 50 THEN 1.0 ELSE 0 END) AS REAL) * 100 / NULLIF(COUNT(T2.income_id), 0) FROM member AS T1 INNER JOIN income AS T2 ON T1.member_id = T2.link_to_member WHERE T1.position = 'Member' AND T1.t_shirt_size = 'Medium'result_hash sha256:9b0af7126766a0df3e2480fa31270bd5146caccf67e4c6d8c1ef7653ff38ce31 recomputed from this JSON: match
record_hash sha256:0f00e3778eca9f1aab77257c5d0580c4c9380eca1d3c01f0448820fc48e33250 recomputed from this JSON: match
from evidence-gold.json, 1 row
| ?column?float8 |
|---|
| 100 |
SELECT
(COUNT(*) FILTER (WHERE i.amount = 50) * 100.0) / COUNT(m.member_id) AS percentage
FROM
member m
LEFT JOIN
income i ON m.member_id = i.link_to_member
WHERE
m.t_shirt_size = 'Medium'
AND m.position = 'Member'result_hash sha256:73d671f4ab1f112381e2ef7cb3f7d5f21a638d7943934fe75a3ca94c7673a00e recomputed from this JSON: match
record_hash sha256:1a73502bb8b2f2b88dbdbd1c1d907a617a7b9fde7da798ffc5afb649bd01aeb4 recomputed from this JSON: match
from evidence-second.json, 1 row
| percentagenumeric |
|---|
| 100.0000000000000000 |
re-run this statement read-only against PostgreSQL 16.15 (Debian 16.15-1.pgdg13+2) on aarch64-unknown-linux-gnu, compiled by gcc (Debian 14.2.0-19) 14.2.0, 64-bit | server=172.17.0.2/32:5432 | database=bird under the session settings and over the data this record's fixture digest names, and compare the two results under R-SET
re-run this statement read-only against PostgreSQL 16.15 (Debian 16.15-1.pgdg13+2) on aarch64-unknown-linux-gnu, compiled by gcc (Debian 14.2.0-19) 14.2.0, 64-bit | server=172.17.0.2/32:5432 | database=bird under the session settings and over the data this record's fixture digest names, and compare the two results under R-SET