berkaysynnada commented on PR #17684:
URL: https://github.com/apache/datafusion/pull/17684#issuecomment-3316177281
```
DataFusion CLI v50.0.0
> WITH source AS (
SELECT
1 AS n,
'' AS a1, '' AS a2, '' AS a3, '' AS a4, '' AS a5, '' AS a6, '' AS
a7, '' AS a8,
'' AS a9, '' AS a10, '' AS a11, '' AS a12, '' AS a13, '' AS a14, ''
AS a15, '' AS a16,
'' AS a17, '' AS a18, '' AS a19, '' AS a20, '' AS a21, '' AS a22, ''
AS a23, '' AS a24,
'' AS a25, '' AS a26, '' AS a27, '' AS a28, '' AS a29, '' AS a30, ''
AS a31, '' AS a32,
'' AS a33, '' AS a34, '' AS a35, '' AS a36, '' AS a37, '' AS a38, ''
AS a39, '' AS a40
)
SELECT
sum(n) OVER (PARTITION BY
a1, a2, a3, a4
)
FROM source;
+----------------------------------------------------------------------------------------------------------------------------------+
| sum(source.n) PARTITION BY [source.a1, source.a2, source.a3, source.a4]
ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING |
+----------------------------------------------------------------------------------------------------------------------------------+
| 1
|
+----------------------------------------------------------------------------------------------------------------------------------+
1 row(s) fetched.
Elapsed 0.039 seconds.
> WITH source AS (
SELECT
1 AS n,
'' AS a1, '' AS a2, '' AS a3, '' AS a4, '' AS a5, '' AS a6, '' AS
a7, '' AS a8,
'' AS a9, '' AS a10, '' AS a11, '' AS a12, '' AS a13, '' AS a14, ''
AS a15, '' AS a16,
'' AS a17, '' AS a18, '' AS a19, '' AS a20, '' AS a21, '' AS a22, ''
AS a23, '' AS a24,
'' AS a25, '' AS a26, '' AS a27, '' AS a28, '' AS a29, '' AS a30, ''
AS a31, '' AS a32,
'' AS a33, '' AS a34, '' AS a35, '' AS a36, '' AS a37, '' AS a38, ''
AS a39, '' AS a40
)
SELECT
sum(n) OVER (PARTITION BY
a1, a2, a3, a4, a5
)
FROM source;
+---------------------------------------------------------------------------------------------------------------------------------------------+
| sum(source.n) PARTITION BY [source.a1, source.a2, source.a3, source.a4,
source.a5] ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING |
+---------------------------------------------------------------------------------------------------------------------------------------------+
| 1
|
+---------------------------------------------------------------------------------------------------------------------------------------------+
1 row(s) fetched.
Elapsed 0.018 seconds.
```
I missed the case where there are no ordering constraints. In that scenario,
ordering_satisfy() returns true for nearly all combinations, so we were still
keeping O(4^n) orderings despite pruning. I've now updated the approach as
greedy.
1) For each partition column, try 4 sort options and pick the first that
works
2) Move to the next column only with the chosen option
3) Stop immediately if any column has no valid option
Can you check this updated version @findepi ?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
