jiayuasu commented on code in PR #124:
URL:
https://github.com/apache/sedona-spatialbench/pull/124#discussion_r3607862330
##########
docs/queries.md:
##########
@@ -259,20 +260,13 @@ JOIN customer c
ON t.t_custkey = c.c_custkey
GROUP BY c.c_custkey, c.c_name, pickup_month
HAVING dropoff_count > 5 -- Only include repeat customers
-ORDER BY monthly_travel_hull_area DESC, c.c_custkey ASC
+ORDER BY dropoff_count DESC, c.c_custkey ASC, pickup_month ASC
Review Comment:
Good catch — fixed in 8f1b62e. Q5 now orders by `monthly_travel_hull_area
DESC` (with `c_custkey`, `pickup_month` as tiebreakers) instead of
`dropoff_count`, across all four implementations plus the docs and notebook.
Ranking by the hull area forces the convex hull to be computed for every group
in order to find the top 100, so the `LIMIT` cannot be short-circuited by a
top-k that skips the spatial work — which was exactly your concern. Regenerated
the sample output on SedonaDB (SF1) to match.
##########
docs/index.md:
##########
@@ -81,7 +81,8 @@ FROM trip t
JOIN building b
ON ST_DWithin(t.t_pickup_loc, b.b_boundary, 500)
GROUP BY b.b_buildingkey, b.b_name
-ORDER BY nearby_pickup_count DESC;
+ORDER BY nearby_pickup_count DESC
Review Comment:
Fixed in 8f1b62e — added `b_buildingkey ASC` before the `LIMIT` in both
`index.md` and `index.zh.md` so the bounded top-100 is deterministic, matching
Q8.
--
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]
