james-willis commented on PR #141:
URL:
https://github.com/apache/sedona-spatialbench/pull/141#issuecomment-5484499448
Withdrawing this. Further testing undercut the premise it was built on.
**1. The SedonaDB override is not redundant — it is load-bearing.** DuckDB
inherits the
base Q5, and DuckDB spatial has no `ST_Collect_Agg` (`duckdb_functions()`
lists
`ST_Collect` as a scalar plus `ST_Envelope_Agg` and `ST_Union_Agg`). So
moving the base to
the aggregate form does not delete an override, it relocates one: minus
SedonaDB, plus
DuckDB. Net count unchanged.
**2. `ST_Collect_Agg` lands in Sedona Spark 1.8.1, not earlier.** Measured,
one Spark
3.5.3 session per version:
| Sedona Spark | `ST_Collect(ARRAY_AGG(...))` | `ST_Collect_Agg(...)` |
|---|---|---|
| 1.7.2 | works | `UNRESOLVED_ROUTINE` |
| 1.8.0 | works | `UNRESOLVED_ROUTINE` |
| 1.8.1 | works | works |
| 1.9.1 | works | works |
The base class is the documented Spark/Sedona dialect and surfaces in
`docs/queries.md`
and the notebooks, but nothing executes it — `get_sql_queries()` maps only
`duckdb` and
`sedonadb`, and there is no Spark runner in `run_benchmark.py`. So the
change buys no
measured benefit and would narrow the documented reference query to Sedona
>= 1.8.1, in a
repo that pins no Sedona version. That trade is not worth it.
Everything else here did check out — at SF1 and SF10 both spellings are
bit-identical in
Sedona Spark 1.9.1 and match `benchmark/answers/sf{1,10}/q5.csv` — but "same
answer,
narrower compatibility, same override count" is not a good enough reason to
touch it.
One finding worth separating out, unrelated to the spelling debate:
`docs/queries.md`,
`docs/queries.zh.md` and `notebooks/queries.ipynb` show Q5 as
`ST_Collect(ST_GeomFromWKB(t.t_dropoffloc))` inside an `sd.sql(...)` call,
and SedonaDB
0.4.1 rejects that outright with `Invalid function 'st_collect'` — the
documented query
cannot be run as written. That looks like a real docs bug and I am happy to
send a
docs-only fix if maintainers want one.
--
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]
