Apache DataFusion SQL Query Engine
docs: add partitioned ClickBench SQL example (#23637)
ByteBaker955f70f3f510
diff --git a/benchmarks/README.md b/benchmarks/README.mdindex de69875a2c..34c67e5151 100644--- a/benchmarks/README.md+++ b/benchmarks/README.md@@ -507,23 +507,35 @@ The runner applies two ClickBench-specific setup steps automatically:runner enables the parquet `binary_as_string` option so those columnsare read as strings.-If you set up ClickBench manually through SQL, use the same `EventDate`-view pattern:+If you set up ClickBench manually through SQL, register the single-file+dataset as follows:```sqlCREATE EXTERNAL TABLE hits_rawSTORED AS PARQUETLOCATION 'benchmarks/data/hits.parquet';+```++For the partitioned dataset, register the directory and enable+`binary_as_string`:++```sql+CREATE EXTERNAL TABLE hits_raw+STORED AS PARQUET+LOCATION 'benchmarks/data/hits_partitioned'+OPTIONS ('binary_as_string' 'true');+```++After registering either dataset as `hits_raw`, create the `hits` view with+the required `EventDate` conversion:+```sqlCREATE VIEW hits ASSELECT * EXCEPT ("EventDate"),CAST(CAST("EventDate" AS INTEGER) AS DATE) AS "EventDate"FROM hits_raw;```-For the partitioned dataset, use `benchmarks/data/hits_partitioned` and-add `OPTIONS ('binary_as_string' 'true')` to the external table statement.-From the repository root, download data and run the default ClickBenchqueries against the single parquet file: