clintropolis commented on code in PR #18628:
URL: https://github.com/apache/druid/pull/18628#discussion_r2430552496
##########
benchmarks/src/test/java/org/apache/druid/benchmark/query/SqlBaseBenchmark.java:
##########
@@ -177,6 +180,12 @@ public enum BenchmarkStringEncodingStrategy
})
protected String complexCompression;
+ @Param({
+ "NONE",
+ "SMILE"
+ })
+ protected ObjectStorageEncoding objectStorageEncoding;
Review Comment:
nit: `jsonObjectStorageEncoding` would make it clearer that this is a
setting that only applies to measuring stuff with json columns
##########
benchmarks/src/test/java/org/apache/druid/benchmark/query/SqlBaseBenchmark.java:
##########
@@ -355,6 +368,11 @@ public void setup() throws JsonProcessingException
private void checkIncompatibleParameters()
{
+ // we only support NONE object storage encoding for auto schema with mmap
segments
+ if (ObjectStorageEncoding.NONE.equals(objectStorageEncoding) &&
!("auto".equals(schemaType)
Review Comment:
just checking for auto is maybe not quite correct, like we definitely want
to run them if 'auto' is set, but a non-auto benchmark dataset could also have
a json column that explicitly defines it as nested/auto too.
I think we want something like this inside this if
```
for (String dataSource : getDatasources()) {
final DimensionsSpec dims =
SqlBenchmarkDatasets.getSchema(dataSource).getDimensionsSpec();
if (dims.getDimensions().stream().noneMatch(x -> x instanceof
AutoTypeColumnSchema || x instanceof NestedDataColumnSchema)) {
System.exit(0);
}
}
```
--
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]
