can i not insert data via INSERT INTO ?

i am trying insert into test (strategy_id, amount) values (1, 10.5)

but i get an error

via UI

no,

it's a derived data store

ProcessingException(errorCode: 150,

`message: PQLParsingError: java.lang.RuntimeException: Unable to convert SqlNode: INSERT INTO`test` (

strategy_id

,

amount

)`

VALUES ROW(1, 10.5) to PinotQuery.Unknown node type: INSERT

at org.apache.pinot.sql.parsers.CalciteSqlParser.compileCalciteSqlToPinotQuery(CalciteSqlParser.java: 282)

at org.apache.pinot.sql.parsers.CalciteSqlParser.compileToPinotQuery(CalciteSqlParser.java: 91)

at org.apache.pinot.sql.parsers.CalciteSqlCompiler.compileToBrokerRequest(CalciteSqlCompiler.java: 33)

at org.apache.pinot.controller.api.resources.PinotQueryResource.getQueryResponse(PinotQueryResource.java: 158)

at org.apache.pinot.controller.api.resources.PinotQueryResource.handlePostSql(PinotQueryResource.java: 131)

at sun.reflect.GeneratedMethodAccessor150.invoke(Unknown Source)

at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java: 43)

at java.lang.reflect.Method.invoke(Method.java: 498)

at org.glassfish.jersey.server.model.internal.ResourceMethodInvocationHandlerFactory.lambda$static$0(ResourceMethodInvocationHandlerFactory.java: 52)

at org.glassfish.jersey.server.model.internal.AbstractJavaResourceMethodDispatcher$1.run(AbstractJavaResourceMethodDispatcher.java: 124)

at org.glassfish.jersey.server.model.internal.AbstractJavaResourceMethodDispatcher.invoke(AbstractJavaResourceMethodDispatcher.java: 167)

at org.glassfish.jersey.server.model.internal.JavaResourceMethodDispatcherProvider$TypeOutInvoker.doDispatch(JavaResourceMethodDispatcherProvider.java: 219)

at org.glassfish.jersey.server.model.internal.AbstractJavaResourceMethodDispatcher.dispatch(AbstractJavaResourceMethodDispatcher.java: 79))

which means you first write to some other source such as Kafka or Blob store (S3, GCS, HDFS) etc and Pinot can ingest from those sources using the respective API's

correct?

Hello, i read that multivalue aggregations mean that we have something like SELECT min(field1), max(field2), max(field3) and so one... so basically more accregations function... so first question is that correct? second question...is there a way to accept two column as input of an aggregator? for example MAX(field1, field2) ... yeah, i know reading that way it could sounds stupid... but maybe we need to follow specific logics inside the aggregator and for this reason accepting more columns as input could be useful.

Hi, Pinot does support multiple aggregations in one query, but that is not multi-value aggregation

do we have aggregators that accept more columns?

@User could you explain it ? i mean.. multiple aggregations is when we have select count(one), count(two) and so on...right?

multi-value aggregation should be myaggr(field1, field2)

no?

sum(m1), sum(m2)

-> multiple aggregations

myAggr(m1, m2)

-> multi-arg aggregations (only distinct takes multiple args, and a new one theta-sketch based index is in PR).

multi-value

-> This refers to columns where a row can have multiple values.

BTW, I until now the execution stack assumed that aggrs take single column (with execption of distinct). I recently enhanced code so that this is no longer a limitation, and now it becomes really easy to add aggrs with multi-args

If you are interested, here's a doc from @User https://app.gitbook.com/@apache-pinot/s/apache-pinot-cookbook/developers-and-contributors/extending-pinot/custom-aggregation-function on how to add your custom aggr

@User yeah i was talking with Kishore yesterday

but one moment when you said "multi-value" what do you really mean "a row can have multiple values" ?

i am a bit confused on it

however i am building something like myAggr(m1,m2)

starting from code like: https://github.com/apache/incubator-pinot/blob/master/pinot-core/src/main/java/org/apache/pinot/core/query/aggregation/function/AvgAggregationFunction.java

so basically something like Avg(m1, m2)

(it is obviously an example...but just to clarify what i am doing)

I think you have few questions, whose answers might be unrelated concepts. Let me go over them one by one.

First multi-valued columns

These are columns that can take multiple values. Consider a schema with columns (company, locations).