Flink SQL : Joining Tables with timestamp in pure SQL syntax

Question

I have some troubles using the SQL syntax from Flink to join multiple tables when at least one of them have a time attribute column.

I have a table Table1 that use the schema (id, value1, rowtime), when rowtime is used as a flink rowtime.

I want to join this table with a table Table2, that use the schema (id, value2). The join must be done on matching id.

In the end, I want to group the result of this join by using a tumbling time window.

Is it possible to do this using only the SQL syntax ?

Here is an example of what I would like to do :

SELECT 
    Table1.id as id, 
    TUMBLE_END(rowtime, INTERVAL '10' SECOND),
    MAX(value1) as value1,
    MAX(value2) as value2       
FROM Table1 JOIN TABLE2 ON Table1.id = Table2.id
GROUP BY Table1.id, TUMBLE(rowtime, INTERVAL '10' SECOND)

But it give me the following error :

2019-11-12 16:37:57.191 [main] ERROR - Cannot generate a valid execution plan for the given query: 

FlinkLogicalCalc(expr#0..6=[{inputs}], id=[$t0], EXPR$1=[$t4], value1=[$t1], value2=[$t2])
  FlinkLogicalWindowAggregate(group=[{0}], value1=[MAX($2)], value2=[MAX($3)])
    FlinkLogicalCalc(expr#0..2=[{inputs}], expr#3=[0], proj#0..1=[{exprs}], value1=[$t3], value2=[$t3])
      FlinkLogicalJoin(condition=[=($0, $2)], joinType=[inner])
        FlinkLogicalTableSourceScan(table=[[Table1]], fields=[id, value1, rowtime], source=[KafkaTableSource(id, value1, rowtime)])
        FlinkLogicalTableSourceScan(table=[[Table2]], fields=[id, value2], source=[Table2_Type(id, value2)])

Rowtime attributes must not be in the input rows of a regular join. As a workaround you can cast the time attributes of input tables to TIMESTAMP before.
Please check the documentation for the set of currently supported SQL features.
org.apache.flink.table.api.TableException: Cannot generate a valid execution plan for the given query: 

FlinkLogicalCalc(expr#0..6=[{inputs}], id=[$t0], EXPR$1=[$t4], value1=[$t1], value2=[$t2])
  FlinkLogicalWindowAggregate(group=[{0}], value1=[MAX($2)], value2=[MAX($3)])
    FlinkLogicalCalc(expr#0..2=[{inputs}], expr#3=[0], proj#0..1=[{exprs}], value1=[$t3], value2=[$t3])
      FlinkLogicalJoin(condition=[=($0, $2)], joinType=[inner])
        FlinkLogicalTableSourceScan(table=[[kafkaDataStream]], fields=[id, value1, rowtime], source=[KafkaTableSource(id, value1, rowtime)])
        FlinkLogicalTableSourceScan(table=[[SensorConfigurationUpdateHTTP]], fields=[id, value2], source=[Table2_Type(id, value2)])

Rowtime attributes must not be in the input rows of a regular join. As a workaround you can cast the time attributes of input tables to TIMESTAMP before.
Please check the documentation for the set of currently supported SQL features.
    at org.apache.flink.table.api.TableEnvironment.runVolcanoPlanner(TableEnvironment.scala:387)
    at org.apache.flink.table.api.TableEnvironment.optimizePhysicalPlan(TableEnvironment.scala:302)
    at org.apache.flink.table.api.StreamTableEnvironment.optimize(StreamTableEnvironment.scala:816)
    at org.apache.flink.table.api.StreamTableEnvironment.writeToSink(StreamTableEnvironment.scala:379)
    at org.apache.flink.table.api.TableEnvironment.insertInto(TableEnvironment.scala:879)
    at org.apache.flink.table.api.Table.insertInto(table.scala:1126)
    ...

I also tried to cast my rowtime into a TIMESTAMP type (as advised by the error message), but then I am no able to process the Time Window anymore. It leads to the following error :

2019-11-12 16:44:52.473 [main] ERROR - Window can only be defined over a time attribute column.
org.apache.flink.table.api.ValidationException: Window can only be defined over a time attribute column.
    at org.apache.flink.table.plan.rules.datastream.DataStreamLogicalWindowAggregateRule.getOperandAsTimeIndicator$1(DataStreamLogicalWindowAggregateRule.scala:84)
    at org.apache.flink.table.plan.rules.datastream.DataStreamLogicalWindowAggregateRule.translateWindowExpression(DataStreamLogicalWindowAggregateRule.scala:89)
    at org.apache.flink.table.plan.rules.common.LogicalWindowAggregateRule.onMatch(LogicalWindowAggregateRule.scala:65)
    at org.apache.calcite.plan.AbstractRelOptPlanner.fireRule(AbstractRelOptPlanner.java:315)
    at org.apache.calcite.plan.hep.HepPlanner.applyRule(HepPlanner.java:556)
    at org.apache.calcite.plan.hep.HepPlanner.applyRules(HepPlanner.java:415)
    at org.apache.calcite.plan.hep.HepPlanner.executeInstruction(HepPlanner.java:252)
    at org.apache.calcite.plan.hep.HepInstruction$RuleInstance.execute(HepInstruction.java:127)
    at org.apache.calcite.plan.hep.HepPlanner.executeProgram(HepPlanner.java:211)
    at org.apache.calcite.plan.hep.HepPlanner.findBestExp(HepPlanner.java:198)
    at org.apache.flink.table.api.TableEnvironment.runHepPlanner(TableEnvironment.scala:360)
    at org.apache.flink.table.api.TableEnvironment.runHepPlannerSequentially(TableEnvironment.scala:326)
    at org.apache.flink.table.api.TableEnvironment.optimizeNormalizeLogicalPlan(TableEnvironment.scala:282)
    at org.apache.flink.table.api.StreamTableEnvironment.optimize(StreamTableEnvironment.scala:813)
    at org.apache.flink.table.api.StreamTableEnvironment.writeToSink(StreamTableEnvironment.scala:379)
    at org.apache.flink.table.api.TableEnvironment.insertInto(TableEnvironment.scala:879)
    at org.apache.flink.table.api.Table.insertInto(table.scala:1126)

Fabian Hueske Fabian Hueske · Accepted Answer · 2019-11-12T16:03:46

The join result cannot include a time attribute, because the join can not guarantee that the order of timestamps is preserved. Flink assumes that both tables are dynamic and can change at any point of time. A new record in table Table2 might join with the first record of Table1, producing results with timestamps in "random" order.

You can change that by adding a temporal constraint to the join. Either you define the query with a time-windowed join or you model Table2 as a temporal table and join Table1 with it.

Flink SQL : Joining Tables with timestamp in pure SQL syntax

1 Answers