[
https://issues.apache.org/jira/browse/FLINK-7062?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16653373#comment-16653373
]
ASF GitHub Bot commented on FLINK-7062:
---------------------------------------
twalthr commented on a change in pull request #6815: [FLINK-7062][cep][table]
Added basic support for MATCH_RECOGNIZE
URL: https://github.com/apache/flink/pull/6815#discussion_r225568576
##########
File path:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/calcite/RelTimeIndicatorConverter.scala
##########
@@ -97,8 +97,46 @@ class RelTimeIndicatorConverter(rexBuilder: RexBuilder)
extends RelShuttle {
LogicalSort.create(input, sort.collation, sort.offset, sort.fetch)
}
- override def visit(`match`: LogicalMatch): RelNode =
- convertMatch(`match`)
+ override def visit(matchRel: LogicalMatch): RelNode = {
+ // visit children and update inputs
+ val input = matchRel.getInput.accept(this)
+
+ // check if input field contains time indicator type
+ // materialize field if no time indicator is present anymore
+ // if input field is already materialized, change to timestamp type
+ val materializer = new RexTimeIndicatorMaterializer(
+ rexBuilder,
+ input.getRowType.getFieldList.map(_.getType))
+
+ // update input expressions
+ val patternDefs =
matchRel.getPatternDefinitions.mapValues(_.accept(materializer))
+ val measures = matchRel.getMeasures.mapValues(_.accept(materializer))
+ val partitionKeys = matchRel.getPartitionKeys.map(_.accept(materializer))
+ val interval = if (matchRel.getInterval != null) {
+ matchRel.getInterval.accept(materializer)
+ } else {
+ null
+ }
+
+ // materialize all output types
+ // TODO allow passing through for rowtime accessor function, once
introduced
+ val output = materializerUtils.materializeTypes(matchRel.getRowType, _ =>
true)
Review comment:
Rename to `outputType`.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Support the basic functionality of MATCH_RECOGNIZE
> --------------------------------------------------
>
> Key: FLINK-7062
> URL: https://issues.apache.org/jira/browse/FLINK-7062
> Project: Flink
> Issue Type: Sub-task
> Components: CEP, Table API & SQL
> Reporter: Dian Fu
> Assignee: Dian Fu
> Priority: Major
> Labels: pull-request-available
>
> In this JIRA, we will support the basic functionality of {{MATCH_RECOGNIZE}}
> in Flink SQL API which includes the support of syntax {{MEASURES}},
> {{PATTERN}} and {{DEFINE}}. This would allow users write basic cep use cases
> with SQL like the following example:
> {code}
> SELECT T.aid, T.bid, T.cid
> FROM MyTable
> MATCH_RECOGNIZE (
> MEASURES
> A.id AS aid,
> B.id AS bid,
> C.id AS cid
> PATTERN (A B C)
> DEFINE
> A AS A.name = 'a',
> B AS B.name = 'b',
> C AS C.name = 'c'
> ) AS T
> {code}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)