[
https://issues.apache.org/jira/browse/APEXMALHAR-1818?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15278884#comment-15278884
]
Julian Hyde commented on APEXMALHAR-1818:
-----------------------------------------
In SQL#1, the algebra is not represented as JSON, but by a tree of Java objects
that are sub-classes of RelNode. It is produced by SqlToRelTranslator and is
operated on by query-transformation rules RelOptRule.
In #2, the schema is provided by implementing the SchemaFactory interface (a
"schema adapter"). Examples of this can be found in calcite/example/csv and
calcite/cassandra.
In testing, there are two tracks. The streaming TCK (see CALCITE-1114) can be
used to verify the runtime behavior of an engine that claims to be streaming
SQL compliant. It is not complete and will be fairly complex. I don't think we
should use it in phase 1.
The other track is simpler: a schema (as JSON) plus a SQL query in, and a DAG
(as JSON) out. Since the inputs and outputs are text, and we don't need a
connection to Kafka or Apex, this is a good basis for simple, fast unit tests.
I suggest that we hand-write maybe 5 SQL queries and their expected DAGs,
testing SELECT, WHERE, and perhaps OVER and GROUP BY.
> Integrate Calcite to support SQL
> --------------------------------
>
> Key: APEXMALHAR-1818
> URL: https://issues.apache.org/jira/browse/APEXMALHAR-1818
> Project: Apache Apex Malhar
> Issue Type: New Feature
> Components: query operators
> Reporter: Amol
> Assignee: Amol
> Labels: roadmap
>
> Once we have ability to generate a subdag, we should take a look at
> integrating Calcite into Apex. The operator that enables populate DAG, should
> use Calcite to generate the DAG, given a SQL query.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
