[
https://issues.apache.org/jira/browse/FLINK-6442?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16149979#comment-16149979
]
ASF GitHub Bot commented on FLINK-6442:
---------------------------------------
Github user lincoln-lil commented on a diff in the pull request:
https://github.com/apache/flink/pull/3829#discussion_r136491885
--- Diff:
flink-libraries/flink-table/src/main/scala/org/apache/flink/table/api/TableEnvironment.scala
---
@@ -502,26 +513,140 @@ abstract class TableEnvironment(val config:
TableConfig) {
* tEnv.sql(s"SELECT * FROM $table")
* }}}
*
- * @param query The SQL query to evaluate.
+ * @param sql The SQL string to evaluate.
* @return The result of the query as Table.
*/
- def sql(query: String): Table = {
+ @deprecated
+ def sql(sql: String): Table = {
val planner = new FlinkPlannerImpl(getFrameworkConfig, getPlanner,
getTypeFactory)
// parse the sql query
- val parsed = planner.parse(query)
+ val parsed = planner.parse(sql)
// validate the sql query
val validated = planner.validate(parsed)
// transform to a relational tree
val relational = planner.rel(validated)
-
new Table(this, LogicalRelNode(relational.rel))
}
/**
+ * Evaluates a SQL Select query on registered tables and retrieves the
result as a
+ * [[Table]].
+ *
+ * All tables referenced by the query must be registered in the
TableEnvironment. But
+ * [[Table.toString]] will automatically register an unique table name
and return the
+ * table name. So it allows to call SQL directly on tables like this:
+ *
+ * {{{
+ * val table: Table = ...
+ * // the table is not registered to the table environment
+ * tEnv.sqlSelect(s"SELECT * FROM $table")
+ * }}}
+ *
+ * @param sql The SQL string to evaluate.
+ * @return The result of the query as Table or null of the DML insert
operation.
+ */
+ def sqlQuery(sql: String): Table = {
+ val planner = new FlinkPlannerImpl(getFrameworkConfig, getPlanner,
getTypeFactory)
+ // parse the sql query
+ val parsed = planner.parse(sql)
+ if (null != parsed && parsed.getKind.belongsTo(SqlKind.QUERY)) {
+ // validate the sql query
+ val validated = planner.validate(parsed)
+ // transform to a relational tree
+ val relational = planner.rel(validated)
+ new Table(this, LogicalRelNode(relational.rel))
+ } else {
+ throw new TableException(
+ "Unsupported sql query! sqlQuery Only accept SELECT, UNION,
INTERSECT, EXCEPT, VALUES, " +
+ "WITH, ORDER_BY, EXPLICIT_TABLE")
+ }
+ }
+
+ /**
+ * Evaluates a SQL statement which must be an SQL Data Manipulation
Language (DML) statement,
+ * such as INSERT, UPDATE or DELETE; or an SQL statement that returns
nothing, such as a DDL
+ * statement;
+ * Currently only support a SQL INSERT statement on registered tables
and has no return value.
+ *
+ * All tables referenced by the query must be registered in the
TableEnvironment. But
+ * [[Table.toString]] will automatically register an unique table name
and return the
+ * table name. So it allows to call SQL directly on tables like this:
+ *
+ * {{{
+ * /// register table sink for insertion
+ * tEnv.registerTableSink("target_table", ...
+ * val sourceTable: Table = ...
+ * // sourceTable is not registered to the table environment
+ * tEnv.sqlInsert(s"INSERT INTO target_table SELECT * FROM
$sourceTable")
+ * }}}
+ *
+ * @param sql The SQL String to evaluate.
+ */
+ def sqlUpdate(sql: String): Unit = {
+ sqlUpdate(sql, QueryConfig.getQueryConfigFromTableEnv(this))
+ }
+
+ /**
+ * Evaluates a SQL statement which must be an SQL Data Manipulation
Language (DML) statement,
+ * such as INSERT, UPDATE or DELETE; or an SQL statement that returns
nothing, such as a DDL
+ * statement;
+ * Currently only support a SQL INSERT statement on registered tables
and has no return value.
+ *
+ * All tables referenced by the query must be registered in the
TableEnvironment. But
+ * [[Table.toString]] will automatically register an unique table name
and return the
+ * table name. So it allows to call SQL directly on tables like this:
+ *
+ * {{{
+ * /// register table sink for insertion
+ * tEnv.registerTableSink("target_table", ...
+ * val sourceTable: Table = ...
+ * // sourceTable is not registered to the table environment
+ * tEnv.sqlInsert(s"INSERT INTO target_table SELECT * FROM
$sourceTable")
+ * }}}
+ *
+ * @param sql The SQL String to evaluate.
+ * @param config The [[QueryConfig]] to use.
+ */
+ def sqlUpdate(sql: String, config: QueryConfig): Unit = {
+ val planner = new FlinkPlannerImpl(getFrameworkConfig, getPlanner,
getTypeFactory)
+ // parse the sql query
+ val parsed = planner.parse(sql)
+ parsed match {
+ case insert: SqlInsert => {
+ // validate the sql query
+ planner.validate(parsed)
+
+ // validate sink table
+ val targetName =
insert.getTargetTable.asInstanceOf[SqlIdentifier].names.get(0)
+ val targetTable = getTable(targetName)
+ if (null == targetTable ||
!targetTable.isInstanceOf[TableSinkTable[_]]) {
+ throw new TableException("SQL INSERT operation need a registered
TableSink Table!")
+ }
+ // validate unsupported partial insertion to sink table
+ val sinkTable = targetTable.asInstanceOf[TableSinkTable[_]]
+ if (null != insert.getTargetColumnList &&
insert.getTargetColumnList.size() !=
--- End diff --
The fields must be in the same order and Calcite will not reorder fields
based on their name.
Current `insert into` implementation equivalent to such sql syntax:
```
INSERT INTO table2
SELECT * FROM table1 ... -- here * represents all columns declared by table2
```
Another `insert into` is partial insert:
```
INSERT INTO table2 (column1, column2, column3, ...)
SELECT column1, column2, column3, ...
FROM table1 ...
```
column headers in the select clause are not used by an insert statement to
match columns up.
> Extend TableAPI Support Sink Table Registration and ‘insert into’ Clause in
> SQL
> -------------------------------------------------------------------------------
>
> Key: FLINK-6442
> URL: https://issues.apache.org/jira/browse/FLINK-6442
> Project: Flink
> Issue Type: New Feature
> Components: Table API & SQL
> Reporter: lincoln.lee
> Assignee: lincoln.lee
> Priority: Minor
>
> Currently in TableAPI there’s only registration method for source table,
> when we use SQL writing a streaming job, we should add additional part for
> the sink, like TableAPI does:
> {code}
> val sqlQuery = "SELECT * FROM MyTable WHERE _1 = 3"
> val t = StreamTestData.getSmall3TupleDataStream(env)
> tEnv.registerDataStream("MyTable", t)
> // one way: invoke tableAPI’s writeToSink method directly
> val result = tEnv.sql(sqlQuery)
> result.writeToSink(new YourStreamSink)
> // another way: convert to datastream first and then invoke addSink
> val result = tEnv.sql(sqlQuery).toDataStream[Row]
> result.addSink(new StreamITCase.StringSink)
> {code}
> From the api we can see the sink table always be a derived table because its
> 'schema' is inferred from the result type of upstream query.
> Compare to traditional RDBMS which support DML syntax, a query with a target
> output could be written like this:
> {code}
> insert into table target_table_name
> [(column_name [ ,...n ])]
> query
> {code}
> The equivalent form of the example above is as follows:
> {code}
> tEnv.registerTableSink("targetTable", new YourSink)
> val sql = "INSERT INTO targetTable SELECT a, b, c FROM sourceTable"
> val result = tEnv.sql(sql)
> {code}
> It is supported by Calcite’s grammar:
> {code}
> insert:( INSERT | UPSERT ) INTO tablePrimary
> [ '(' column [, column ]* ')' ]
> query
> {code}
> I'd like to extend Flink TableAPI to support such feature. see design doc:
> https://goo.gl/n3phK5
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
