danny0405 commented on a change in pull request #9952:
[FLINK-14321][sql-parser] Support to parse watermark statement in SQL DDL
URL: https://github.com/apache/flink/pull/9952#discussion_r338362915
##########
File path:
flink-table/flink-sql-parser/src/main/java/org/apache/flink/sql/parser/ddl/SqlCreateTable.java
##########
@@ -310,9 +322,67 @@ private void printIndent(SqlWriter writer) {
public List<SqlNode> columnList = new ArrayList<>();
public SqlNodeList primaryKeyList = SqlNodeList.EMPTY;
public List<SqlNodeList> uniqueKeysList = new ArrayList<>();
+ @Nullable public SqlWatermark watermark;
}
public String[] fullTableName() {
return tableName.names.toArray(new String[0]);
}
+
+ //
-------------------------------------------------------------------------------------
+
+ private static final class ColumnValidator {
+
+ private final Set<String> allColumnNames = new HashSet<>();
+
Review comment:
The `ColumnValidator` makes me a little confused, i'm wondering do we need
the nested field replicates checking ? I mean during this phrase, what we can
really do is just a name checking, during the column expression conversion
(SqlNode to RexNode), we would finally have a validation though, do you think
this check is necessary ? A little redundant from my side.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
