Github user ijokarumawak commented on a diff in the pull request:
https://github.com/apache/nifi/pull/2162#discussion_r166296034
--- Diff:
nifi-nar-bundles/nifi-standard-bundle/nifi-standard-processors/src/main/java/org/apache/nifi/processors/standard/QueryDatabaseTable.java
---
@@ -150,8 +151,13 @@ public QueryDatabaseTable() {
final List<PropertyDescriptor> pds = new ArrayList<>();
pds.add(DBCP_SERVICE);
pds.add(DB_TYPE);
- pds.add(TABLE_NAME);
+ pds.add(new PropertyDescriptor.Builder()
+ .fromPropertyDescriptor(TABLE_NAME)
+ .description("The name of the database table to be
queried. When a custom query is used, this property is used to alias the query
and appears as an attribute on the FlowFile.")
+ .build());
--- End diff --
Please update `AbstractDatabaseFetchProcessor.onPropertyModified` so that
it clears `setupComplete` flag when this TABLE_NAME property is updated, too.
Without that, when a custom query is used, `columnTypeMap` will not be
populated again if the processor is reconfigured with different table name
alias and restarted, that prevent the maxValueColumn to be captured correctly,
and produce duplicated query result over and over.
---
