gjacoby126 commented on a change in pull request #935:
URL: https://github.com/apache/phoenix/pull/935#discussion_r512847398
##########
File path:
phoenix-core/src/main/java/org/apache/phoenix/schema/MetaDataClient.java
##########
@@ -3048,7 +3048,13 @@ public boolean isViewReferenced() {
* the counter as NULL_COUNTER for extra safety.
*/
EncodedCQCounter cqCounterToBe = tableType == PTableType.VIEW ?
NULL_COUNTER : cqCounter;
- PTable table = new PTableImpl.Builder()
+ PTable table;
+ //better to use the table sent back from the server so we get an
accurate DDL
+ // timestamp, which is server-generated.
+ if (result.getTable() != null ) {
Review comment:
It seems strange that System.Catalog is the source of truth, we've just
been there on the server, and yet the client copy controls. Will look closer at
MetadataClient as you suggest to understand why.
Are we missing tests? Because assuming the client-side copy _should_
control, my replacing it with the server-side copy should have broken tests
somewhere, and it doesn't appear to.
----------------------------------------------------------------
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:
us...@infra.apache.org
