chrajeshbabu commented on code in PR #1437:
URL: https://github.com/apache/phoenix/pull/1437#discussion_r873246623
##########
phoenix-core/src/main/java/org/apache/phoenix/iterate/TableResultIterator.java:
##########
@@ -143,6 +143,20 @@ public TableResultIterator(MutationState mutationState,
Scan scan, ScanMetricsHo
ScanUtil.setScanAttributesForClient(scan, table,
plan.getContext().getConnection());
}
+ public TableResultIterator(MutationState mutationState, Scan scan,
ScanMetricsHolder scanMetricsHolder,
Review Comment:
@gjacoby126 @joshelser
Actually I thought of adding the special constructor to use in phoenix spark
connector to avoid fetching the metadata and avoid using the query plan but
it's difficult to avoid using the query plan because we need the row projector,
query context etc.. So I have found a simple solution of sharing the table meta
data across the worker and cache it so that we need not fetch the metadata
everytime in the workers and avoid the unnecessary load on RS holding the
system catalog table. Here is the PR created under phoenix-connectors
https://github.com/apache/phoenix-connectors/pull/80 with that changes in place
this special constructor not required so will remove these special constructor
changes and update the PR.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
