netudima commented on code in PR #4536:
URL: https://github.com/apache/cassandra/pull/4536#discussion_r2695279621
##########
src/java/org/apache/cassandra/db/memtable/Flushing.java:
##########
@@ -245,7 +245,8 @@ public static SSTableMultiWriter
createFlushWriter(ColumnFamilyStore cfs,
new SerializationHeader(true,
flushSet.metadata(),
flushSet.columns(),
-
flushSet.encodingStats()),
+
flushSet.encodingStats(),
+
flushSet.columnsChangedAfterCreation()),
Review Comment:
I worried about cases when we change a value column type in the middle of a
memtable population, so one row in it has one type and another has another one
and then we flush both rows to the same SSTable, especially we change from a
fixed length size to variable length size. We do no have a type info per row,
so it would not be possible to differentiate types in these two cases to be
able to read such values correctly for different rows without converting them
to the same type during a write time.
But it looks like it is not a problem anymore because we do not allow to
change types (https://issues.apache.org/jira/browse/CASSANDRA-12443) anymore as
well as remove and add a column with a different type is limited to
serialization compatible types only
(https://issues.apache.org/jira/browse/CASSANDRA-16905)...
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
