mike-tr-adamson commented on code in PR #2540: URL: https://github.com/apache/cassandra/pull/2540#discussion_r1290042672
########## src/java/org/apache/cassandra/index/sai/disk/v1/sortedterms/SortedTermsWriter.java: ########## @@ -49,9 +43,9 @@ * <p> Review Comment: We have two separate uses for the sorted terms. The partition keys version is only ever used for getting partition keys from row Ids. This doesn't require any order at all. The clustering keys version can look up row ids from clustering keys but only within a partition where the clustering keys are ordered. So, we have a partitioned case where keys must be ordered within a partition and a non-partitioned case where they don't need to be ordered at all. This is leading me to think that we have a case here that is similar to the `BlockPackedReader` and the `MonotonicBlockPackedReader` (not literally), where we may need to enforce different rules. In the case of the clustering keys, we should be forcing order within partitions, and the partition keys should not allow calls to the partitioned lookup at all. I'll get these changes in. I have no real objection to the use of 'term' here. This is a generic package that is storing terms. If we want to stop using term we'll need to do it across SAI. -- 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]
