[
https://issues.apache.org/jira/browse/PHOENIX-7015?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17814614#comment-17814614
]
ASF GitHub Bot commented on PHOENIX-7015:
-----------------------------------------
haridsv commented on code in PR #1813:
URL: https://github.com/apache/phoenix/pull/1813#discussion_r1479250015
##########
phoenix-core/src/main/java/org/apache/phoenix/util/CDCUtil.java:
##########
@@ -110,4 +109,25 @@ public static Scan initForRawScan(Scan scan) {
}
return scan;
}
+
+ public static int compareCellFamilyAndQualifier(byte[] columnFamily1,
+ byte[] columnQual1,
+ byte[] columnFamily2,
+ byte[] columnQual2) {
+ int familyNameComparison = CDCUtil.compare(columnFamily1,
columnFamily2);
+ if (familyNameComparison != 0) {
+ return familyNameComparison;
+ }
+ return CDCUtil.compare(columnQual1, columnQual2);
Review Comment:
Oh, I didn't know it was added in Java 9. There is also a
`org.apache.hadoop.io.FastByteComparisons#compareTo`? I found a copy of this
class in Phoenix that is named as `DescVarLengthFastByteComparisons` that is
already being used in several places. I think it is using some native
comparator so should be more efficient, besides reducing the code duplication.
There is also
[org.apache.avro.io.BinaryData#compareBytes](https://github.com/apache/avro/blob/d143d6262f7d6c688e695d1656f8609605835604/lang/java/avro/src/main/java/org/apache/avro/io/BinaryData.java#L183)
(requires anew dependency for Phoenix, but it is an existing dependency for
HBase)
> Extend UncoveredGlobalIndexRegionScanner for CDC region scanner usecase
> -----------------------------------------------------------------------
>
> Key: PHOENIX-7015
> URL: https://issues.apache.org/jira/browse/PHOENIX-7015
> Project: Phoenix
> Issue Type: Sub-task
> Reporter: Viraj Jasani
> Priority: Major
>
> For CDC region scanner usecase, extend UncoveredGlobalIndexRegionScanner to
> CDCUncoveredGlobalIndexRegionScanner. The new region scanner for CDC performs
> raw scan to index table and retrieve data table rows from index rows.
> Using the time range, it can form a JSON blob to represent changes to the row
> including pre and/or post row images.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
