stoty commented on code in PR #1353:
URL: https://github.com/apache/phoenix/pull/1353#discussion_r843433977
##########
phoenix-core/src/main/java/org/apache/phoenix/schema/SaltingUtil.java:
##########
@@ -123,4 +130,29 @@ public static void
addRegionStartKeyToScanStartAndStopRows(byte[] startKey, byte
scan.setStopRow(newStopRow);
}
}
+
+ public static void checkTableIsSalted(Admin admin, TableName tableName,
int buckets) throws IOException, SQLException {
+ // Heuristics to check if and existing HBase table can be a salted
Phoenix table
+ // with the specified number of buckets.
+ // May give false negatives.
+ byte[] maxStartKey = new byte[] {0} ;
+ List<RegionInfo> regionInfos =
+ admin.getRegions(tableName);
+ for (RegionInfo regionInfo : regionInfos) {
+ if (Bytes.compareTo(regionInfo.getStartKey(), maxStartKey) > 0) {
+ maxStartKey = regionInfo.getStartKey();
+ }
+ }
+ byte lastRegionStartSaltKey = maxStartKey[0];
+ if (lastRegionStartSaltKey == (byte) (buckets - 1)) {
Review Comment:
Technically it is possible, but it would be a lot of work as we'd need to
start from the grammar to add a force option.
I've handled a few such cases, and in every case it was a user error.
I have given this some thought, and I cannot imagine such a scenario.
If the user has merged his last regions on his table that otherwise matches
the salt bucket count, and wants to create the table, it still wouldn't work
correctly, and they are better off fixing their splits first, and re-creating
the Phoenix table second.
--
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]
