simhadri-g commented on code in PR #4431:
URL: https://github.com/apache/hive/pull/4431#discussion_r1239746172
##########
iceberg/iceberg-handler/src/main/java/org/apache/iceberg/mr/hive/HiveIcebergStorageHandler.java:
##########
@@ -493,18 +503,42 @@ private String getStatsSource() {
}
private Path getStatsPath(Table table) {
- return new Path(table.location() + STATS + table.name() +
table.currentSnapshot().snapshotId());
+ return getStatsPath(table, table.currentSnapshot().snapshotId());
}
- private void invalidateStats(Path statsPath) {
+ private Path getStatsPath(Table table, long snapshotId) {
+ return new Path(table.location() + STATS + table.name() + snapshotId);
+ }
+
+ private boolean checkAndInvalidateStats(Table tbl) {
+ Path statsPath = getStatsPath(tbl);
try {
FileSystem fs = statsPath.getFileSystem(conf);
if (fs.exists(statsPath)) {
- fs.delete(statsPath, true);
+ // Analyze table and stats updater thread
+ return fs.delete(statsPath, true);
}
} catch (IOException e) {
LOG.error("Failed to invalidate stale column stats: {}", e.getMessage());
}
+ return false;
+ }
+
+ private void checkAndMergeStats(ColumnStatistics statsObjNew, Table tbl,
+ org.apache.hadoop.hive.ql.metadata.Table hmsTable, ColumnStatsDesc
columnStatsDesc) {
+ Long previousSnapshotId = tbl.currentSnapshot().parentId();
+ if (previousSnapshotId != null && canProvideColStatistics(tbl,
previousSnapshotId)) {
+ ColumnStatistics statsObjOld = readColStats(tbl, getStatsPath(tbl,
previousSnapshotId));
+ if (statsObjOld != null && statsObjOld.getStatsObjSize() != 0 &&
!statsObjNew.getStatsObj().isEmpty()) {
+ try {
+ MetaStoreServerUtils.mergeColStats(statsObjNew, statsObjOld);
+ StatsSetupConst.setColumnStatsState(hmsTable.getParameters(),
columnStatsDesc.getColName());
+ } catch (InvalidObjectException e) {
+ StatsSetupConst.removeColumnStatsState(hmsTable.getParameters(),
columnStatsDesc.getColName());
Review Comment:
If there is an exception, merging stats won't complete as a result the
stats would be inaccurate.
--
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]
