deniskuzZ commented on code in PR #4431:
URL: https://github.com/apache/hive/pull/4431#discussion_r1238432254
##########
iceberg/iceberg-handler/src/main/java/org/apache/iceberg/mr/hive/HiveIcebergStorageHandler.java:
##########
@@ -457,17 +462,20 @@ public List<ColumnStatisticsObj>
getColStatistics(org.apache.hadoop.hive.ql.meta
Table table = IcebergTableUtil.getTable(conf, hmsTable.getTTable());
String statsPath = getStatsPath(table).toString();
LOG.info("Using stats from puffin file at: {}", statsPath);
+ return readColStats(table, statsPath).equals(null) ? null :
readColStats(table, statsPath).getStatsObj();
+ }
+
+ private ColumnStatistics readColStats(Table table, String statsPath) {
try (PuffinReader reader =
Puffin.read(table.io().newInputFile(statsPath)).build()) {
List<BlobMetadata> blobMetadata = reader.fileMetadata().blobs();
- Map<BlobMetadata, List<ColumnStatistics>> collect =
-
Streams.stream(reader.readAll(blobMetadata)).collect(Collectors.toMap(Pair::first,
- blobMetadataByteBufferPair -> SerializationUtils.deserialize(
-
ByteBuffers.toByteArray(blobMetadataByteBufferPair.second()))));
- return collect.get(blobMetadata.get(0)).get(0).getStatsObj();
+ Map<BlobMetadata, List<ColumnStatistics>> collect =
Streams.stream(reader.readAll(blobMetadata)).collect(
+ Collectors.toMap(Pair::first, blobMetadataByteBufferPair ->
SerializationUtils.deserialize(
+ ByteBuffers.toByteArray(blobMetadataByteBufferPair.second()))));
+ return collect.get(blobMetadata.get(0)).get(0);
Review Comment:
should we have a safe check for position indexes? what if the stats file
would be corrupted or empty
--
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]
