deniskuzZ commented on code in PR #4785: URL: https://github.com/apache/hive/pull/4785#discussion_r1377240091
########## iceberg/iceberg-handler/src/main/java/org/apache/iceberg/mr/hive/HiveIcebergMetaHook.java: ########## @@ -997,6 +1011,62 @@ public void postGetTable(org.apache.hadoop.hive.metastore.api.Table hmsTable) { } } + @Override + public void preDropPartition(org.apache.hadoop.hive.metastore.api.Table hmsTable, + EnvironmentContext context, + List<org.apache.commons.lang3.tuple.Pair<Integer, byte[]>> partExprs) + throws MetaException { + Table icebergTbl = IcebergTableUtil.getTable(conf, hmsTable); + DeleteFiles deleteFiles = icebergTbl.newDelete(); + List<Expression> expressions = partExprs.stream().map(partExpr -> { + ExprNodeDesc exprNodeDesc = SerializationUtilities + .deserializeObjectWithTypeInformation(partExpr.getRight(), true); + SearchArgument sarg = ConvertAstToSearchArg.create(conf, (ExprNodeGenericFuncDesc) exprNodeDesc); + return HiveIcebergFilterFactory.generateFilterExpression(sarg); + }).collect(Collectors.toList()); + PartitionsTable partitionsTable = (PartitionsTable) MetadataTableUtils + .createMetadataTableInstance(icebergTbl, MetadataTableType.PARTITIONS); + List<PartitionData> partitionList = Lists.newArrayList(); + try (CloseableIterable<FileScanTask> fileScanTasks = partitionsTable.newScan().planFiles()) { + fileScanTasks.forEach(task -> { + partitionList.addAll(Sets.newHashSet(CloseableIterable.transform(task.asDataTask().rows(), row -> { + StructProjection data = row.get(0, StructProjection.class); + PartitionSpec pSpec = icebergTbl.spec(); + PartitionData partitionData = new PartitionData(pSpec.partitionType()); + for (int index = 0; index < pSpec.fields().size(); index++) { + partitionData.set(index, data.get(index, Object.class)); + } + return partitionData; + }))); + }); + + Set<PartitionData> refinedPartList = Sets.newHashSet(); + for (Expression expr : expressions) { + refinedPartList.addAll(partitionList.stream().filter(partitionData -> { + ResidualEvaluator resEval = ResidualEvaluator.of(icebergTbl.spec(), expr, false); + return resEval.residualFor(partitionData).isEquivalentTo(Expressions.alwaysTrue()); + }).collect(Collectors.toList())); + } + + Expression partitionSetFilter = Expressions.alwaysFalse(); + for (PartitionData partitionData : refinedPartList) { + Expression partFilter = Expressions.alwaysTrue(); + for (int i = 0; i < icebergTbl.spec().fields().size(); i += 1) { + PartitionField field = icebergTbl.spec().fields().get(i); + partFilter = Expressions.and( + partFilter, Expressions.equal(field.name(), partitionData.get(i, Object.class))); + } + partitionSetFilter = Expressions.or(partitionSetFilter, partFilter); + } + + deleteFiles.deleteFromRowFilter(partitionSetFilter); + deleteFiles.commit(); + } catch (IOException e) { + throw new MetaException(String.format("Error while fetching the partitions due to: %s", e)); + } + context.putToProperties("dropPartitionSkip", "true"); Review Comment: should we introduce a constant for this? -- 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: gitbox-unsubscr...@hive.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: gitbox-unsubscr...@hive.apache.org For additional commands, e-mail: gitbox-h...@hive.apache.org