godfreyhe commented on code in PR #20084: URL: https://github.com/apache/flink/pull/20084#discussion_r934294704
########## flink-formats/flink-csv/src/main/java/org/apache/flink/formats/csv/util/CsvFormatStatisticsReportUtil.java: ########## @@ -0,0 +1,82 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.formats.csv.util; + +import org.apache.flink.core.fs.FileStatus; +import org.apache.flink.core.fs.FileSystem; +import org.apache.flink.core.fs.Path; +import org.apache.flink.table.plan.stats.TableStats; + +import java.io.BufferedReader; +import java.io.File; +import java.io.InputStreamReader; +import java.nio.charset.StandardCharsets; +import java.nio.file.Files; +import java.util.List; + +/** Utils for Csv format statistics report. */ +public class CsvFormatStatisticsReportUtil { + public static TableStats getTableStatistics(List<Path> files) { + // For Csv format, it's a heavy operation to obtain accurate statistics by scanning all + // files. So, We obtain the estimated statistics by sampling, the specific way is to + // sample the first 100 lines and calculate their row size, then compare row size with + // total file size to get the estimated row count. + final int totalSampleLineCnt = 100; + try { + long totalFileSize = 0; + int sampledRowCnt = 0; + long sampledRowSize = 0; + for (Path file : files) { + FileSystem fs = FileSystem.get(file.toUri()); + FileStatus status = fs.getFileStatus(file); + totalFileSize += status.getLen(); + + // sample the line size + if (sampledRowCnt < totalSampleLineCnt) { + try (InputStreamReader isr = + new InputStreamReader( + Files.newInputStream(new File(file.toUri()).toPath())); + BufferedReader br = new BufferedReader(isr)) { + String line; + while (sampledRowCnt < totalSampleLineCnt + && (line = br.readLine()) != null) { + sampledRowCnt += 1; + sampledRowSize += (line.getBytes(StandardCharsets.UTF_8).length + 1); + } + } + } + } + + // If line break is "\r\n", br.readLine() will ignore '\n' which make sampledRowSize + // smaller than totalFileSize. This will influence test result. + if (sampledRowCnt < totalSampleLineCnt) { + sampledRowSize = totalFileSize; + } + if (sampledRowSize == 0) { + return TableStats.UNKNOWN; + } + + int realSampledLineCnt = Math.min(totalSampleLineCnt, sampledRowCnt); + long estimatedRowCount = totalFileSize * realSampledLineCnt / sampledRowSize; + return new TableStats(estimatedRowCount); + } catch (Exception e) { + return TableStats.UNKNOWN; Review Comment: add some log for exception case ########## flink-connectors/flink-connector-hive/src/main/java/org/apache/flink/connectors/hive/HiveTableSource.java: ########## @@ -261,6 +273,99 @@ public DynamicTableSource copy() { return source; } + @Override + public TableStats reportStatistics() { + try { + // only support BOUNDED source + if (isStreamingSource()) { + return TableStats.UNKNOWN; + } + if (flinkConf.get(FileSystemConnectorOptions.SOURCE_REPORT_STATISTICS) + != FileSystemConnectorOptions.FileStatisticsType.ALL) { + return TableStats.UNKNOWN; + } + + HiveSourceBuilder sourceBuilder = + new HiveSourceBuilder(jobConf, flinkConf, tablePath, hiveVersion, catalogTable) + .setProjectedFields(projectedFields) + .setLimit(limit); + int threadNum = + flinkConf.get(HiveOptions.TABLE_EXEC_HIVE_LOAD_PARTITION_SPLITS_THREAD_NUM); + List<HiveTablePartition> hivePartitionsToRead = + getAllPartitions( + jobConf, + hiveVersion, + tablePath, + catalogTable.getPartitionKeys(), + remainingPartitions); + BulkFormat<RowData, HiveSourceSplit> defaultBulkFormat = + sourceBuilder.createDefaultBulkFormat(); + List<HiveSourceSplit> inputSplits = + HiveSourceFileEnumerator.createInputSplits( + 0, hivePartitionsToRead, threadNum, jobConf); + if (inputSplits.size() != 0) { + TableStats tableStats; + if (defaultBulkFormat instanceof FileBasedStatisticsReportableInputFormat) { + // If HiveInputFormat's variable useMapRedReader is false, Flink will use hadoop + // mapRed record to read data. + tableStats = + ((FileBasedStatisticsReportableInputFormat) defaultBulkFormat) + .reportStatistics( + inputSplits.stream() + .map(FileSourceSplit::path) + .collect(Collectors.toList()), + catalogTable.getSchema().toRowDataType()); + } else { + // If HiveInputFormat's variable useMapRedReader is true, Hive using MapRed + // InputFormat to read data. + tableStats = + getMapRedInputFormatStatistics( + inputSplits, catalogTable.getSchema().toRowDataType()); + } + if (limit == null) { + // If no limit push down, return recompute table stats. + return tableStats; + } else { + // If table have limit push down, return new table stats without table column + // stats. + long newRowCount = Math.min(limit, tableStats.getRowCount()); + return new TableStats(newRowCount); + } + } else { + return new TableStats(0); + } + + } catch (Exception e) { + return TableStats.UNKNOWN; + } + } + + private TableStats getMapRedInputFormatStatistics( + List<HiveSourceSplit> inputSplits, DataType producedDataType) { + // TODO now we assume that one hive external table has only one storage file format + String serializationLib = + inputSplits + .get(0) + .getHiveTablePartition() + .getStorageDescriptor() + .getSerdeInfo() + .getSerializationLib() + .toLowerCase(); + List<Path> files = + inputSplits.stream().map(FileSourceSplit::path).collect(Collectors.toList()); + // Now we only support Parquet, Orc formats. + if (serializationLib.contains("parquet")) { + return ParquetFormatStatisticsReportUtil.getTableStatistics( + files, producedDataType, jobConf, hiveVersion.startsWith("3")); + } else if (serializationLib.contains("orc")) { + return OrcFormatStatisticsReportUtil.getTableStatistics( + files, producedDataType, jobConf); + } else { + // Now, only support Orc and Parquet Formats. + return TableStats.UNKNOWN; Review Comment: add some log here ########## flink-connectors/flink-connector-hive/src/main/java/org/apache/flink/connectors/hive/HiveTableSource.java: ########## @@ -261,6 +273,99 @@ public DynamicTableSource copy() { return source; } + @Override + public TableStats reportStatistics() { + try { + // only support BOUNDED source + if (isStreamingSource()) { + return TableStats.UNKNOWN; + } + if (flinkConf.get(FileSystemConnectorOptions.SOURCE_REPORT_STATISTICS) + != FileSystemConnectorOptions.FileStatisticsType.ALL) { + return TableStats.UNKNOWN; + } + + HiveSourceBuilder sourceBuilder = + new HiveSourceBuilder(jobConf, flinkConf, tablePath, hiveVersion, catalogTable) + .setProjectedFields(projectedFields) + .setLimit(limit); + int threadNum = + flinkConf.get(HiveOptions.TABLE_EXEC_HIVE_LOAD_PARTITION_SPLITS_THREAD_NUM); + List<HiveTablePartition> hivePartitionsToRead = + getAllPartitions( + jobConf, + hiveVersion, + tablePath, + catalogTable.getPartitionKeys(), + remainingPartitions); + BulkFormat<RowData, HiveSourceSplit> defaultBulkFormat = + sourceBuilder.createDefaultBulkFormat(); + List<HiveSourceSplit> inputSplits = + HiveSourceFileEnumerator.createInputSplits( + 0, hivePartitionsToRead, threadNum, jobConf); + if (inputSplits.size() != 0) { + TableStats tableStats; + if (defaultBulkFormat instanceof FileBasedStatisticsReportableInputFormat) { + // If HiveInputFormat's variable useMapRedReader is false, Hive using Flink's + // InputFormat to read data. + tableStats = + ((FileBasedStatisticsReportableInputFormat) defaultBulkFormat) + .reportStatistics( + inputSplits.stream() + .map(FileSourceSplit::path) + .collect(Collectors.toList()), + catalogTable.getSchema().toRowDataType()); + } else { + // If HiveInputFormat's variable useMapRedReader is true, Hive using MapRed + // InputFormat to read data. + tableStats = + getMapRedInputFormatStatistics( + inputSplits, catalogTable.getSchema().toRowDataType()); + } + if (limit == null) { + // If no limit push down, return recompute table stats. + return tableStats; + } else { + // If table have limit push down, return new table stats without table column + // stats. + long newRowCount = Math.min(limit, tableStats.getRowCount()); + return new TableStats(newRowCount); + } + } else { + return TableStats.UNKNOWN; + } + + } catch (Exception e) { + return TableStats.UNKNOWN; Review Comment: +1 ########## flink-connectors/flink-connector-files/src/main/java/org/apache/flink/connector/file/table/FileSystemTableSource.java: ########## @@ -359,11 +359,19 @@ public TableStats reportStatistics() { splits.stream().map(FileSourceSplit::path).collect(Collectors.toList()); if (bulkReaderFormat instanceof FileBasedStatisticsReportableInputFormat) { - return ((FileBasedStatisticsReportableInputFormat) bulkReaderFormat) - .reportStatistics(files, producedDataType); + TableStats tableStats = + ((FileBasedStatisticsReportableInputFormat) bulkReaderFormat) + .reportStatistics(files, producedDataType); + return limit == null Review Comment: add some log for exception case ########## flink-connectors/flink-connector-hive/src/main/java/org/apache/flink/connectors/hive/HiveTableSource.java: ########## @@ -261,6 +273,99 @@ public DynamicTableSource copy() { return source; } + @Override + public TableStats reportStatistics() { + try { + // only support BOUNDED source + if (isStreamingSource()) { + return TableStats.UNKNOWN; + } + if (flinkConf.get(FileSystemConnectorOptions.SOURCE_REPORT_STATISTICS) + != FileSystemConnectorOptions.FileStatisticsType.ALL) { + return TableStats.UNKNOWN; + } + + HiveSourceBuilder sourceBuilder = + new HiveSourceBuilder(jobConf, flinkConf, tablePath, hiveVersion, catalogTable) + .setProjectedFields(projectedFields) + .setLimit(limit); + int threadNum = + flinkConf.get(HiveOptions.TABLE_EXEC_HIVE_LOAD_PARTITION_SPLITS_THREAD_NUM); + List<HiveTablePartition> hivePartitionsToRead = + getAllPartitions( + jobConf, + hiveVersion, + tablePath, + catalogTable.getPartitionKeys(), + remainingPartitions); + BulkFormat<RowData, HiveSourceSplit> defaultBulkFormat = + sourceBuilder.createDefaultBulkFormat(); + List<HiveSourceSplit> inputSplits = + HiveSourceFileEnumerator.createInputSplits( + 0, hivePartitionsToRead, threadNum, jobConf); + if (inputSplits.size() != 0) { + TableStats tableStats; + if (defaultBulkFormat instanceof FileBasedStatisticsReportableInputFormat) { + // If HiveInputFormat's variable useMapRedReader is false, Flink will use hadoop + // mapRed record to read data. + tableStats = + ((FileBasedStatisticsReportableInputFormat) defaultBulkFormat) + .reportStatistics( + inputSplits.stream() + .map(FileSourceSplit::path) + .collect(Collectors.toList()), + catalogTable.getSchema().toRowDataType()); + } else { + // If HiveInputFormat's variable useMapRedReader is true, Hive using MapRed + // InputFormat to read data. + tableStats = + getMapRedInputFormatStatistics( + inputSplits, catalogTable.getSchema().toRowDataType()); + } + if (limit == null) { Review Comment: Since we has built the HiveSourceBuilder with limit, is `limit` also needed here ? ########## flink-formats/flink-orc/src/main/java/org/apache/flink/orc/util/OrcFormatStatisticsReportUtil.java: ########## @@ -0,0 +1,214 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.orc.util; + +import org.apache.flink.core.fs.Path; +import org.apache.flink.table.plan.stats.ColumnStats; +import org.apache.flink.table.plan.stats.TableStats; +import org.apache.flink.table.types.DataType; +import org.apache.flink.table.types.logical.LogicalType; +import org.apache.flink.table.types.logical.RowType; + +import org.apache.hadoop.conf.Configuration; +import org.apache.orc.ColumnStatistics; +import org.apache.orc.DateColumnStatistics; +import org.apache.orc.DecimalColumnStatistics; +import org.apache.orc.DoubleColumnStatistics; +import org.apache.orc.IntegerColumnStatistics; +import org.apache.orc.OrcConf; +import org.apache.orc.OrcFile; +import org.apache.orc.Reader; +import org.apache.orc.StringColumnStatistics; +import org.apache.orc.TimestampColumnStatistics; +import org.apache.orc.TypeDescription; +import org.apache.orc.impl.ColumnStatisticsImpl; + +import java.io.IOException; +import java.sql.Date; +import java.util.HashMap; +import java.util.List; +import java.util.Map; + +/** Utils for Orc format statistics report. */ +public class OrcFormatStatisticsReportUtil { + public static TableStats getTableStatistics( + List<Path> files, DataType producedDataType, Configuration hadoopConfig) { + try { + long rowCount = 0; + Map<String, ColumnStatistics> columnStatisticsMap = new HashMap<>(); + RowType producedRowType = (RowType) producedDataType.getLogicalType(); + for (Path file : files) { + rowCount += + updateStatistics(hadoopConfig, file, columnStatisticsMap, producedRowType); + } + + Map<String, ColumnStats> columnStatsMap = + convertToColumnStats(rowCount, columnStatisticsMap, producedRowType); + + return new TableStats(rowCount, columnStatsMap); + } catch (Exception e) { + return TableStats.UNKNOWN; Review Comment: add some log for exception ########## flink-connectors/flink-connector-hive/src/main/java/org/apache/flink/connectors/hive/HiveTableSource.java: ########## @@ -261,6 +273,99 @@ public DynamicTableSource copy() { return source; } + @Override + public TableStats reportStatistics() { + try { + // only support BOUNDED source + if (isStreamingSource()) { + return TableStats.UNKNOWN; + } + if (flinkConf.get(FileSystemConnectorOptions.SOURCE_REPORT_STATISTICS) + != FileSystemConnectorOptions.FileStatisticsType.ALL) { + return TableStats.UNKNOWN; + } + + HiveSourceBuilder sourceBuilder = + new HiveSourceBuilder(jobConf, flinkConf, tablePath, hiveVersion, catalogTable) + .setProjectedFields(projectedFields) + .setLimit(limit); + int threadNum = + flinkConf.get(HiveOptions.TABLE_EXEC_HIVE_LOAD_PARTITION_SPLITS_THREAD_NUM); + List<HiveTablePartition> hivePartitionsToRead = + getAllPartitions( + jobConf, + hiveVersion, + tablePath, + catalogTable.getPartitionKeys(), + remainingPartitions); + BulkFormat<RowData, HiveSourceSplit> defaultBulkFormat = + sourceBuilder.createDefaultBulkFormat(); + List<HiveSourceSplit> inputSplits = + HiveSourceFileEnumerator.createInputSplits( + 0, hivePartitionsToRead, threadNum, jobConf); + if (inputSplits.size() != 0) { + TableStats tableStats; + if (defaultBulkFormat instanceof FileBasedStatisticsReportableInputFormat) { + // If HiveInputFormat's variable useMapRedReader is false, Flink will use hadoop + // mapRed record to read data. + tableStats = + ((FileBasedStatisticsReportableInputFormat) defaultBulkFormat) + .reportStatistics( + inputSplits.stream() + .map(FileSourceSplit::path) + .collect(Collectors.toList()), + catalogTable.getSchema().toRowDataType()); + } else { + // If HiveInputFormat's variable useMapRedReader is true, Hive using MapRed + // InputFormat to read data. + tableStats = + getMapRedInputFormatStatistics( + inputSplits, catalogTable.getSchema().toRowDataType()); + } + if (limit == null) { + // If no limit push down, return recompute table stats. + return tableStats; + } else { + // If table have limit push down, return new table stats without table column + // stats. + long newRowCount = Math.min(limit, tableStats.getRowCount()); + return new TableStats(newRowCount); + } + } else { + return new TableStats(0); + } + + } catch (Exception e) { + return TableStats.UNKNOWN; + } + } + + private TableStats getMapRedInputFormatStatistics( + List<HiveSourceSplit> inputSplits, DataType producedDataType) { + // TODO now we assume that one hive external table has only one storage file format + String serializationLib = + inputSplits + .get(0) + .getHiveTablePartition() + .getStorageDescriptor() + .getSerdeInfo() + .getSerializationLib() + .toLowerCase(); + List<Path> files = + inputSplits.stream().map(FileSourceSplit::path).collect(Collectors.toList()); + // Now we only support Parquet, Orc formats. + if (serializationLib.contains("parquet")) { Review Comment: case sensitive? ########## flink-connectors/flink-connector-hive/src/test/java/org/apache/flink/connectors/hive/HiveTableSourceStatisticsReportTest.java: ########## @@ -0,0 +1,393 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.flink.connectors.hive; + +import org.apache.flink.table.api.SqlDialect; +import org.apache.flink.table.catalog.hive.HiveCatalog; +import org.apache.flink.table.catalog.hive.HiveTestUtils; +import org.apache.flink.table.plan.stats.ColumnStats; +import org.apache.flink.table.plan.stats.TableStats; +import org.apache.flink.table.planner.plan.stats.FlinkStatistic; +import org.apache.flink.table.planner.utils.StatisticsReportTestBase; +import org.apache.flink.table.types.DataType; +import org.apache.flink.table.utils.DateTimeUtils; + +import org.junit.jupiter.api.AfterEach; +import org.junit.jupiter.api.BeforeEach; +import org.junit.jupiter.api.Test; +import org.junit.jupiter.api.io.TempDir; + +import java.io.File; +import java.math.BigDecimal; +import java.sql.Date; +import java.util.HashMap; +import java.util.List; +import java.util.Map; + +import static org.assertj.core.api.Assertions.assertThat; + +/** Test for statistics functionality in {@link HiveTableSource}. */ +public class HiveTableSourceStatisticsReportTest extends StatisticsReportTestBase { + + private static HiveCatalog hiveCatalog; + private static final String catalogName = "hive"; + private static final String dbName = "db1"; + private static final String sourceTable = "sourceTable"; + + @BeforeEach + public void setup(@TempDir File file) throws Exception { + super.setup(file); + hiveCatalog = HiveTestUtils.createHiveCatalog(); + hiveCatalog.open(); + + tEnv.getConfig().setSqlDialect(SqlDialect.HIVE); + tEnv.registerCatalog(catalogName, hiveCatalog); + tEnv.useCatalog(catalogName); + tEnv.executeSql(String.format("create database %s", dbName)); + + tEnv.executeSql( + String.format( + "create table %s.%s.%s ( %s )", + catalogName, + dbName, + sourceTable, + String.join(", ", ddlTypesMapToStringList(ddlTypesMap())))); + + DataType dataType = + tEnv.from(String.format("%s.%s.%s", catalogName, dbName, sourceTable)) + .getResolvedSchema() + .toPhysicalRowDataType(); + tEnv.fromValues(dataType, getData()) + .executeInsert(String.format("%s.%s.%s", catalogName, dbName, sourceTable)) + .await(); + } + + @AfterEach + public void after() { + super.after(); + if (null != hiveCatalog) { + hiveCatalog.close(); + } + } + + @Override + protected String[] properties() { + return new String[0]; + } + + @Test + public void testMapRedCsvFormatHiveTableSourceStatisticsReport() { + FlinkStatistic statistic = + getStatisticsFromOptimizedPlan( + String.format("select * from %s.%s.%s", catalogName, dbName, sourceTable)); + assertThat(statistic.getTableStats()).isEqualTo(TableStats.UNKNOWN); + } + + @Test + public void testFlinkOrcFormatHiveTableSourceStatisticsReport() throws Exception { + tEnv.executeSql( + String.format( + "create table hive.db1.orcTable ( %s ) stored as orc", + String.join(", ", ddlTypesMapToStringList(ddlTypesMap())))); + tEnv.executeSql( + String.format( + "insert into hive.db1.orcTable select * from %s.%s.%s", + catalogName, dbName, sourceTable)) + .await(); + + // Hive to read Orc file + FlinkStatistic statistic = + getStatisticsFromOptimizedPlan( + "select * from hive.db1.orcTable where f_smallint > 100"); + assertHiveTableOrcFormatTableStatsEquals(statistic.getTableStats(), 3, 1L); + } + + @Test + public void testFlinkParquetFormatHiveTableSourceStatisticsReport() throws Exception { + tEnv.executeSql( + String.format( + "create table hive.db1.parquetTable ( %s ) stored as parquet", + String.join(", ", ddlTypesMapToStringList(ddlTypesMap())))); + tEnv.executeSql( + String.format( + "insert into hive.db1.parquetTable select * from %s.%s.%s", + catalogName, dbName, sourceTable)) + .await(); + + // Hive to read Parquet file + FlinkStatistic statistic = + getStatisticsFromOptimizedPlan( + "select * from hive.db1.parquetTable where f_smallint > 100"); + assertHiveTableParquetFormatTableStatsEquals(statistic.getTableStats(), 3, 1L); + } + + @Test + public void testMapRedOrcFormatHiveTableSourceStatisticsReport() throws Exception { + // Use mapRed parquet format. + tEnv.getConfig().set(HiveOptions.TABLE_EXEC_HIVE_FALLBACK_MAPRED_READER, true); + tEnv.executeSql( + String.format( + "create table hive.db1.orcTable ( %s ) stored as orc", + String.join(", ", ddlTypesMapToStringList(ddlTypesMap())))); + tEnv.executeSql( + String.format( + "insert into hive.db1.orcTable select * from %s.%s.%s", + catalogName, dbName, sourceTable)) + .await(); + + // Hive to read Orc file + FlinkStatistic statistic = + getStatisticsFromOptimizedPlan( + "select * from hive.db1.orcTable where f_smallint > 100"); + assertHiveTableOrcFormatTableStatsEquals(statistic.getTableStats(), 3, 1L); + } + + @Test + public void testMapRedParquetFormatHiveTableSourceStatisticsReport() throws Exception { + // Use mapRed parquet format. + tEnv.getConfig().set(HiveOptions.TABLE_EXEC_HIVE_FALLBACK_MAPRED_READER, true); + tEnv.executeSql( + String.format( + "create table hive.db1.parquetTable ( %s ) stored as parquet", + String.join(", ", ddlTypesMapToStringList(ddlTypesMap())))); + tEnv.executeSql( + String.format( + "insert into hive.db1.parquetTable select * from %s.%s.%s", + catalogName, dbName, sourceTable)) + .await(); + + // Hive to read Parquet file. + FlinkStatistic statistic = + getStatisticsFromOptimizedPlan( + "select * from hive.db1.parquetTable where f_smallint > 100"); + assertHiveTableParquetFormatTableStatsEquals(statistic.getTableStats(), 3, 1L); + } + + @Test + public void testHiveTableSourceWithLimitPushDown() { Review Comment: add a case with limit push down and statistics is unknown -- 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]
