fhan688 commented on code in PR #3541: URL: https://github.com/apache/fluss/pull/3541#discussion_r3490815452
########## fluss-lake/fluss-lake-hudi/src/main/java/org/apache/fluss/lake/hudi/source/HudiSortedRecordReader.java: ########## @@ -0,0 +1,433 @@ +/* + * 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.fluss.lake.hudi.source; + +import org.apache.fluss.config.Configuration; +import org.apache.fluss.lake.hudi.utils.HudiTableInfo; +import org.apache.fluss.lake.source.RecordReader; +import org.apache.fluss.lake.source.SortedRecordReader; +import org.apache.fluss.metadata.TablePath; +import org.apache.fluss.record.GenericRecord; +import org.apache.fluss.record.LogRecord; +import org.apache.fluss.row.BinaryString; +import org.apache.fluss.row.Decimal; +import org.apache.fluss.row.GenericRow; +import org.apache.fluss.row.InternalRow; +import org.apache.fluss.types.DataTypeRoot; +import org.apache.fluss.utils.CloseableIterator; +import org.apache.fluss.utils.InternalRowUtils; + +import org.apache.flink.table.api.Schema; +import org.apache.flink.table.types.AbstractDataType; +import org.apache.flink.table.types.DataType; +import org.apache.flink.table.types.logical.DecimalType; +import org.apache.flink.table.types.logical.LocalZonedTimestampType; +import org.apache.flink.table.types.logical.LogicalType; +import org.apache.flink.table.types.logical.TimestampType; +import org.apache.hudi.configuration.FlinkOptions; +import org.apache.hudi.source.ExpressionPredicates; + +import javax.annotation.Nullable; + +import java.io.IOException; +import java.util.ArrayList; +import java.util.Arrays; +import java.util.Collections; +import java.util.Comparator; +import java.util.HashMap; +import java.util.List; +import java.util.Map; +import java.util.stream.Collectors; + +import static org.apache.fluss.lake.hudi.HudiLakeCatalog.SYSTEM_COLUMNS; + +/** Sorted Hudi record reader for primary key table union read. */ +public class HudiSortedRecordReader implements SortedRecordReader { + + private static final String DELIMITER = ","; + + private final @Nullable RecordReader delegate; + private final SortOrder sortOrder; + + public HudiSortedRecordReader( + Configuration hudiConfig, + TablePath tablePath, + @Nullable HudiSplit hudiSplit, + @Nullable int[][] project, + List<ExpressionPredicates.Predicate> predicates) + throws Exception { + this.delegate = + hudiSplit == null + ? null + : new HudiRecordReader( + hudiConfig, tablePath, hudiSplit, project, predicates); + this.sortOrder = createSortOrder(hudiConfig, tablePath, project); + } + + @Override + public CloseableIterator<LogRecord> read() throws IOException { + if (delegate == null) { + return CloseableIterator.wrap(Collections.emptyIterator()); + } + CloseableIterator<LogRecord> iterator = delegate.read(); + List<LogRecord> records = new ArrayList<>(); + try { + while (iterator.hasNext()) { + records.add(copyRecord(iterator.next(), sortOrder.producedTypes)); + } + } finally { + iterator.close(); + } + records.sort( + (record1, record2) -> + compareRows( + record1.getRow(), + record2.getRow(), + sortOrder.keyPositionsInRecord, + sortOrder.keyTypes)); + return CloseableIterator.wrap(records.iterator()); + } + + @Override + public Comparator<InternalRow> order() { + return (row1, row2) -> + compareRows(row1, row2, sortOrder.keyPositionsInKey, sortOrder.keyTypes); + } + + static boolean canSortByRecordKey(HudiTableInfo hudiTableInfo, @Nullable int[][] project) { + RecordKeyInfo recordKeyInfo = resolveRecordKeyInfo(hudiTableInfo); + if (project == null) { + return true; + } + for (int originalPosition : recordKeyInfo.keyOriginalPositions) { + if (findProducedPosition(originalPosition, project) < 0) { + return false; + } + } + return true; + } + + private static SortOrder createSortOrder( + Configuration hudiConfig, TablePath tablePath, @Nullable int[][] project) + throws IOException { + try (HudiTableInfo hudiTableInfo = HudiTableInfo.create(tablePath, hudiConfig)) { + return resolveSortOrder(hudiTableInfo, project); + } + } + + private static SortOrder resolveSortOrder( + HudiTableInfo hudiTableInfo, @Nullable int[][] project) { + RecordKeyInfo recordKeyInfo = resolveRecordKeyInfo(hudiTableInfo); + int[] keyPositions = new int[recordKeyInfo.keyOriginalPositions.length]; + for (int i = 0; i < recordKeyInfo.keyOriginalPositions.length; i++) { + int keyPosition = findProducedPosition(recordKeyInfo.keyOriginalPositions[i], project); + if (keyPosition < 0) { + throw new IllegalArgumentException( + "Can not find Hudi record key field in projected fields."); + } + keyPositions[i] = keyPosition; + } + return new SortOrder( + recordKeyInfo.keyTypes, + keyPositions, + producedTypes(recordKeyInfo.userDataTypes, project)); + } + + private static RecordKeyInfo resolveRecordKeyInfo(HudiTableInfo hudiTableInfo) { + String recordKeyFields = + hudiTableInfo.getTableOptions().get(FlinkOptions.RECORD_KEY_FIELD.key()); + if (recordKeyFields == null || recordKeyFields.trim().isEmpty()) { + throw new IllegalArgumentException( + "Hudi record key field is required for sorted primary key table reader."); + } + + Map<String, DataType> dataTypesByName = new HashMap<>(); + Map<String, Integer> positionsByName = new HashMap<>(); + List<DataType> userDataTypes = new ArrayList<>(); + int position = 0; + for (Schema.UnresolvedColumn column : + hudiTableInfo.getHudiTable().getUnresolvedSchema().getColumns()) { + if (SYSTEM_COLUMNS.containsKey(column.getName())) { + continue; + } + DataType dataType = getDataType(column); + dataTypesByName.put(column.getName(), dataType); + positionsByName.put(column.getName(), position++); + userDataTypes.add(dataType); + } + + List<String> keyFields = + Arrays.stream(recordKeyFields.split(DELIMITER)) + .map(String::trim) + .filter(key -> !key.isEmpty()) + .collect(Collectors.toList()); + + List<DataType> keyTypes = new ArrayList<>(keyFields.size()); + int[] keyOriginalPositions = new int[keyFields.size()]; + for (int i = 0; i < keyFields.size(); i++) { + String key = keyFields.get(i); + DataType dataType = dataTypesByName.get(key); + Integer originalPosition = positionsByName.get(key); + if (dataType == null || originalPosition == null) { + throw new IllegalArgumentException( + "Can not find Hudi record key field " + key + "."); + } + keyTypes.add(dataType); + keyOriginalPositions[i] = originalPosition; + } + return new RecordKeyInfo(userDataTypes, keyTypes, keyOriginalPositions); + } + + private static DataType getDataType(Schema.UnresolvedColumn column) { + AbstractDataType<?> dataType; + if (column instanceof Schema.UnresolvedPhysicalColumn) { + dataType = ((Schema.UnresolvedPhysicalColumn) column).getDataType(); + } else if (column instanceof Schema.UnresolvedMetadataColumn) { + dataType = ((Schema.UnresolvedMetadataColumn) column).getDataType(); + } else { + throw new IllegalStateException("Unexpected column kind: " + column.getClass()); + } + if (dataType instanceof DataType) { + return (DataType) dataType; + } + throw new IllegalArgumentException("Unsupported Hudi column data type: " + dataType); + } + + private static int findProducedPosition(int originalPosition, @Nullable int[][] project) { + if (project == null) { + return originalPosition; + } + for (int i = 0; i < project.length; i++) { + if (project[i].length > 0 && project[i][0] == originalPosition) { + return i; + } + } + return -1; + } + + private static List<DataType> producedTypes( + List<DataType> userDataTypes, @Nullable int[][] project) { + if (project == null) { + return userDataTypes; + } + List<DataType> producedTypes = new ArrayList<>(project.length); + for (int[] projectPath : project) { + if (projectPath.length > 0) { + producedTypes.add(userDataTypes.get(projectPath[0])); + } + } + return producedTypes; + } + + private static LogRecord copyRecord(LogRecord record, List<DataType> producedTypes) { + return new GenericRecord( + record.logOffset(), + record.timestamp(), + record.getChangeType(), + copyRow(record.getRow(), producedTypes)); + } + + private static InternalRow copyRow(InternalRow row, List<DataType> fieldTypes) { + GenericRow copiedRow = new GenericRow(fieldTypes.size()); + for (int i = 0; i < fieldTypes.size(); i++) { + if (row.isNullAt(i)) { + copiedRow.setField(i, null); + } else { + copiedRow.setField(i, copyField(row, i, fieldTypes.get(i).getLogicalType())); + } + } + return copiedRow; + } + + private static Object copyField(InternalRow row, int pos, LogicalType logicalType) { + switch (logicalType.getTypeRoot()) { + case BOOLEAN: + return row.getBoolean(pos); + case TINYINT: + return row.getByte(pos); + case SMALLINT: + return row.getShort(pos); + case INTEGER: + case DATE: + case TIME_WITHOUT_TIME_ZONE: + return row.getInt(pos); + case BIGINT: + return row.getLong(pos); + case FLOAT: + return row.getFloat(pos); + case DOUBLE: + return row.getDouble(pos); + case CHAR: + case VARCHAR: + return row.getString(pos).copy(); + case DECIMAL: + DecimalType decimalType = (DecimalType) logicalType; + Decimal decimal = + row.getDecimal(pos, decimalType.getPrecision(), decimalType.getScale()); + return decimal.copy(); + case TIMESTAMP_WITHOUT_TIME_ZONE: + TimestampType timestampType = (TimestampType) logicalType; + return row.getTimestampNtz(pos, timestampType.getPrecision()); + case TIMESTAMP_WITH_LOCAL_TIME_ZONE: + LocalZonedTimestampType timestampLtzType = (LocalZonedTimestampType) logicalType; + return row.getTimestampLtz(pos, timestampLtzType.getPrecision()); + case BINARY: + case VARBINARY: + byte[] bytes = row.getBytes(pos); + return Arrays.copyOf(bytes, bytes.length); + default: + throw new IllegalArgumentException( + "Unsupported Hudi row field type: " + logicalType); + } + } + + private static int compareRows( + InternalRow row1, InternalRow row2, int[] keyPositions, List<DataType> keyTypes) { + for (int pos = 0; pos < keyTypes.size(); pos++) { + DataType keyType = keyTypes.get(pos); + int result = + compareValues( + getField(row1, keyPositions[pos], keyType), + getField(row2, keyPositions[pos], keyType), + keyType.getLogicalType()); + if (result != 0) { + return result; + } + } + return 0; + } + + private static Object getField(InternalRow row, int pos, DataType dataType) { + LogicalType logicalType = dataType.getLogicalType(); + switch (logicalType.getTypeRoot()) { + case BOOLEAN: + return row.getBoolean(pos); Review Comment: Fixed at the compareRows entry point. Since Hudi record keys map to Fluss primary key fields, null key values are invalid and now fail fast with a clear state check. -- 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]
