skywalker0618 commented on code in PR #18636: URL: https://github.com/apache/hudi/pull/18636#discussion_r3196562416
########## hudi-flink-datasource/hudi-flink1.18.x/src/main/java/org/apache/hudi/table/format/cow/vector/reader/NestedPrimitiveColumnReader.java: ########## @@ -0,0 +1,639 @@ +/* + * 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.hudi.table.format.cow.vector.reader; + +import org.apache.hudi.table.format.cow.utils.IntArrayList; +import org.apache.hudi.table.format.cow.vector.ParquetDecimalVector; +import org.apache.hudi.table.format.cow.vector.position.LevelDelegation; + +import org.apache.flink.formats.parquet.vector.reader.ColumnReader; +import org.apache.flink.table.data.TimestampData; +import org.apache.flink.table.data.columnar.vector.heap.HeapBooleanVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapByteVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapBytesVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapDoubleVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapFloatVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapIntVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapLongVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapShortVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapTimestampVector; +import org.apache.flink.table.data.columnar.vector.writable.WritableColumnVector; +import org.apache.flink.table.types.logical.LogicalType; + +import org.apache.parquet.bytes.ByteBufferInputStream; +import org.apache.parquet.bytes.BytesInput; +import org.apache.parquet.bytes.BytesUtils; +import org.apache.parquet.column.ColumnDescriptor; +import org.apache.parquet.column.Encoding; +import org.apache.parquet.column.page.DataPage; +import org.apache.parquet.column.page.DataPageV1; +import org.apache.parquet.column.page.DataPageV2; +import org.apache.parquet.column.page.DictionaryPage; +import org.apache.parquet.column.page.PageReader; +import org.apache.parquet.column.values.ValuesReader; +import org.apache.parquet.column.values.rle.RunLengthBitPackingHybridDecoder; +import org.apache.parquet.io.ParquetDecodingException; +import org.apache.parquet.schema.PrimitiveType; +import org.apache.parquet.schema.Type; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import java.io.ByteArrayInputStream; +import java.io.IOException; +import java.util.ArrayList; +import java.util.List; + +import static org.apache.parquet.column.ValuesType.DEFINITION_LEVEL; +import static org.apache.parquet.column.ValuesType.REPETITION_LEVEL; +import static org.apache.parquet.column.ValuesType.VALUES; + +/** + * Reader to read a single primitive leaf column that participates in a nested (Dremel) structure. + * + * <p>Vendored from Apache Flink 2.1 (FLINK-35702, {@code + * org.apache.flink.formats.parquet.vector.reader.NestedPrimitiveColumnReader}). Only the package + * and the Hudi-local {@link ParquetDecimalVector} / {@link LevelDelegation} / {@link IntArrayList} + * imports are changed; the algorithm is untouched. The companion Hudi-specific {@code + * Int64TimestampColumnReader} / {@code FixedLenBytesColumnReader} behaviours stay at the leaf- + * reader creation boundary in {@code ParquetSplitReaderUtil} (lands in a follow-up PR), not inside + * this class — keeping it a faithful copy of upstream. + */ +public class NestedPrimitiveColumnReader implements ColumnReader<WritableColumnVector> { + private static final Logger LOG = LoggerFactory.getLogger(NestedPrimitiveColumnReader.class); + + private final IntArrayList repetitionLevelList = new IntArrayList(0); + private final IntArrayList definitionLevelList = new IntArrayList(0); + + private final PageReader pageReader; + private final ColumnDescriptor descriptor; + private final Type type; + private final LogicalType logicalType; + + /** The dictionary, if this column has dictionary encoding. */ + private final ParquetDataColumnReader dictionary; + + /** Maximum definition level for this column. */ + private final int maxDefLevel; + + private boolean isUtcTimestamp; + + /** Total number of values read. */ + private long valuesRead; + + /** + * value that indicates the end of the current page. That is, if valuesRead == + * endOfPageValueCount, we are at the end of the page. + */ + private long endOfPageValueCount; + + /** If true, the current page is dictionary encoded. */ + private boolean isCurrentPageDictionaryEncoded; + + private int definitionLevel; + private int repetitionLevel; + + /** Repetition/Definition/Value readers. */ + private IntIterator repetitionLevelColumn; + + private IntIterator definitionLevelColumn; + private ParquetDataColumnReader dataColumn; + + /** Total values in the current page. */ + private int pageValueCount; + + // flag to indicate if there is no data in parquet data page + private boolean eof = false; + + private boolean isFirstRow = true; + + private Object lastValue; + + public NestedPrimitiveColumnReader( + ColumnDescriptor descriptor, + PageReader pageReader, + boolean isUtcTimestamp, + Type parquetType, + LogicalType logicalType) + throws IOException { + this.descriptor = descriptor; + this.type = parquetType; + this.pageReader = pageReader; + this.maxDefLevel = descriptor.getMaxDefinitionLevel(); + this.isUtcTimestamp = isUtcTimestamp; + this.logicalType = logicalType; + + DictionaryPage dictionaryPage = pageReader.readDictionaryPage(); + if (dictionaryPage != null) { + try { + this.dictionary = + ParquetDataColumnReaderFactory.getDataColumnReaderByTypeOnDictionary( + parquetType.asPrimitiveType(), + dictionaryPage.getEncoding().initDictionary(descriptor, dictionaryPage), + isUtcTimestamp); + this.isCurrentPageDictionaryEncoded = true; + } catch (IOException e) { + throw new IOException( + String.format("Could not decode the dictionary for %s", descriptor), e); + } + } else { + this.dictionary = null; + this.isCurrentPageDictionaryEncoded = false; + } + } + + // This won't call, will actually call readAndNewVector Review Comment: Fixed ########## hudi-flink-datasource/hudi-flink1.18.x/src/main/java/org/apache/hudi/table/format/cow/vector/reader/NestedPrimitiveColumnReader.java: ########## @@ -0,0 +1,639 @@ +/* + * 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.hudi.table.format.cow.vector.reader; + +import org.apache.hudi.table.format.cow.utils.IntArrayList; +import org.apache.hudi.table.format.cow.vector.ParquetDecimalVector; +import org.apache.hudi.table.format.cow.vector.position.LevelDelegation; + +import org.apache.flink.formats.parquet.vector.reader.ColumnReader; +import org.apache.flink.table.data.TimestampData; +import org.apache.flink.table.data.columnar.vector.heap.HeapBooleanVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapByteVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapBytesVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapDoubleVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapFloatVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapIntVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapLongVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapShortVector; +import org.apache.flink.table.data.columnar.vector.heap.HeapTimestampVector; +import org.apache.flink.table.data.columnar.vector.writable.WritableColumnVector; +import org.apache.flink.table.types.logical.LogicalType; + +import org.apache.parquet.bytes.ByteBufferInputStream; +import org.apache.parquet.bytes.BytesInput; +import org.apache.parquet.bytes.BytesUtils; +import org.apache.parquet.column.ColumnDescriptor; +import org.apache.parquet.column.Encoding; +import org.apache.parquet.column.page.DataPage; +import org.apache.parquet.column.page.DataPageV1; +import org.apache.parquet.column.page.DataPageV2; +import org.apache.parquet.column.page.DictionaryPage; +import org.apache.parquet.column.page.PageReader; +import org.apache.parquet.column.values.ValuesReader; +import org.apache.parquet.column.values.rle.RunLengthBitPackingHybridDecoder; +import org.apache.parquet.io.ParquetDecodingException; +import org.apache.parquet.schema.PrimitiveType; +import org.apache.parquet.schema.Type; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import java.io.ByteArrayInputStream; +import java.io.IOException; +import java.util.ArrayList; +import java.util.List; + +import static org.apache.parquet.column.ValuesType.DEFINITION_LEVEL; +import static org.apache.parquet.column.ValuesType.REPETITION_LEVEL; +import static org.apache.parquet.column.ValuesType.VALUES; + +/** + * Reader to read a single primitive leaf column that participates in a nested (Dremel) structure. + * + * <p>Vendored from Apache Flink 2.1 (FLINK-35702, {@code + * org.apache.flink.formats.parquet.vector.reader.NestedPrimitiveColumnReader}). Only the package + * and the Hudi-local {@link ParquetDecimalVector} / {@link LevelDelegation} / {@link IntArrayList} + * imports are changed; the algorithm is untouched. The companion Hudi-specific {@code + * Int64TimestampColumnReader} / {@code FixedLenBytesColumnReader} behaviours stay at the leaf- + * reader creation boundary in {@code ParquetSplitReaderUtil} (lands in a follow-up PR), not inside + * this class — keeping it a faithful copy of upstream. + */ +public class NestedPrimitiveColumnReader implements ColumnReader<WritableColumnVector> { + private static final Logger LOG = LoggerFactory.getLogger(NestedPrimitiveColumnReader.class); + + private final IntArrayList repetitionLevelList = new IntArrayList(0); + private final IntArrayList definitionLevelList = new IntArrayList(0); + + private final PageReader pageReader; + private final ColumnDescriptor descriptor; + private final Type type; + private final LogicalType logicalType; + + /** The dictionary, if this column has dictionary encoding. */ + private final ParquetDataColumnReader dictionary; + + /** Maximum definition level for this column. */ + private final int maxDefLevel; + + private boolean isUtcTimestamp; + + /** Total number of values read. */ + private long valuesRead; + + /** + * value that indicates the end of the current page. That is, if valuesRead == + * endOfPageValueCount, we are at the end of the page. + */ + private long endOfPageValueCount; + + /** If true, the current page is dictionary encoded. */ + private boolean isCurrentPageDictionaryEncoded; + + private int definitionLevel; + private int repetitionLevel; + + /** Repetition/Definition/Value readers. */ + private IntIterator repetitionLevelColumn; + + private IntIterator definitionLevelColumn; + private ParquetDataColumnReader dataColumn; + + /** Total values in the current page. */ + private int pageValueCount; + + // flag to indicate if there is no data in parquet data page + private boolean eof = false; + + private boolean isFirstRow = true; + + private Object lastValue; + + public NestedPrimitiveColumnReader( + ColumnDescriptor descriptor, + PageReader pageReader, + boolean isUtcTimestamp, + Type parquetType, + LogicalType logicalType) + throws IOException { + this.descriptor = descriptor; + this.type = parquetType; + this.pageReader = pageReader; + this.maxDefLevel = descriptor.getMaxDefinitionLevel(); + this.isUtcTimestamp = isUtcTimestamp; + this.logicalType = logicalType; + + DictionaryPage dictionaryPage = pageReader.readDictionaryPage(); + if (dictionaryPage != null) { + try { + this.dictionary = + ParquetDataColumnReaderFactory.getDataColumnReaderByTypeOnDictionary( + parquetType.asPrimitiveType(), + dictionaryPage.getEncoding().initDictionary(descriptor, dictionaryPage), + isUtcTimestamp); + this.isCurrentPageDictionaryEncoded = true; + } catch (IOException e) { + throw new IOException( + String.format("Could not decode the dictionary for %s", descriptor), e); + } + } else { + this.dictionary = null; + this.isCurrentPageDictionaryEncoded = false; + } + } + + // This won't call, will actually call readAndNewVector + @Override + public void readToVector(int readNumber, WritableColumnVector vector) throws IOException { + throw new UnsupportedOperationException("This function should no be called."); Review Comment: Fixed -- 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]
