liaorui commented on code in PR #8096: URL: https://github.com/apache/inlong/pull/8096#discussion_r1206201379
########## inlong-sort/sort-flink/sort-flink-v1.13/sort-connectors/hive/src/main/java/org/apache/inlong/sort/hive/filesystem/HadoopPathBasedPartFileWriter.java: ########## @@ -0,0 +1,788 @@ +/* + * 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.inlong.sort.hive.filesystem; + +import static org.apache.inlong.sort.base.Constants.SINK_MULTIPLE_DATABASE_PATTERN; +import static org.apache.inlong.sort.base.Constants.SINK_MULTIPLE_ENABLE; +import static org.apache.inlong.sort.base.Constants.SINK_MULTIPLE_FORMAT; +import static org.apache.inlong.sort.base.Constants.SINK_MULTIPLE_TABLE_PATTERN; +import static org.apache.inlong.sort.hive.HiveOptions.HIVE_SCHEMA_SCAN_INTERVAL; + +import java.util.Arrays; +import java.util.HashMap; +import java.util.Iterator; +import java.util.List; +import java.util.Map; +import java.util.function.Function; +import javax.annotation.Nullable; +import org.apache.commons.lang3.tuple.ImmutablePair; +import org.apache.commons.lang3.tuple.ImmutableTriple; +import org.apache.commons.lang3.tuple.Pair; +import org.apache.commons.lang3.tuple.Triple; +import org.apache.flink.annotation.VisibleForTesting; +import org.apache.flink.connectors.hive.FlinkHiveException; +import org.apache.flink.core.io.SimpleVersionedSerializer; +import org.apache.flink.formats.hadoop.bulk.HadoopFileCommitter; +import org.apache.flink.formats.hadoop.bulk.HadoopFileCommitterFactory; +import org.apache.flink.formats.hadoop.bulk.HadoopPathBasedBulkWriter; +import org.apache.flink.shaded.jackson2.com.fasterxml.jackson.databind.JsonNode; +import org.apache.flink.streaming.api.functions.sink.filesystem.AbstractPartFileWriter; +import org.apache.flink.streaming.api.functions.sink.filesystem.BucketWriter; +import org.apache.flink.streaming.api.functions.sink.filesystem.InProgressFileWriter; +import org.apache.flink.streaming.api.functions.sink.filesystem.WriterProperties; + +import org.apache.flink.table.catalog.ObjectIdentifier; +import org.apache.flink.table.catalog.hive.client.HiveShim; +import org.apache.flink.table.data.GenericRowData; +import org.apache.flink.table.data.RowData; +import org.apache.flink.table.data.binary.BinaryRowData; +import org.apache.flink.table.types.logical.RowType; +import org.apache.flink.util.ExceptionUtils; +import org.apache.hadoop.conf.Configuration; + +import java.io.IOException; +import java.nio.ByteBuffer; +import java.nio.ByteOrder; +import java.nio.charset.Charset; +import java.nio.charset.StandardCharsets; +import org.apache.hadoop.fs.Path; +import org.apache.hadoop.hive.ql.exec.FileSinkOperator; +import org.apache.hadoop.hive.ql.exec.FileSinkOperator.RecordWriter; +import org.apache.hadoop.io.Writable; +import org.apache.hadoop.mapred.JobConf; +import org.apache.inlong.sort.base.dirty.DirtyOptions; +import org.apache.inlong.sort.base.dirty.DirtySinkHelper; +import org.apache.inlong.sort.base.dirty.DirtyType; +import org.apache.inlong.sort.base.dirty.sink.DirtySink; +import org.apache.inlong.sort.base.format.DynamicSchemaFormatFactory; +import org.apache.inlong.sort.base.format.JsonDynamicSchemaFormat; +import org.apache.inlong.sort.base.metric.sub.SinkTableMetricData; +import org.apache.inlong.sort.base.sink.PartitionPolicy; +import org.apache.inlong.sort.base.sink.SchemaUpdateExceptionPolicy; +import org.apache.inlong.sort.hive.HiveBulkWriterFactory; +import org.apache.inlong.sort.hive.util.CacheHolder; +import org.apache.inlong.sort.hive.util.HiveTableUtil; +import org.apache.inlong.sort.hive.HiveWriterFactory; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +/** + * The part-file writer that writes to the specified hadoop path. + */ +public class HadoopPathBasedPartFileWriter<IN, BucketID> extends AbstractPartFileWriter<IN, BucketID> { + + private static final Logger LOG = LoggerFactory.getLogger(HadoopPathBasedPartFileWriter.class); + + private final InLongHadoopPathBasedBulkWriter writer; + + private final HadoopFileCommitter fileCommitter; + + private BucketID bucketID; + private Path targetPath; + private Path inProgressPath; + private final HiveShim hiveShim; + private final String hiveVersion; + + private final boolean sinkMultipleEnable; + private final String sinkMultipleFormat; + + private final String databasePattern; + private final String tablePattern; + + private final SchemaUpdateExceptionPolicy schemaUpdatePolicy; + + private final PartitionPolicy partitionPolicy; + + private transient JsonDynamicSchemaFormat jsonFormat; + + @Nullable + private final transient SinkTableMetricData metricData; + + private final DirtyOptions dirtyOptions; + + private final @Nullable DirtySink<Object> dirtySink; + + public HadoopPathBasedPartFileWriter(final BucketID bucketID, + Path targetPath, + Path inProgressPath, + InLongHadoopPathBasedBulkWriter writer, + HadoopFileCommitter fileCommitter, + long createTime, + HiveShim hiveShim, + String hiveVersion, + boolean sinkMultipleEnable, + String sinkMultipleFormat, + String databasePattern, + String tablePattern, + @Nullable SinkTableMetricData metricData, + DirtyOptions dirtyOptions, + @Nullable DirtySink<Object> dirtySink, + SchemaUpdateExceptionPolicy schemaUpdatePolicy, + PartitionPolicy partitionPolicy) { + super(bucketID, createTime); + + this.bucketID = bucketID; + this.targetPath = targetPath; + this.inProgressPath = inProgressPath; + this.writer = writer; + this.fileCommitter = fileCommitter; + this.hiveShim = hiveShim; + this.hiveVersion = hiveVersion; + this.sinkMultipleEnable = sinkMultipleEnable; + this.sinkMultipleFormat = sinkMultipleFormat; + this.databasePattern = databasePattern; + this.tablePattern = tablePattern; + this.metricData = metricData; + this.schemaUpdatePolicy = schemaUpdatePolicy; + this.dirtyOptions = dirtyOptions; + this.dirtySink = dirtySink; + this.partitionPolicy = partitionPolicy; + } + + @Override + public void write(IN element, long currentTime) { + String databaseName = null; + String tableName = null; + int recordNum = 1; + int recordSize = 0; + JsonNode rootNode = null; + ObjectIdentifier identifier = null; + HashMap<ObjectIdentifier, Long> ignoreWritingTableMap = CacheHolder.getIgnoreWritingTableMap(); + try { + if (sinkMultipleEnable) { + GenericRowData rowData = (GenericRowData) element; + if (jsonFormat == null) { + jsonFormat = (JsonDynamicSchemaFormat) DynamicSchemaFormatFactory.getFormat(sinkMultipleFormat); + } + byte[] rawData = (byte[]) rowData.getField(0); + rootNode = jsonFormat.deserialize(rawData); + LOG.debug("root node: {}", rootNode); + boolean isDDL = jsonFormat.extractDDLFlag(rootNode); + if (isDDL) { + // Ignore ddl change for now + return; + } + databaseName = jsonFormat.parse(rootNode, databasePattern); + tableName = jsonFormat.parse(rootNode, tablePattern); + + List<Map<String, Object>> physicalDataList = HiveTableUtil.jsonNode2Map( + jsonFormat.getPhysicalData(rootNode)); + recordNum = physicalDataList.size(); + + identifier = HiveTableUtil.createObjectIdentifier(databaseName, tableName); + + // ignore writing data into this table + if (ignoreWritingTableMap.containsKey(identifier)) { + return; + } + + List<String> pkListStr = jsonFormat.extractPrimaryKeyNames(rootNode); + RowType schema = jsonFormat.extractSchema(rootNode, pkListStr); + HiveWriterFactory writerFactory = getHiveWriterFactory(identifier, schema, hiveVersion); + + // parse the real location of hive table + Path inProgressFilePath = getInProgressPath(writerFactory); + + LOG.debug("in progress file path: {}", inProgressFilePath); + Pair<RecordWriter, Function<RowData, Writable>> pair = getRecordWriterAndRowConverter(writerFactory, + inProgressFilePath); + // reset record writer and row converter + writer.setRecordWriter(pair.getLeft()); + writer.setRowConverter(pair.getRight()); + writer.setInProgressPath(inProgressFilePath); + + boolean replaceLineBreak = writerFactory.getStorageDescriptor().getInputFormat() + .contains("TextInputFormat"); + + for (Map<String, Object> record : physicalDataList) { + // check and alter hive table if schema has changed + boolean changed = checkSchema(identifier, writerFactory, schema); + if (changed) { + // remove cache and reload hive writer factory + CacheHolder.getFactoryMap().remove(identifier); + writerFactory = HiveTableUtil.getWriterFactory(hiveShim, hiveVersion, identifier); + assert writerFactory != null; + FileSinkOperator.RecordWriter recordWriter = writerFactory.createRecordWriter( + inProgressFilePath); + Function<RowData, Writable> rowConverter = writerFactory.createRowDataConverter(); + writer.setRecordWriter(recordWriter); + writer.setRowConverter(rowConverter); + CacheHolder.getRecordWriterHashMap().put(inProgressFilePath, recordWriter); + CacheHolder.getRowConverterHashMap().put(inProgressFilePath, rowConverter); + } + + LOG.debug("record: {}", record); + LOG.debug("columns : {}", Arrays.deepToString(writerFactory.getAllColumns())); + LOG.debug("types: {}", Arrays.deepToString(writerFactory.getAllTypes())); + Pair<GenericRowData, Long> rowDataPair = HiveTableUtil.getRowData(record, + writerFactory.getAllColumns(), writerFactory.getAllTypes(), replaceLineBreak); + GenericRowData genericRowData = rowDataPair.getLeft(); + recordSize += rowDataPair.getRight(); Review Comment: Fixed already. -- 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: commits-unsubscr...@inlong.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
