rdsr commented on a change in pull request #843: [WIP] InputFormat support for Iceberg URL: https://github.com/apache/incubator-iceberg/pull/843#discussion_r394483967
########## File path: mr/src/main/java/org/apache/iceberg/mr/IcebergInputFormat.java ########## @@ -0,0 +1,494 @@ +/* + * 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.iceberg.mr; + +import com.google.common.collect.Iterators; +import com.google.common.collect.Lists; +import com.google.common.collect.Sets; +import java.io.Closeable; +import java.io.DataInput; +import java.io.DataOutput; +import java.io.IOException; +import java.util.Arrays; +import java.util.Iterator; +import java.util.List; +import java.util.Set; +import org.apache.hadoop.conf.Configuration; +import org.apache.hadoop.fs.BlockLocation; +import org.apache.hadoop.fs.FileSystem; +import org.apache.hadoop.fs.Path; +import org.apache.hadoop.io.Writable; +import org.apache.hadoop.mapreduce.InputFormat; +import org.apache.hadoop.mapreduce.InputSplit; +import org.apache.hadoop.mapreduce.Job; +import org.apache.hadoop.mapreduce.JobContext; +import org.apache.hadoop.mapreduce.RecordReader; +import org.apache.hadoop.mapreduce.TaskAttemptContext; +import org.apache.iceberg.CombinedScanTask; +import org.apache.iceberg.DataFile; +import org.apache.iceberg.FileScanTask; +import org.apache.iceberg.PartitionField; +import org.apache.iceberg.PartitionSpec; +import org.apache.iceberg.Schema; +import org.apache.iceberg.SchemaParser; +import org.apache.iceberg.StructLike; +import org.apache.iceberg.Table; +import org.apache.iceberg.TableProperties; +import org.apache.iceberg.TableScan; +import org.apache.iceberg.avro.Avro; +import org.apache.iceberg.catalog.Catalog; +import org.apache.iceberg.catalog.TableIdentifier; +import org.apache.iceberg.data.GenericRecord; +import org.apache.iceberg.data.Record; +import org.apache.iceberg.data.avro.DataReader; +import org.apache.iceberg.data.parquet.GenericParquetReaders; +import org.apache.iceberg.exceptions.RuntimeIOException; +import org.apache.iceberg.expressions.Expression; +import org.apache.iceberg.hadoop.HadoopInputFile; +import org.apache.iceberg.hadoop.HadoopTables; +import org.apache.iceberg.hive.HiveCatalogs; +import org.apache.iceberg.io.CloseableIterable; +import org.apache.iceberg.io.InputFile; +import org.apache.iceberg.orc.ORC; +import org.apache.iceberg.parquet.Parquet; +import org.apache.iceberg.types.TypeUtil; +import org.apache.iceberg.types.Types; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + + +public class IcebergInputFormat<T> extends InputFormat<Void, T> { + private static final Logger LOG = LoggerFactory.getLogger(IcebergInputFormat.class); + + static final String AS_OF_TIMESTAMP = "iceberg.mr.as.of.time"; + static final String CASE_SENSITIVE = "iceberg.mr.case.sensitive"; + static final String FILTER_EXPRESSION = "iceberg.mr.filter.expression"; + static final String IN_MEMORY_DATA_MODEL = "iceberg.mr.in.memory.data.model"; + static final String READ_SCHEMA = "iceberg.mr.read.schema"; + static final String REUSE_CONTAINERS = "iceberg.mr.case.sensitive"; + static final String SNAPSHOT_ID = "iceberg.mr.snapshot.id"; + static final String SPLIT_SIZE = "iceberg.mr.split.size"; + static final String TABLE_PATH = "iceberg.mr.table.path"; + static final String TABLE_SCHEMA = "iceberg.mr.table.schema"; + private static final String LOCALITY = "iceberg.mr.locality"; + + private transient List<InputSplit> splits; + + public enum InMemoryDataModel { + PIG, + HIVE, + DEFAULT // Default data model is of Iceberg Generics + } + + /** + * Configures the {@code Job} to use the {@code IcebergInputFormat} and + * returns a helper to add further configuration. + * + * @param job the {@code Job} to configure + */ + public static ConfigBuilder configure(Job job) { + job.setInputFormatClass(IcebergInputFormat.class); + return new ConfigBuilder(job.getConfiguration()); + } + + public static class ConfigBuilder { + private final Configuration conf; + + public ConfigBuilder(Configuration conf) { + this.conf = conf; + } + + public ConfigBuilder readFrom(String path) { + conf.set(TABLE_PATH, path); + Table table = getTable(conf); + conf.set(TABLE_SCHEMA, SchemaParser.toJson(table.schema())); + return this; + } + + public ConfigBuilder filter(Expression expression) { + conf.set(FILTER_EXPRESSION, SerializationUtil.serializeToBase64(expression)); + return this; + } + + public ConfigBuilder project(Schema schema) { + conf.set(READ_SCHEMA, SchemaParser.toJson(schema)); + return this; + } + + public ConfigBuilder reuseContainers(boolean reuse) { + conf.setBoolean(REUSE_CONTAINERS, reuse); + return this; + } + + public ConfigBuilder caseSensitive(boolean caseSensitive) { + conf.setBoolean(CASE_SENSITIVE, caseSensitive); + return this; + } + + public ConfigBuilder snapshotId(long snapshotId) { + conf.setLong(SNAPSHOT_ID, snapshotId); + return this; + } + + public ConfigBuilder asOfTime(long asOfTime) { + conf.setLong(AS_OF_TIMESTAMP, asOfTime); + return this; + } + + public ConfigBuilder splitSize(long splitSize) { + conf.setLong(SPLIT_SIZE, splitSize); + return this; + } + + public ConfigBuilder locality(boolean localityPreferred) { + conf.setBoolean(LOCALITY, localityPreferred); + return this; + } + + public ConfigBuilder inMemoryDataModel(InMemoryDataModel inMemoryDataModel) { + conf.set(IN_MEMORY_DATA_MODEL, inMemoryDataModel.name()); + return this; + } + } + + @Override + public List<InputSplit> getSplits(JobContext context) { + if (splits != null) { + LOG.info("Returning cached splits: {}", splits.size()); + return splits; + } + + Configuration conf = context.getConfiguration(); + Table table = getTable(conf); + TableScan scan = table.newScan() + .caseSensitive(conf.getBoolean(CASE_SENSITIVE, true)); + long snapshotId = conf.getLong(SNAPSHOT_ID, -1); + if (snapshotId != -1) { + scan = scan.useSnapshot(snapshotId); + } + long asOfTime = conf.getLong(AS_OF_TIMESTAMP, -1); + if (asOfTime != -1) { + scan = scan.asOfTime(asOfTime); + } + long splitSize = conf.getLong(SPLIT_SIZE, -1); + if (splitSize != -1) { + scan = scan.option(TableProperties.SPLIT_SIZE, String.valueOf(splitSize)); + } + String schemaStr = conf.get(READ_SCHEMA); + if (schemaStr != null) { + scan.project(SchemaParser.fromJson(schemaStr)); + } + + // TODO add a filter parser to get rid of Serialization + Expression filterExpression = SerializationUtil.deserializeFromBase64(conf.get(FILTER_EXPRESSION)); + if (filterExpression != null) { + scan = scan.filter(filterExpression); + } + + splits = Lists.newArrayList(); + try (CloseableIterable<CombinedScanTask> tasksIterable = scan.planTasks()) { + tasksIterable.forEach(task -> splits.add(new IcebergSplit(conf, task))); + } catch (IOException e) { + throw new RuntimeIOException(e, "Failed to close table scan: %s", scan); + } + + return splits; + } + + @Override + public RecordReader<Void, T> createRecordReader(InputSplit split, TaskAttemptContext context) { + return new IcebergRecordReader<>(); + } + + private static final class IcebergRecordReader<T> extends RecordReader<Void, T> { + private TaskAttemptContext context; + private Iterator<FileScanTask> tasks; + private Iterator<T> currentIterator; + private T currentRow; + private Schema expectedSchema; + private Schema tableSchema; + private InMemoryDataModel inMemoryDataModel; + private Closeable currentCloseable; + private boolean reuseContainers; + private boolean caseSensitive; + + @Override + public void initialize(InputSplit split, TaskAttemptContext newContext) { + Configuration conf = newContext.getConfiguration(); + CombinedScanTask task = ((IcebergSplit) split).task; Review comment: If the splits are only created using Iceberg apis, I'd imagine we will only be getting IcebergSpits , no? ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
