amogh-jahagirdar commented on code in PR #16285: URL: https://github.com/apache/iceberg/pull/16285#discussion_r3678823876
########## core/src/main/java/org/apache/iceberg/ColumnFileStruct.java: ########## @@ -0,0 +1,271 @@ +/* + * 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; + +import java.io.Serializable; +import java.nio.ByteBuffer; +import java.util.Arrays; +import java.util.List; +import org.apache.iceberg.avro.SupportsIndexProjection; +import org.apache.iceberg.relocated.com.google.common.base.MoreObjects; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.apache.iceberg.relocated.com.google.common.collect.Sets; +import org.apache.iceberg.types.Types; +import org.apache.iceberg.util.ArrayUtil; +import org.apache.iceberg.util.ByteBuffers; + +/** Mutable {@link StructLike} implementation of {@link ColumnFile}. */ +class ColumnFileStruct extends SupportsIndexProjection implements ColumnFile, Serializable { + private static final Types.StructType BASE_TYPE = + Types.StructType.of( + ColumnFile.FORMAT_VERSION, + ColumnFile.FIELD_IDS, + ColumnFile.LOCATION, + ColumnFile.FILE_FORMAT, + ColumnFile.FILE_SIZE_IN_BYTES, + ColumnFile.KEY_METADATA, + ColumnFile.SPLIT_OFFSETS); + + private int formatVersion = -1; + private int[] fieldIds = null; + private String location = null; + private FileFormat fileFormat = null; + private long fileSizeInBytes = -1L; + private byte[] keyMetadata = null; + private long[] splitOffsets = null; + + /** Used by internal readers to instantiate this class with a projection schema. */ + ColumnFileStruct(Types.StructType projection) { + super(BASE_TYPE, projection); + } + + ColumnFileStruct( + int formatVersion, + List<Integer> fieldIds, + String location, + FileFormat fileFormat, + long fileSizeInBytes, + ByteBuffer keyMetadata, + List<Long> splitOffsets) { + super(BASE_TYPE.fields().size()); + this.formatVersion = formatVersion; + this.fieldIds = ArrayUtil.toIntArray(fieldIds); + this.location = location; + this.fileFormat = fileFormat; + this.fileSizeInBytes = fileSizeInBytes; + this.keyMetadata = ByteBuffers.toByteArray(keyMetadata); + this.splitOffsets = ArrayUtil.toLongArray(splitOffsets); + } + + /** Copy constructor. */ + private ColumnFileStruct(ColumnFileStruct toCopy) { + super(toCopy); + this.formatVersion = toCopy.formatVersion; + this.fieldIds = + toCopy.fieldIds != null ? Arrays.copyOf(toCopy.fieldIds, toCopy.fieldIds.length) : null; + this.location = toCopy.location; + this.fileFormat = toCopy.fileFormat; + this.fileSizeInBytes = toCopy.fileSizeInBytes; + this.keyMetadata = + toCopy.keyMetadata != null + ? Arrays.copyOf(toCopy.keyMetadata, toCopy.keyMetadata.length) + : null; + this.splitOffsets = + toCopy.splitOffsets != null + ? Arrays.copyOf(toCopy.splitOffsets, toCopy.splitOffsets.length) + : null; + } + + /** Constructor for Java serialization. */ + ColumnFileStruct() { + super(BASE_TYPE.fields().size()); + } + + @Override + public int formatVersion() { + return formatVersion; + } + + @Override + public List<Integer> fieldIds() { + return fieldIds != null ? ArrayUtil.toUnmodifiableIntList(fieldIds) : null; + } + + @Override + public String location() { + return location; + } + + @Override + public FileFormat fileFormat() { + return fileFormat; + } + + @Override + public long fileSizeInBytes() { + return fileSizeInBytes; + } + + @Override + public ByteBuffer keyMetadata() { + return keyMetadata != null ? ByteBuffer.wrap(keyMetadata) : null; + } + + @Override + public List<Long> splitOffsets() { + return splitOffsets != null ? ArrayUtil.toUnmodifiableLongList(splitOffsets) : null; + } + + @Override + public ColumnFile copy() { + return new ColumnFileStruct(this); + } + + @Override + protected <T> T internalGet(int pos, Class<T> javaClass) { + return javaClass.cast(getByPos(pos)); + } + + private Object getByPos(int pos) { + return switch (pos) { + case 0 -> formatVersion; + case 1 -> fieldIds(); + case 2 -> location; + case 3 -> fileFormat != null ? fileFormat.toString() : null; Review Comment: I think I've noticed this on other PRs for TrackedFile for instance, when would fileFormat be null? It's required right? ########## core/src/main/java/org/apache/iceberg/TrackedFileStruct.java: ########## @@ -130,9 +137,11 @@ class TrackedFileStruct extends SupportsIndexProjection implements TrackedFile, this.keyMetadata = ByteBuffers.toByteArray(keyMetadata); this.splitOffsets = ArrayUtil.toLongArray(splitOffsets); this.equalityIds = ArrayUtil.toIntArray(equalityIds); + this.columnFiles = columnFiles != null ? Lists.newArrayList(columnFiles) : null; } /** Copy constructor. */ + @SuppressWarnings("CyclomaticComplexity") Review Comment: Any reasonable way to tighten this, or is it just inherent complexity from all the null handling we need to do? ########## core/src/main/java/org/apache/iceberg/ColumnFile.java: ########## @@ -0,0 +1,94 @@ +/* + * 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; + +import java.nio.ByteBuffer; +import java.util.List; +import org.apache.iceberg.types.Types; + +interface ColumnFile { + Types.NestedField FORMAT_VERSION = + Types.NestedField.required( + 161, "format_version", Types.IntegerType.get(), "Format version of this column file"); + Types.NestedField FIELD_IDS = + Types.NestedField.required( + 162, + "field_ids", + Types.ListType.ofRequired(163, Types.IntegerType.get()), + "Live field IDs in this column file"); + Types.NestedField LOCATION = + Types.NestedField.required( + 164, "location", Types.StringType.get(), "Location of the column file"); + Types.NestedField FILE_FORMAT = + Types.NestedField.required( + 165, + "file_format", + Types.StringType.get(), + "String file format name for this column file"); + Types.NestedField FILE_SIZE_IN_BYTES = + Types.NestedField.required( + 166, "file_size_in_bytes", Types.LongType.get(), "Total column file size in bytes"); + Types.NestedField KEY_METADATA = + Types.NestedField.optional( + 167, + "key_metadata", + Types.BinaryType.get(), + "Implementation-specific key metadata for encryption"); + Types.NestedField SPLIT_OFFSETS = + Types.NestedField.optional( + 168, + "split_offsets", + Types.ListType.ofRequired(169, Types.LongType.get()), + "Split offsets for the data file"); + + static Types.StructType schema() { + return Types.StructType.of( + FORMAT_VERSION, + FIELD_IDS, + LOCATION, + FILE_FORMAT, + FILE_SIZE_IN_BYTES, + KEY_METADATA, + SPLIT_OFFSETS); + } + + /** Returns the format version of this column file. */ + int formatVersion(); + + /** Returns the field IDs contained in this column file. */ + List<Integer> fieldIds(); + + /** Returns the location of this column file. */ + String location(); + + /** Returns the format of this column file. */ + FileFormat fileFormat(); + + /** Returns the total size of this column file in bytes. */ + long fileSizeInBytes(); + + /** Returns encryption key metadata, or null if this column file is not encrypted. */ + ByteBuffer keyMetadata(); + + /** Returns the list of recommended split locations for this column file, or null. */ + List<Long> splitOffsets(); Review Comment: I remember there was a discussion around if we need split offsets per column file or not, what was the conclusion on that? I feel like it makes sense to have them, there may be different files with varying schemas and as a result there would be different boundaries that would be set appropriately for each file. It's additional complexity, but not much? ########## core/src/main/java/org/apache/iceberg/TrackedFileStruct.java: ########## @@ -167,6 +176,13 @@ private TrackedFileStruct(TrackedFileStruct toCopy, Set<Integer> statsIds) { toCopy.equalityIds != null ? Arrays.copyOf(toCopy.equalityIds, toCopy.equalityIds.length) : null; + this.columnFiles = Review Comment: Minor: Could we express this without the Java stream? It's just a bit on the hot path when reading entries so all those allocations from the additional objects that streams create may add up. ########## core/src/main/java/org/apache/iceberg/TrackingBuilder.java: ########## @@ -25,13 +25,14 @@ class TrackingBuilder { private final long newSnapshotId; private final Long snapshotId; - private final Long dataSequenceNumber; Review Comment: Does this need to move? -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
