RussellSpitzer commented on a change in pull request #3983: URL: https://github.com/apache/iceberg/pull/3983#discussion_r830120764
########## File path: spark/v3.2/spark/src/main/java/org/apache/iceberg/spark/actions/Spark3ZOrderUDF.java ########## @@ -0,0 +1,241 @@ +/* + * Licensed 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.spark.actions; + +import java.io.IOException; +import java.io.Serializable; +import java.nio.ByteBuffer; +import java.nio.charset.CharsetEncoder; +import java.nio.charset.StandardCharsets; +import org.apache.iceberg.util.ZOrderByteUtils; +import org.apache.spark.sql.Column; +import org.apache.spark.sql.expressions.UserDefinedFunction; +import org.apache.spark.sql.functions; +import org.apache.spark.sql.types.BinaryType; +import org.apache.spark.sql.types.BooleanType; +import org.apache.spark.sql.types.ByteType; +import org.apache.spark.sql.types.DataType; +import org.apache.spark.sql.types.DataTypes; +import org.apache.spark.sql.types.DateType; +import org.apache.spark.sql.types.DoubleType; +import org.apache.spark.sql.types.FloatType; +import org.apache.spark.sql.types.IntegerType; +import org.apache.spark.sql.types.LongType; +import org.apache.spark.sql.types.ShortType; +import org.apache.spark.sql.types.StringType; +import org.apache.spark.sql.types.TimestampType; +import scala.collection.Seq; + +class SparkZOrder implements Serializable { + private final int STRING_KEY_LENGTH = 128; + + private final byte[] TINY_EMPTY = new byte[Byte.BYTES]; + private final byte[] SHORT_EMPTY = new byte[Short.BYTES]; + private final byte[] INT_EMPTY = new byte[Integer.BYTES]; + private final byte[] LONG_EMPTY = new byte[Long.BYTES]; + private final byte[] FLOAT_EMPTY = new byte[Float.BYTES]; + private final byte[] DOUBLE_EMPTY = new byte[Double.BYTES]; + + transient private ThreadLocal<ByteBuffer> outputBuffer; + transient private ThreadLocal<byte[][]> inputHolder; + transient private ThreadLocal<ByteBuffer>[] inputBuffers; + transient private ThreadLocal<CharsetEncoder> encoder; + + private final int numCols; + + private int inputCol = 0; + private int totalBytes = 0; + + SparkZOrder(int numCols) { + this.numCols = numCols; + } + + private void readObject(java.io.ObjectInputStream in) + throws IOException, ClassNotFoundException { + in.defaultReadObject(); + inputBuffers = new ThreadLocal[numCols]; + inputHolder = ThreadLocal.withInitial(() -> new byte[numCols][]); + encoder = ThreadLocal.withInitial(() -> StandardCharsets.UTF_8.newEncoder()); + } + + + private ByteBuffer outputBuffer(int size) { + if (outputBuffer == null) { + // May over allocate on concurrent calls + outputBuffer = ThreadLocal.withInitial(() -> ByteBuffer.allocate(size)); + } + return outputBuffer.get(); + } + + private ByteBuffer inputBuffer(int position, int size){ + if (inputBuffers[position] == null) { + // May over allocate on concurrent calls + inputBuffers[position] = ThreadLocal.withInitial(() -> ByteBuffer.allocate(size)); + } + return inputBuffers[position].get(); + } + + byte[] interleaveBits(Seq<byte[]> scalaBinary) { + byte[][] columnsBinary = scala.collection.JavaConverters.seqAsJavaList(scalaBinary) + .toArray(inputHolder.get()); + return ZOrderByteUtils.interleaveBits(columnsBinary, totalBytes, outputBuffer(totalBytes)); + } + + private UserDefinedFunction tinyToOrderedBytesUDF() { + int position = inputCol; Review comment: Do you have any ideas how this would look? I'm not sure we can make it much better ... but maybe I lack vision -- 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]
