mjsax commented on code in PR #21408: URL: https://github.com/apache/kafka/pull/21408#discussion_r2778078789
########## streams/src/main/java/org/apache/kafka/streams/state/ValueTimestampHeaders.java: ########## @@ -0,0 +1,177 @@ +/* + * 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.kafka.streams.state; + +import org.apache.kafka.common.header.Headers; +import org.apache.kafka.common.header.internals.RecordHeaders; +import org.apache.kafka.streams.state.internals.HeadersDeserializer; + +import java.util.Objects; + +/** + * Combines a value with its timestamp and associated record headers. + * + * @param <V> the value type + */ +public final class ValueTimestampHeaders<V> { + + private final V value; + private final long timestamp; + //visible for test + volatile Headers headers; + private final byte[] rawHeaders; + + private ValueTimestampHeaders(final V value, final long timestamp, final Headers headers) { + this.value = value; + this.timestamp = timestamp; + this.headers = headers == null ? new RecordHeaders() : headers; + this.rawHeaders = null; + } + + private ValueTimestampHeaders(final V value, final long timestamp, final byte[] rawHeaders) { + this.value = value; + this.timestamp = timestamp; + this.headers = null; + this.rawHeaders = rawHeaders; + } + + /** + * Create a new {@link ValueTimestampHeaders} instance if the provided {@code value} is not {@code null}. + * + * @param value the value + * @param timestamp the timestamp + * @param headers the headers (may be {@code null}, treated as empty) + * @param <V> the type of the value + * @return a new {@link ValueTimestampHeaders} instance if the provided {@code value} is not {@code null}; + * otherwise {@code null} is returned + */ + public static <V> ValueTimestampHeaders<V> make(final V value, + final long timestamp, + final Headers headers) { + if (value == null) { + return null; + } + return new ValueTimestampHeaders<>(value, timestamp, headers); + } + + /** + * Create a new {@link ValueTimestampHeaders} instance. + * The provided {@code value} may be {@code null}. + * + * @param value the value (may be {@code null}) + * @param timestamp the timestamp + * @param headers the headers (may be {@code null}, treated as empty) + * @param <V> the type of the value + * @return a new {@link ValueTimestampHeaders} instance + */ + public static <V> ValueTimestampHeaders<V> makeAllowNullable(final V value, + final long timestamp, + final Headers headers) { + return new ValueTimestampHeaders<>(value, timestamp, headers); + } + + /** + * Return the wrapped {@code value} of the given {@code valueTimestampHeaders} parameter + * if the parameter is not {@code null}. + * + * @param valueTimestampHeaders a {@link ValueTimestampHeaders} instance; can be {@code null} + * @param <V> the type of the value + * @return the wrapped {@code value} of {@code valueTimestampHeaders} if not {@code null}; otherwise {@code null} + */ + public static <V> V getValueOrNull(final ValueTimestampHeaders<V> valueTimestampHeaders) { + return valueTimestampHeaders == null ? null : valueTimestampHeaders.value; + } + + /** + * <strong>Internal use only.</strong> This method is used by the deserialization infrastructure Review Comment: Was thinking about this more. And have a better idea (I believe). It's actually a good thing to have the public facing `ValueTimestampHeaders` being a `final` class so might be better to not change. Overall, lazy deserialization should be an impl detail the user does not see, and maybe even `ValueTimestampHeaders` does not need to know about it -- it seems to be a leaky abstraction if we add a `makeWithRawHeaders()` method (even if we add it on a different, non-public class). It seems better to fully embed lazy deserialization inside `HeadersDeserializer`. I would propose to change its `deserialize(...)` method: ``` public Headers deserialize(final String topic, final byte[] data) { return new LazyRecordHeaders(data); } ``` We just return a new "headers" object, and push the deserialization inside `LazyRecordHeaders` (a new class we add: not sure if this extend `RecordHeaders` or just implement `Headers` interface directly) itself. The constructor would just store `data` byte array, and when the object is accessed, we deserialize internally. Thoughts? \cc @aliehsaeedii -- 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]
