On Thu, 16 Jul 2026 10:46:42 GMT, Jaikiran Pai <[email protected]> wrote:
> Can I please get a review of this doc-only change which documents the purpose > of the `jdk.internal.value.Deserializer` annotation? > > > > > --------- > - [x] I confirm that I make this contribution in accordance with the [OpenJDK > Interim AI Policy](https://openjdk.org/legal/ai). LGTM after reviewing my comments src/java.base/share/classes/jdk/internal/value/Deserializer.java line 41: > 39: * Non-record value classes, unless they implement the {@code > writeReplace()} > 40: * and {@code readObject()} methods, cannot be serialized or > deserialized. Certain pre-existing > 41: * {@code Serializable} identity classes within the JDK are migrated to > value classes. Such a value Suggestion: * {@code Serializable} identity classes within the JDK are value classes in preview-mode, and must remain compatible. Such a value src/java.base/share/classes/jdk/internal/value/Deserializer.java line 54: > 52: * future value object persistence would be handled by other mechanisms. > The {@code Deserializer} > 53: * annotation isn't for general purpose usage, even in the classes that > belong to the JDK; it is > 54: * meant to be used by a very select few classes and the legacy > serialization places several Suggestion: * meant to be used only by a very select few classes and the legacy serialization places several ------------- Changes requested by vklang (no project role). PR Review: https://git.openjdk.org/valhalla/pull/2647#pullrequestreview-4713027653 PR Review Comment: https://git.openjdk.org/valhalla/pull/2647#discussion_r3594849159 PR Review Comment: https://git.openjdk.org/valhalla/pull/2647#discussion_r3594857679
