On Thu, 27 Aug 2026 18:32:04 GMT, Alan Bateman <[email protected]> wrote:
>> Main changes: >> >> - Adding `(SSLEngine | SSLSocket | >> SSLServerSocket).getSupportedNamedGroups()` public API methods. >> - Adding `ExtendedSSLSession.getNegotiatedNamedGroup()` public API method. >> >> Additional side-issues being addressed: >> >> - Fixing a subtle bug in the implementation in JFR TLSHandshakeEvent: the >> event wasn't recorded on the client side for the TLSv1.2 resumed session >> (abbreviated handshake) and on the server side for TLSv1.2 full handshake. >> For a resumed handshake the client produces the last Finished in TLSv1.2, >> unlike the full handshake where the server produces the last Finished. In >> TLSv1.3 the client produces the last Finished for both resumed and full >> handshake. >> - Repurpose currently unused NamedGroup.namesOf() method >> >> --------- >> - [x] I confirm that I make this contribution in accordance with the >> [OpenJDK Interim AI Policy](https://openjdk.org/legal/ai). > > src/java.base/share/classes/javax/net/ssl/SSLEngine.java line 1508: > >> 1506: * @since 28 >> 1507: */ >> 1508: public String [] getSupportedNamedGroups() { > > I assume this is returning a String[] instead of a List<String> or > Set<String> because many of the existing method do this. For completeness, > the API docs could say if the names are unique if the ordering is significant. > > Can you confirm that returning an empty array wouldn't work for the "not > implemented by provider" case? - Yes, returning `String[] ` for consistency with existing methods. - Not sure if we should say that the names are unique and the ordering is significant: similar methods such as `getSupportedCipherSuites()` and `getSupportedProtocols()` don't document such requirement. While SunJSSE returns unique values in order of preference, other providers may choose to do otherwise. I need to think about it more. - Please elaborate about the `empty array ` question, not sure I understand. Technically this call can return an empty array, that would be a valid return value. ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/32560#discussion_r3875578564
