jsr368-experts@jms-spec.java.net

[jsr368-experts] Re: [jms-spec users] Re: JMS 2.0 errata: final request for comments

From: Nigel Deakin <nigel.deakin_at_oracle.com>
Date: Fri, 02 Jan 2015 13:52:39 +0000

On 02/01/2015 13:19, Werner Keil wrote:
>
> On Fri, Jan 2, 2015 at 11:39 AM, Nigel Deakin <nigel.deakin_at_oracle.com <mailto:nigel.deakin_at_oracle.com>> wrote:
>
> On 23/12/2014 14:34, Werner Keil wrote:
>
> Looks OK to me as well.
>
> One question about https://java.net/jira/browse/__JMS_SPEC-157 <https://java.net/jira/browse/JMS_SPEC-157>
>
> It is proposed that the javadoc for the Connection object’s three createSession methods be changed to make it
> consistent
> with the Java EE 6 specification and with JMS 2.0 section 12.2.
>
> Why does this refer to Java EE 6? Is this an aspect that has not changed with Java EE 7 since then?
>
>
> The JIRA issue referenced explains the background to this issue, and particularly how JMS 2.0 incorporated various
> incorporate pieces of specification relating to JMS that were previously in Java EE 6, but accidentally introduced
> an incompatible change when doing so.
>
> This issue is about maintaining compatibility between Java EE versions. We need to refer to Java EE 6 when deciding
> what changes are needed now since it is Java EE 6 that defines the behaviour that Java EE 7 (including JMS 2.0)
> needs to be compatible with.
>
>
> Would that be something for a JMS 2.0 MR (since Java EE 7 contains 2.0 and is unlikely to include 2.1 "on the fly") or
> would we rather address that with 2.1, hence it wouldn't be changed before Java EE 8 (or solutions that apply JMS 2.1 on
> top of Java EE 7, that likely depends on how we use CDI, at the moment CDI 2 aims for Java SE 8+ so it's less likely
> you'd get CDI 2 without EE 8 or a standalone SE 8 app;-)

Yes, compatibility violations in JMS 2.0 need to be corrected immediately (in the JMS 2.0 errata that I am proposing)
since they prevent vendors implementing JMS 2.0.

Nigel