Here's what I have done regarding these comments. (No need to read this email unless you're particularly interested).
The updated draft spec is here:
http://java.net/projects/jms-spec/sources/repository/content/jms2.0/specification/word/JMS20.pdf
On 12/03/2012 11:11, Nigel Deakin wrote:
> On 29/02/2012 12:01, Nigel Deakin wrote:
>> The JMS 2.0 Early Draft is now posted for formal review. The review period ends on 29 March 2012.
>> http://jcp.org/aboutJava/communityprocess/edr/jsr343/index.html
>
> I've received these comments from a colleague at Oracle (they were actually commented in a previous draft so I'll need
> to check they still apply):
>
> p. 11 seems a little dated
This is a reference to Chapter 1 "Introduction". This whole chapter would benefit from rewriting. This is a big enough
task to merit logging as a JIRA issue:
http://java.net/jira/browse/JMS_SPEC-86
Chapter 1 "Introduction" is a little dated and requires rewriting
>
> 1.3, p 14 J2EE -> Java EE
This whole section requires rewriting (see above), but this small change has been made.
>
> 1.4.3, p. 14 EJB 2.0 - replace with current release or just remove version number
I've removed the "2.0" from all references to EJB 2.0 (given that it is ancient history).
>
> p. 16 starement -> statement
Was already fixed in the Early Draft
>
> 2.2 p.17 systems -> system's native client API
This is just a missing apostophe. Now fixed.
>
> 2.3 their -> its underlying messaging technology
This is the sentence "It is expected that each JMS provider will differ significantly in its underlying messaging
technology."
Yes, the phrase "each JMS provider" is singular. So it's "its", not "their".
>
> Figure 2.2 It would be helpful to show the cardinality of these
> relationships
This is an old diagram that shows the relationship between a connection factory, connection, session, message producer,
message consumer etc. Showing the cardinality would be useful though not essential.
However this whole section 2.5 "JMS Interfaces" needs to be updated to introduce the interfaces of the simplified API as
well.
I've logged this as a JIRA issue:
Section 2.5 "Interfaces" needs updating to introduce the simplified API
> 3.4.12 "client specified values" is confusing in view of 2 preceding
> paragraphs.
This section refers to the "client specified values" of JMSDeliveryMode, JMSExpiration and JMSPriority, despite the
previous section describing these as being set by the "send method" and *not* the client. This is indeed confusing.
I've substantially reworded both sections 3.4.11 and 3.4.12 (and extended table 3.1) to clarify this further.
> 3.5 Section name is missing in cross-reference
Fixed.
>
> 4.3.8, third paragraph. This is confusing.
> Can only one ExceptionListener be registered?
Yes
> How does
> "onMessage()" fit in (or was "onException()" meant)?
I wrote onMessage instead of onException in two places. This has been fixed.
> Why would one want to register an ExceptionListener with
> multiple connections? Wouldn't it be better to create a listener
> per connection? Can more than one ExceptionListener be set
> on a connection?
Yes, only one exception listener may be registered with a given connection. I hope that now I've made the corrections
mentioned above this is clear.
You can register the same exception listener with multiple connections. I can't see any particular reason for doing
that. The only reason I mentioned it is to clarify that calls to an exception listener for multiple connections are not
serialized.
> p. 87 J2EE -> Java EE
This is the sentence: "In the future, an integration point for JMS integration into J2EE platforms will be provided
using the J2EE Connector Architecture.
I've decided to simply delete this. We'll come back to the issue of JCA support as a separate issue (JMS_SPEC-25)
>
> p. 100 nessages -> messages
Fixed.
> There is some bug in the template that causes the footer for chapter
> 11 to be used for the 2 appendixes as well.
Now fixed.
Nigel