A comment on the comments on the comments:
On 02/04/2013 1:58 PM, Ian Robinson wrote:
> - "Pg 30 - code sample - Kind of an awkward example because the test
> bean is injected but the UserTransaction and TransactionManager are
> not. I know that CDI supports injection of UT, so it could have an @Inject
> on it, but I don't think CDI supports injecting the TM."
> It doesn't and it shouldn't ever try! TM is not an interface that
> applications have access to so there is (deliberately) no
> lookup/injection method specified. I guess a slightly less awkward
> pseudocode could add:
>
> transactionManager = getTransactionManager();
>
> where the function is some internal helper class provided by the
> container.
Right, I was not suggesting that CDI should support injecting the TM,
although it is still used by some applications from before TSR was
introduced and many servers still make it available (and also because
tests like this one have no other way to suspend and resume
transactions!). The point was just that although the bean variable was
injected and filled in, the TM and UT variables were left uninitialized
which seemed a little odd. Although it wouldn't compile, an ellipsis
might even do the trick, e.g. TransactionManager tm = ... ;
Not a big deal either way, though.
> From: Paul Parkinson <paul.parkinson_at_oracle.com>
> To: users_at_jta-spec.java.net,
> Cc: Joseph Bergmark <bergmark_at_us.ibm.com>, Bill Shannon
> <bill.shannon_at_oracle.com>, Kevin Sutter <sutter_at_us.ibm.com>, michael
> keith <michael.keith_at_oracle.com>
> Date: 30/03/2013 00:31
> Subject: [jta-spec users] Re: latest (and hopefully final) spec and
> javadoc uploaded
> ------------------------------------------------------------------------
>
>
>
> Hello Ian,
>
> Thanks so much. It was a pleasure working with you as always.
>
> Mike Keith's eagle eye noticed a few formatting and typo type issues
> listed below. They probably could've been pushed to the next release
> but I figured we've gotten this far, we may as well do the extra
> cleanup now.
>
> Unless I hear otherwise the rev I've just uploaded (which I'll send
> out a separate post/email on) will be the final.
>
> Cheers and hooray for JTA 1.2 (JTA 1.3 and ordered resource commit
> here we come ;-) )...
> Paul
>
>
> Overall - "transaction manager" capitalization is inconsistent.
> Sometimes it is capitalized and sometimes not. Same goes for "resource
> manager".
>
> Pg 8 - lookup example - lookup throws a checked NamingException that
> must be either handled or declared
>
> Pg 13 - the bullets at the top of the page are actually micro question
> marks???!
>
> Pg 15 - Figure lists old versions of JDBC and JMS - should not be any
> versions there
>
> Pg 18 - 6th line - "multiple method invocation" - should be pluralized
>
> Pg 19 - State transition diagram - Not sure why recover() is there
>
> Pg 21 - code example - // dissociate xid1 to ... -> *disassociate*
> xid1 *from* ...
>
> Pg 22 - "Failures Recovery" heading - should not be pluralized
>
> Pg 22 - "representing each of the Resource Manager" - should be pluralized
>
> Pg 24 - code sample - xid1NewBranch variable was not declared with a type
>
> Pg 30 - code sample - Kind of an awkward example because the test bean
> is injected but the UserTransaction and TransactionManager are not. I
> know that CDI supports injection of UT, so it could have an @Inject
> on it, but I don't think CDI supports injecting the TM.
> - The declared variable "Transaction transaction" gets assigned the
> suspended tx, but later on in the code we see "resume(tx)"
> - line: "// tx2 committed, assert notransaction scope is active" -
> need a space for "no transaction"
> - line: "// tx1 commit, assert no transactionscope is active"- need a
> space for "transaction scope"
> - line: "should have thrownContextNotActiveException" - need a space
> for "thrown ContextNotActiveException"
>
> Pg 34 - figure - shows a Java Application obtaining a resource using
> "javax.tbd.getResource". This is invalid and should just be replaced
> with some words like "look up resource" or "get resource from
> factory", or something
>
> Pg 35 - First line - "The code sample below illustrate..." - should be
> pluralized
> - NamingException not handled from lookup method
>
> - section 4.2, 1. "... annotated with Transactional(TxType.REQUIRED)
> ..." - missing @
>
>
>
> On Mar 29, 2013, at 10:47 AM, Ian Robinson wrote:
>
> Hi Paul - this looks good to me. I have no further comments on this draft.
> Thanks, and nice job in getting it there.
>
> Regards,
> Ian/
> /
>
>
>
>
> From: Paul Parkinson <_paul.parkinson_at_oracle.com_
> <mailto:paul.parkinson_at_oracle.com>>
> To: _users_at_jta-spec.java.net_ <mailto:users_at_jta-spec.java.net>,
> Cc: Joseph Bergmark <_bergmark_at_us.ibm.com_
> <mailto:bergmark_at_us.ibm.com>>, Bill Shannon <_bill.shannon_at_oracle.com_
> <mailto:bill.shannon_at_oracle.com>>, Kevin Sutter <_sutter_at_us.ibm.com_
> <mailto:sutter_at_us.ibm.com>>
> Date: 27/03/2013 16:20
> Subject: [jta-spec users] latest (and hopefully final) spec and
> javadoc uploaded
> ------------------------------------------------------------------------
>
>
> _
> __http://java.net/projects/jta-spec/sources/spec-source-repository/content/JTA1.2Specification.pdf?rev=11__
> __http://java.net/projects/jta-spec/sources/spec-source-repository/content/JTA1.2Javadoc.zip?rev=11_
>
>
> Unless stated otherwise above:
> IBM United Kingdom Limited - Registered in England and Wales with
> number 741598.
> Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6
> 3AU
>
>
> Unless stated otherwise above:
> IBM United Kingdom Limited - Registered in England and Wales with
> number 741598.
> Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU