[
https://issues.apache.org/jira/browse/WSCOMMONS-181?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12484485
]
Glen Daniels commented on WSCOMMONS-181:
----------------------------------------
Yes, fix for this is on the way! I'm just confirming a clean Axis build right
now then I'll check in.
The problem was that way too many places were relying on the
match-on-localName-only behavior of getFirstChildWithName(), when they should
either have been searching for full QNames or, better yet (as in this case)
just using another API like getHeader() which doesn't even do string
comparisons.
Sorry for the breakage, clearly our test coverage isn't quite adequate yet in
Axiom.
> SOAPHeader not serialised in Sandesha unit tests
> ------------------------------------------------
>
> Key: WSCOMMONS-181
> URL: https://issues.apache.org/jira/browse/WSCOMMONS-181
> Project: WS-Commons
> Issue Type: Bug
> Components: AXIOM
> Reporter: David Illsley
> Assigned To: Glen Daniels
> Priority: Critical
>
> The Sandesha unit tests fail when running from a clean maven repository. The
> problem is that the SOAPHeader is being omitted when it is serialised to the
> wire (the header appears in the in-memory object model before then).
> I took a look at SOAPEnvelopeImpl and it looks like the recent chenges to
> getFirstChildWithName have broken the logic in internalSerialize(). It
> currently assumes that if it passes in a new QName("Header") that either a
> SOAP1.1 or a SOAP 1.2 header element will be found. In reality, with these
> changes neither will.
> I did a quick hack locally to change the QName to add the SOAP 1.1 namespace
> and the unit tests then passed.
> I'm not an Axiom expert and I don't know what the best fix for this is (or
> why this problem isn't breaking the Axis2 tests! so I don't want to bodge a
> fix for this myself.
> Glen, I know you made some changes around this area, could you take a look?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
