Sam, here's the latest table of failing TCK tests: (See attached file: TCKIssues.html). Every failure except the attachment failure has been reported. Most have TCK problem #'s associated with them. There is a chance that these failures are hiding other failures, but Rich went through and fudged the AXIS code in his sandbox to get rid of some of these (Calendar, boolean accessor, full bean ctor) and he got most of the tests to pass.

There's the outstanding doc/lit issue. We have yet to convince the Sun folks this is a TCK issue (or they have yet to convince us that it is not).

Since I haven't implemented attachment support, yet, issues may arise there. I do have a potential one already. Remember my point #3 on doc/lit? I've copied it below to refresh your memory. This same issue applies to attachment support. It's only in the binding that we see mime info, so the binding forces the SEI to change. Granted, given the way most folks write/use WSDL, the scenario in point #3 is unlikely, but it IS possible. Do y'all think it's worth bringing this up as a TCK issue since the spec is faulty, or should I just ignore the issue and code it with the assumption that there will be only one binding per portType?



3. Once upon a time JAX-RPC said that, if the binding info affects the signatures of the SEI methods, then the name of the SEI is derived from the name of the binding instead of the portType. JAX-RPC no longer says that. The implication of the absence of that statement is that the SEI will be the same no matter the binding. Consider a WSDL that contains the following:
- a portType named "PT"
- an rpc/encoded binding for "PT"
- a doc/lit binding for "PT"
The name of the SEI for BOTH these bindings will be derived from the portType "PT". Unless the SEI is the same for both, we have an interface clash. Therefore, the fact that a binding is doc/lit must not affect the SEI.

Russell Butek
[EMAIL PROTECTED]

Please respond to [EMAIL PROTECTED]

To: [EMAIL PROTECTED]
cc:
Subject: [Fwd: jaxrpc issues]





Rich, Russell - what's the outlook for more issues?

-------- Original Message --------
Subject: jaxrpc issues
Date: Mon, 15 Jul 2002 13:37:49 -0400
From: Lance Andersen
To: Sam Ruby , Davanum Srinivas

Hi Sam, Dims,

I saw Rich's email regarding your jaxrpc issues.  Outside of the doc/lit
issue which needs to be worked via Rahul from a spec perspective, is
there any other outstanding  issues that might be coming my way?  I am
trying to get the work wrapped up on the jaxrpc patch so this info would
be helpful.

Dims, can you please answer the same question regarding SAAJ?

Regards,
Lance

--
Lance Andersen                       email: [EMAIL PROTECTED]
Sun Microsystems Inc.                phone: (781) 442-2037
1 Network Drive, UBUR02-301          fax  : (781) 442-1610
Burlington, MA 01803






Title: TCKIssues
 
Test  TCK issue Status
jaxrpc/api/javax_xml_rpc./Service/Client.java#
GetPortsPosTest2 
4. getPorts should throw ServiceException wsdl
Note: Axis was changed to pass this test
TCK bug #4709354 assigned
jaxrpc/api/javax_xml_rpc/Call/Client.java#
SetGetPropertyTest1_from_standalone(passes but w/ Stub issue)
8.  Stub/Call get/setProperty issue TCK bug #4708407 assigned
jaxrpc/api/javax_xml_rpc/Stub/Client.java#
setPropertyTest5_from_standalone
8.  Stub/Call get/setProperty issue TCK bug #4708407 assigned
jaxrpc/ee/j2w/attachmentstest/Client.java#
AttachmentsTest_from_standalone
AXIS needs work Russell is working on it
jaxrpc/ee/j2w/inheritedinterfacetest/Client.java#
inheritedinterfaceTest_from_standalone
12. Client.class is stale reported, hopefully will go away in next version of TCK
jaxrpc/ee/j2w/marshalltest/Client.java#
MarshallStandardJavaClassesTest_from_standalone
5.  Roundtrip issues TCK bug #4708425 assigned
jaxrpc/ee/j2w/marshalltest/Client.java#
MarshallJavaBeanTest_from_standalone
11.  Calendar compare issue TCK bug #4714327 assigned
jaxrpc/ee/j2w/marshalltest/Client.java#
MarshallValueTypeTest_from_standalone
11.  Calendar compare issue TCK bug #4714327 assigned
jaxrpc/ee/j2w/marshalltest/Client.java#
MarshallJavaArrayTest_from_standalone
5.  Roundtrip issues TCK bug #4708425 assigned
jaxrpc/ee/j2w/marshalltest/Client.java#
MarshallJavaMultiArrayTest_from_standalone
5.  Roundtrip issues TCK bug #4708425 assigned
jaxrpc/ee/j2w/marshalltest/Client.java#
MarshallServiceExceptionTest_from_standalone
Axis problem with Java2WSDL and Exceptions Resolved: Scheu Fixed Axis
jaxrpc/ee/w2j/document/literal/marshaltest/Client.java#
MarshallSimpleTypesTest_from_standalone
6.  wrapped doc/lit
jaxrpc/ee/w2j/document/literal/marshaltest/Client.java#
MarshallArraysOfSimpleTypesTest_from_standalone
6.  wrapped doc/lit
jaxrpc/ee/w2j/document/literal/simpletest/Client.java#
SimpleTest_from_standalone
6.  wrapped doc/lit
jaxrpc/ee/w2j/document/literal/stockquote/Client.java#
invokeStockQuoteService_from_standalone
6.  wrapped doc/lit
jaxrpc/ee/w2j/rpc/encoded/marsalltest/Client.java#
MarshallAllSimpleXMLDataTypeTest_from_standalone
7.  Full bean constructor issue
(Some of the marshal tests also have issues 2 and 11, but they are hidden by 7)
TCK bug #4708351 assigned
(TCK bug #4708425 assigned, TCK bug #4714327 assigned)
jaxrpc/ee/w2j/rpc/encoded/marsalltest/Client.java#
MarshallAllStructsTest_from_standalone
7.  Full bean constructor issue TCK bug #4708365 assigned
jaxrpc/ee/w2j/rpc/encoded/marsalltest/Client.java#
MarshallSequenceStructsTest_from_standalone
7.  Full bean constructor issue TCK bug #4708365 assigned
jaxrpc/ee/w2j/rpc/encoded/marsalltest/Client.java#
MarshallWSDLRestrictedArrayTest_from_standalone
7.  Full bean constructor issue TCK bug #4708365 assigned
jaxrpc/ee/w2j/rpc/encoded/marsalltest/Client.java#
MarshallSOAPRestrictedArrayTest_from_standalone
7.  Full bean constructor issue TCK bug #4708365 assigned
jaxrpc/ee/w2j/rpc/encoded/marsalltest/Client.java#
MarshallExceptionsTest_from_standalone
7.  Full bean constructor issue TCK bug #4708365 assigned
jaxrpc/ee/w2j/rpc/encoded/marsalltest/Client.java#
MarshallEnumerationsTest_from_standalone
7.  Full bean constructor issue TCK bug #4708365 assigned
jaxrpc/ee/w2j/rpc/encoded/marsalltest/Client.java#
MarshallAllSimpleXMLDataTypesTest_from_standalone
7.  Full bean constructor issue TCK bug #4708365 assigned
jaxrpc/ee/w2j/rpc/encoded/parametermodetest/Client.java#
InSimpleTypesTest_from_standalone
11.  Calendar issue TCK bug #4714327 assigned
jaxrpc/ee/w2j/rpc/encoded/parametermodetest/Client.java#
OutSimpleTypesTest_from_standalone
11.  Calendar issue TCK bug #4714327 assigned
jaxrpc/ee/w2j/rpc/encoded/parametermodetest/Client.java#
InOutSimpleTypesTest_from_standalone
11.  Calendar issue TCK bug #4714327 assigned
jaxrpc/ee/w2j/rpc/encoded/parametermodetest/Client.java#
InStructTest_from_standalone
2.  Boolean value issue TCK bug #4708351 assigned
jaxrpc/ee/w2j/rpc/encoded/xmlnamemappingtest/Client.java#
JavaKeywordsTest_from_standalone
3. get<port> xml name 
7. Full bean constructor issue
Scheu looking into this one
TCK bug #4708351 assigned
jaxrpc/ee/w2j/rpc/encoded/xmlnamemappingtest/Client.java#
JavaNamingConventionsTest_from_standalone
3. get<port> xml name
7. Full bean constructor issue
Scheu looking into this one
TCK bug #4708351 assigned

 

Issue Explanations

Issue: 2 Boolean value

The TCK expects getVarBooloon instead of isVarBoolean for a bean property that is a boolean value.

Issue: 3 get<port> xml name

TCK expect getxMLNameMappingTestService method on the service and Axis has getXMLNameMappingTestService.
Sun doesn't see a problem.  We may need to change Axis to comply.  (Note the WSDL port name is xMLNameMappingTestService.) Rich Scheuerle is investigating.

Issue: 4 getPorts() problem

TCK expects a call to getPorts() to return a null iterator if there is no wsdl.  The specification clearly indicates that an exception should be thrown.  Axis was changed to comply with the TCK.  The Axis change should be backed out once the TCK is fixed.

Issue: 5. Roundtrip Issues

Problems with roundtripping Calendar/Date.

Issue: 6 Wrapped Doc/Lit

The TCK and Axis do not agree on the way to interpret wrapped doc/lit.
 

Issue: 7 Full Bean Constructor

The TCK relies on full constructors in the beans, which is not compliant with the specification.
 

Issue: 8 Stub/Call Get/Set Property Issue

The TCK expects an exception when it sets/gets an unknown property.  AXIS does not throw an exception.
 

Issue: 11 Calendar Issue

The compareCalendars method in the TCK should compare the canonical form of the calendars.
 

Issue: 12 Client.class is stale

The Client.class file for the inherited interface test does not match the Client.java.  Recompiling Client.java fixes the problem.
 

Reply via email to