- From: David Illsley <david.illsley@uk.ibm.com>
- Date: Fri, 27 Jan 2006 07:45:39 +0000
- To: <paul.downey@bt.com>
- Cc: public-ws-addressing-tests@w3.org, public-ws-addressing-tests-request@w3.org
- Message-ID: <OF5AC6D871.D855F072-ON80257102.007D1E00-80257103.002A9A8F@uk.ibm.com>
OK, in attempt to be a little more clear than on the phone, and in an
effort to have the reasoning behind the proposed changes archived, here is
a small explanation. (in addition to sample messages)
Tests 1233,1234 and 1235 (I didn't notice 1235 before because it hasn't
been implemented so isn't red) include the XPaths:
soap12:Envelope/soap12:Body/soap12:Fault/soap12:Code/soap12:Value/(resolve-QName(.,.)
= xs:QName('echo:EmptyEchoString'))
This appears to be an invalid requirement per the SOAP 1.2 Spec Part 1,
Section 5.4.6 [1], any custom faultcode should be a subcode. I would
therefore suggest the XPath:
soap12:Envelope/soap12:Body/soap12:Fault/soap12:Code/soap12:Subcode/soap12:Value/(resolve-QName(.,.)
= xs:QName('echo:EmptyEchoString'))
Sample Message:
<soapenv:Envelope xmlns:soapenv="http://www.w3.org/2003/05/soap-envelope"
xmlns:soapenc="http://www.w3.org/2003/05/soap-encoding"
xmlns:xsd="http://www.w3.org/2001/XMLSchema"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns:wsa="http://www.w3.org/2005/08/addressing">
<soapenv:Header>
<wsa:To>http://www.w3.org/2005/08/addressing/anonymous</wsa:To>
<CustomerKey xmlns="http://example.org/customer"
wsa:IsReferenceParameter="true">Fault#123456789</CustomerKey>
<wsa:Action>http://www.w3.org/2005/08/addressing/fault</wsa:Action>
<wsa:MessageID>uuid:E0A0E4EA-0108-4000-E000-08F0C0A80145</wsa:MessageID>
<wsa:RelatesTo>urn:uuid:5eddfb1b-c4c1-472f-af13-5730f5933a5c</wsa:RelatesTo>
</soapenv:Header> <soapenv:Body>
<soapenv:Fault>
<soapenv:Code>
<soapenv:Value>soapenv:Sender</soapenv:Value>
<soapenv:Subcode>
<soapenv:Value
xmlns:echo="http://example.org/echo">echo:EmptyEchoString</soapenv:Value>
</soapenv:Subcode>
</soapenv:Code>
<soapenv:Reason> <soapenv:Text xml:lang="en-US">String to echo was
empty.</soapenv:Text> </soapenv:Reason>
</soapenv:Fault>
</soapenv:Body>
</soapenv:Envelope>
Additionally, because there are issues when mapping a SOAP 1.2 programming
model to SOAP1.1 because SOAP 1.1 does not support subcodes we should
change the XPaths such that the implementation should be able to choose
which Code or Subcode to use, thus the XPaths should allow soap12:Sender
or echo:EmptyEchoString for tests 1133, 1134
Messages from Mike:
<s:Envelope xmlns:s="http://schemas.xmlsoap.org/soap/envelope/" xmlns:a="
http://www.w3.org/2005/08/addressing">
<s:Header>
<a:Action
s:mustUnderstand="1">http://www.w3.org/2005/08/addressing/fault</a:Action>
<a:RelatesTo>urn:uuid:c29a2099-60b5-4ec2-a270-eb1cf97baa62</a:RelatesTo>
</s:Header>
<s:Body>
<s:Fault>
<faultcode xmlns:a="http://example.org/echo
">a:EmptyEchoString</faultcode>
<faultstring xml:lang="en-US">The supplied string was of zero
length.</faultstring>
</s:Fault>
</s:Body>
</s:Envelope>
Or:
<s:Envelope xmlns:s="http://schemas.xmlsoap.org/soap/envelope/" xmlns:a="
http://www.w3.org/2005/08/addressing">
<s:Header>
<a:Action
s:mustUnderstand="1">http://www.w3.org/2005/08/addressing/fault</a:Action>
<a:RelatesTo>urn:uuid:c29a2099-60b5-4ec2-a270-eb1cf97baa62</a:RelatesTo>
</s:Header>
<s:Body>
<s:Fault>
<faultcode>s:Sender</faultcode>
<faultstring xml:lang="en-US">The supplied string was of zero
length.</faultstring>
</s:Fault>
</s:Body>
</s:Envelope>
David
[1] http://www.w3.org/TR/soap12-part1/#faultcodes
David Illsley
Web Services Development
IBM Hursley Park, SO21 2JN
+44 (0)1962 815049 (Int. 245049)
david.illsley@uk.ibm.com
<paul.downey@bt.com>
Sent by: public-ws-addressing-tests-request@w3.org
26/01/2006 21:24
To
<public-ws-addressing-tests@w3.org>
cc
Subject
Minutes - WS-Addressing Test Suite TF call 26 January 2006
Present:
David Illsley (IBM)
Mike Vernal (Microsoft)
Jonathan Marsh (Microsoft)
Arun Gupta (Sun)
Paul Downey (BT)
pauld: OK, where do we go from here?
arun: start publishing public endpoints!
arun: Sun's: http://soapinterop.java.sun.com:8080/index.html
public endpoints: http://www.w3.org/2002/ws/addr/testsuite/endpoints/
pauld: will contact giovani, invite him to join
jonathan: suspects sun, IBM and Microsoft are likely to be all be green or
good enough to be discussed at Cannes
.. understands that Glen is pretty tied up for the next few weeks
discussion: do we have enough implementations?
.. Sonic is only SOAP 1.1 at the moment
.. which might not be good enough for CR
arun: Sonic did mention at the F2F possible SOAP 1.2 in the near future
jonathan: is there anywhere else we can drum up support?
pauld: could be motivate to provide one, but we should flag this up to the
community
arun: others may be interested
jonathan: WSO2 might be a source of effort
jonathan: spells out missing parts, refresh endpoints; implement optional
and other tests; analyse red boxes
jonathan: offers help turn red boxes green, particularly where cause may
be log file weirdness
Arun: would like to refresh his logs, in particular sun to sun
pauld: offers assistance in helping people run the observer themselves
jonathan: want's to improve report generation to include merger of
override where errors introduced by the harness, or is deemed 'good
enough'
pauld: happy to accept fresh logs and re-run the report
mike: at the event we periodically regenerated the reports, can we have
that now?
pauld: babbles incoherently about why that might be tricky
pauld: would need log files to be available at a publicly visible uri
.. what can we do for next week?
.. seems like we'll have public endpoints refreshed by early next week
pauld: .. we should work towards having a refreshed report to discuss on
thursday
David: outlines issue with 1233
david: http://www.w3.org/2002/ws/addr/testsuite/testcases/#test1233
david: http://www.w3.org/TR/soap12-part1/#faultcodes
david: problem seems to be with the XPath
pauld: .. 1234 has the same problem
pauld: agreed to change the XPath on these two issues?
mike: distinction between SOAP 1.1 and 1.2 here
jonathan: there's ambiguity in SOAP 1.1 and we should allow vendors to
send either case
pauld: david is press-ganged into providing some simple test message
examples to express this problem
arun: I need to log off early today, so will be logging out now
jonathan: what about the new test cases? will they case problems?
david: could raise issues with SOAP processing, in particular 1248 is
tricky to send
.. would need a canned interaction
pauld: still keen to publish a canned implementation ..
mike: what's the difference between OPTIONAL and INFORMATIONAL
pauld: INFORMATIONAL not required for CR
jonathan: will create a third bucket on the report for INFORMATIONAL tests
discussion of issues list, rolling agendas and task lists.
pauld: happy to track list of tasks on the end of the minutes/agendas
.. let's keep up the momentum from the F2F ..
Current aim: to have a green report for Cannes, 27th February!
http://www.w3.org/2002/ws/addr/testsuite/report/
Task List:
all: update your public endpoints ASAP
all: provide logs on the Web (HTTPS/BA secured if necessary)
pauld: flag possible lack of 4 working implementations to the WG
all: solicit resource / additional implementations
pauld: will contact giovani, invite him to join our calls
david: to send simple example messages for SOAP faultcodes to the list
pauld: to resolve XPaths for faultcodes in 1133,1233, 1134,1234
jonathan: separate OPTIONAL/INFORMATIONAL on the report
jonathan: allow certain results to be overridden with qualification
pauld: will publish his canned endpoint
pauld: will work on automatically refreshing the report
-----
Previous meetings:
http://lists.w3.org/Archives/Public/public-ws-addressing-tests/2006Jan/0026.html
http://lists.w3.org/Archives/Public/public-ws-addressing-tests/2006Jan/0010.html
http://lists.w3.org/Archives/Public/public-ws-addressing-tests/2005Dec/0033.html
http://lists.w3.org/Archives/Public/public-ws-addressing-tests/2005Dec/0016.html
http://lists.w3.org/Archives/Public/public-ws-addressing-tests/2005Dec/0003.html
http://lists.w3.org/Archives/Public/public-ws-addressing-tests/2005Nov/0001.html
http://lists.w3.org/Archives/Public/public-ws-addressing-tests/2005Oct/0004.html
Received on Friday, 27 January 2006 07:46:18 UTC