Bug / Issue Tracking Service
Bugzilla – Bug 4331
WSDL WG comment 1
Last modified: 2007-03-07 18:37:04 UTC
WSDL WG comment 1: As in WSDL 2.0 component designators, this spec recommends the creation of an identifier from the targetNamespace of the WSDL 1.1 document, and that this identifier can be resolved without considering imports and includes. Unlike WSDL 2.0, in WSDL 1.1 the targetNamespace is not required, and although there is no wsdl11:include, we have some evidence that some customers have used multiple wsdl11:imports of the same namespace (which can be the same as the targetNamespace) and different locations to modularlize their documents – and that a number of popular tools actually support this “abuse? of import. These situations demonstrate the limits of the assumption of a 1-1 correspondence between a WSDL 1.1 document and a WSDL 1.1 targetNamespace. The spec’s recommendation to construct an identifier using the targetNamespace doesn’t work in these situations. The spec should at least note situations (edge cases) which conflict with the advice about creation of an element identifier from the targetNamespace.
[13:30] paulc: 1. Change text into one or more notes to make it more prominent. [13:30] * Jonathan doesn't matter much, but referencing the BP doesn't seem very helpful in this regard. [13:30] paulc: 2. Add a note saying this spec does not work if there is no TargetNamespace in the WSDL 1.1 document [13:31] paulc: 3. Make the current text into a second note to cover other corner cases that do not work. [13:31] dorchard: 1. Note: wsdl 1.1 EIs require are targetNamespace so wsdl documents without a targetNamespace will not have EIs. [13:33] abbie: Issue 4331 resolved with the propsal above items 1,2,3 [13:34] cferris: RESOLUTION: 4331 resolved with 1, 2 and 3 above [13:34] cferris: rrsagent, where am i? [13:34] RRSAgent: See http://www.w3.org/2007/03/07-ws-policy-irc#T18-34-53