-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Doesn't seem to make a difference - will continue the chase ...
Johan James Carman wrote: > Yeah, you need to qualify names contributed in other modules. Prefix them > with the module id (and '.'). > > -----Original Message----- > From: Johan Lindquist [mailto:[EMAIL PROTECTED] > Sent: Wednesday, July 19, 2006 10:41 AM > To: user@hivemind.apache.org > Subject: Re: MethodInterceptorFactory and multiple interceptors > > Somewhat simplified :) > > > <implementation service-id="com.acme.Service"> > <interceptor name="webservice1" after="jms1" > service-id="hivemind.lib.MethodInterceptorFactory"> > <impl object="service:WebServiceMethodInterceptor" /> > </interceptor> > </implementation> > > Also tried fully qualified: > > <implementation service-id="com.acme.Service"> > <interceptor name="webservice1" after="integration.jms.jms1" > service-id="hivemind.lib.MethodInterceptorFactory"> > <impl object="service:WebServiceMethodInterceptor" /> > </interceptor> > </implementation> > > Same error with both. > > Johan > > James Carman wrote: >>> What syntax are you using in the hivemodule-webservice.xml file? >>> >>> >>> -----Original Message----- >>> From: Johan Lindquist [mailto:[EMAIL PROTECTED] >>> Sent: Wednesday, July 19, 2006 10:30 AM >>> To: user@hivemind.apache.org >>> Subject: Re: MethodInterceptorFactory and multiple interceptors >>> >>> Ok, i tried the name attribute - does this require a fully qualified >>> name or something? >>> >>> Currently getting the following error: >>> >>> 2006-07-19 16:28:21,919 ERROR com.acme.Service1 [error] Error at >>> > file:/home/johan/forge/projects/integration/target/test-classes/META-INF/hiv >>> emodule-webservice.xml, >>> line 61, column 121: Unknown interceptor contribution dependency >>> 'integration.jms.jms7' (for 'integration.webservice.webservice7'). >>> >>> Cheers, >>> >>> Johan >>> >>> James Carman wrote: >>>>> To be honest, that "name" attribute could be documented a bit better, >>>>> especially with respect to this factory. >>>>> >>>>> -----Original Message----- >>>>> From: Johan Lindquist [mailto:[EMAIL PROTECTED] >>>>> Sent: Wednesday, July 19, 2006 10:19 AM >>>>> To: user@hivemind.apache.org >>>>> Subject: Re: MethodInterceptorFactory and multiple interceptors >>>>> >>>>> What happens if you contribute twice to the service point outside of the >>>>> service point definition? >>>>> >>>>> Johan >>>>> >>>>> James Carman wrote: >>>>>>> In my test case for multiple method interceptors, I use multiple like >>>>> this: >>>>>>> <module id="hivemind.lib.test" version="1.0.0"> >>>>>>> >>>>>>> <service-point id="FortuneCookie" >>>>>>> interface="org.apache.hivemind.lib.impl.FortuneCookie"> >>>>>>> <create-instance >>>>>>> class="org.apache.hivemind.lib.impl.FortuneCookieImpl"/> >>>>>>> <interceptor service-id="hivemind.lib.MethodInterceptorFactory" >>>>>>> name="first"> >>>>>>> <impl >>>>>>> object="instance:org.apache.hivemind.lib.impl.SuffixMethodInterceptor" >>> /> >>>>>>> </interceptor> >>>>>>> <interceptor service-id="hivemind.lib.MethodInterceptorFactory" >>>>>>> name="second"> >>>>>>> <impl >>>>>>> object="instance:org.apache.hivemind.lib.impl.SuffixMethodInterceptor" >>> /> >>>>>>> </interceptor> >>>>>>> </service-point> >>>>>>> >>>>>>> </module> >>>>>>> >>>>>>> -----Original Message----- >>>>>>> From: Johan Lindquist [mailto:[EMAIL PROTECTED] >>>>>>> Sent: Wednesday, July 19, 2006 9:22 AM >>>>>>> To: user@hivemind.apache.org >>>>>>> Subject: MethodInterceptorFactory and multiple interceptors >>>>>>> >>>>>>> Hi All, >>>>>>> >>>>>>> I have recently been using the MethodInterceptorFactory to create >>>>>>> interceptors based on the AOP Alliance libraries. >>>>>>> >>>>>>> Now, it is all fine until I realised that I can not have two different >>>>>>> interceptors (using the MethodInterceptorFactory service id) added to >>>>>>> the same service point. >>>>>>> >>>>>>> This would make sense if you for example referred to the logging >>>>>>> interceptor - two of it would be kinda dum. But since the >>>>>>> MethodInterceptorFactory is simply providing the plumbing for the real >>>>>>> implementation, 2 or more of this particular contribution should be > fine >>>>>>> right? >>>>>>> >>>>>>> Not sure if this is an easy problem to solve or not? >>>>>>> >>>>>>> Cheers, >>>>>>> >>>>>>> johan >>>>>>> >>>>> -- >>>>> you too? >>> -- >>> you too? > > -- > you too? - -- you too? -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEvkg41Tv8wj7aQ34RAkpXAJ4xzBcW0o80m69Hb/uf5JyJ4AyulACfXc4M NIWK6/t8woLHGYBIG9vhdoY= =OROc -----END PGP SIGNATURE-----
