|Re: [jakartaee-tck-dev] [glassfish-dev] 1 failure remaining: App client signature test|
We would have to look into this specifically, especially to see how easy it is to replace the context in weld-se-shaded. Hopefully we can look at this tomorrow.
Alternatively maybe Weld can look at an update as well? I guess this same change is needed for WildFly and potentially OpenLiberty too.
Thanks, https://github.com/eclipse-ee4j/glassfish/issues/24014 is the tracking issue for GlassFish.
On Tue, Jun 21, 2022 at 7:08 PM Scott Marlow <smarlow@xxxxxxxxxx> wrote:
On 6/21/22 12:32 PM, Scott Marlow wrote:
On 6/21/22 10:55 AM, Scott Marlow wrote:
Hmm, I don't see the failure due to signature test tool bug https://github.com/jtulach/netbeans-apitest/pull/12 not being fixed in a release yet. :(
I'm guessing that the updated jakarta.annotation classes are not on the appclient container classpath.
The appclient container has an older version of jakarta.annotation classes via glassfish/lib/appclient/weld-se-shaded.jar:
grep -rli jakarta.annotation --include=*.jar
Should the GlassFish appclient container be updated to prefer the glassfish/modules/jakarta.annotation-api.jar over the glassfish/lib/appclient/weld-se-shaded.jar contents?
It looks like we are using the latest released version of weld-se-shaded.jar which contains jakarta.annotation classes from last December I think.
On 6/21/22 10:40 AM, arjan tijms wrote:
The platform TCK seems to have only one failure remaining with GF 7 nightly:
Is anyone already looking into this?
Kind regards,Arjan Tijms
_______________________________________________ glassfish-dev mailing list glassfish-dev@xxxxxxxxxxx To unsubscribe from this list, visit https://www.eclipse.org/mailman/listinfo/glassfish-dev
Back to the top