Definitely +1 to a smoke test (just look for non-failure). Maybe a single test class smoke testing all of v2. Maybe should be in "bats" actually.
On Thu, Jan 9, 2025 at 4:24 AM Colvin Cowie <colvin.cowie....@gmail.com> wrote: > Hi, > > Okay yep, I can probably raise a JIRA and a PR tomorrow. I checked on 9.7 > and 9.4, but looking at the source I expect it's always been this way. > On the testing front, I think it's a good idea to have at least one "smoke" > test on every API just to check that things serialize correctly, even > without doing an exhaustive test - as you say the logic of the code is > mostly exercised via the v1 API. > > Cheers > > > On Wed, 8 Jan 2025 at 16:50, Jason Gerlowski <gerlowsk...@gmail.com> > wrote: > > > Hi Colvin, > > > > Yep - that's a result of the missing '@Inject' annotation. What > > version are you seeing this on? If you're up for it, can you file a > > JIRA ticket with the details and hopefully we can address it in an > > upcoming release? > > > > > I guess that there's no tests for going through the v2 API? > > > > I suspect you're correct. Testing the v2 APIs has been a bit of a > > balancing act that we're still trying to feel out. It's not really > > feasible from either a "test run duration" or a "maintenance" > > perspective to cover v2 as comprehensively as we currently cover v1. > > We've tried to accommodate for this by structuring things so that the > > v1 APIs make use of the v2 code, under the theory that tests using v1 > > APIs will be testing the v2 code "implicitly". But that doesn't catch > > everything - as you've found with this Jersey dispatching bug. > > > > Maybe we should reconsider the testing strategy here, and try to > > ensure that all v2 APIs are called in at least one test. > > > > Best, > > > > Jason > > > > On Wed, Jan 8, 2025 at 10:25 AM Colvin Cowie <colvin.cowie....@gmail.com > > > > wrote: > > > > > > Hello, I'm trying to use the v2 > > > > > > https://solr.apache.org/guide/solr/latest/deployment-guide/collection-management.html#collectionprop > > > API. v1 works fine, but v2 fails with a glassfish error. I won't paste > > the > > > full stacktrace here because it's mostly noise. > > > > > > I assume it's because > > org.apache.solr.handler.admin.api.CollectionProperty > > > doesn't have an @Inject annotation on its constructor > > > < > > > https://github.com/apache/solr/blob/main/solr/core/src/java/org/apache/solr/handler/admin/api/CollectionProperty.java#L41 > > > > > > - and I guess that there's no tests for going through the v2 API? Or is > > it > > > tested, and I'm missing something? > > > > > > ---- > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > *2025-01-08 14:29:13.089 WARN (qtp1429351083-106-localhost-1) [c: s: > > r: > > > > x: t:localhost-1] o.g.j.s.ServerRuntime$Responder An exception > mapping > > did > > > > not successfully produce and processed a response. Logging the > > exception > > > > propagated to the default exception mapper. => MultiException stack 1 > > of > > > > 1java.lang.NoSuchMethodException: Could not find a suitable > > constructor in > > > > org.apache.solr.handler.admin.api.CollectionProperty > > > > class.org.glassfish.hk2.api.MultiException: A MultiException has 1 > > > > exceptions. They are:1. java.lang.NoSuchMethodException: Could not > > find a > > > > suitable constructor in > > > > org.apache.solr.handler.admin.api.CollectionProperty class. at > > > > org.jvnet.hk2.internal.Collector.throwIfErrors(Collector.java:65) > > > > ~[?:?]...Caused by: java.lang.NoSuchMethodException: Could not find a > > > > suitable constructor in > > > > org.apache.solr.handler.admin.api.CollectionProperty class. at > > > > > > > org.glassfish.jersey.inject.hk2.JerseyClassAnalyzer.getConstructor(JerseyClassAnalyzer.java:168) > > > > ~[?:?] at > > > > org.jvnet.hk2.internal.Utilities.getConstructor(Utilities.java:157) > > ~[?:?] > > > > at org.jvnet.hk2.internal.Utilities.justCreate(Utilities.java:1043) > > ~[?:?] > > > > ... 71 more2025-01-08 14:29:13.106 WARN > > (qtp1429351083-106-localhost-1) > > > > [c: s: r: x: t:localhost-1] o.g.j.i.Errors The following warnings > have > > been > > > > detected: WARNING: HK2 service reification failed for > > > > [org.apache.solr.handler.admin.api.CollectionProperty] with an > > > > exception:MultiException stack 1 of 2java.lang.NoSuchMethodException: > > Could > > > > not find a suitable constructor in > > > > org.apache.solr.handler.admin.api.CollectionProperty class. at > > > > > > > org.glassfish.jersey.inject.hk2.JerseyClassAnalyzer.getConstructor(JerseyClassAnalyzer.java:168)...MultiException > > > > stack 2 of 2java.lang.IllegalArgumentException: Errors were > discovered > > > > while reifying SystemDescriptor( > > > > implementation=org.apache.solr.handler.admin.api.CollectionProperty > > > > contracts={org.apache.solr.handler.admin.api.CollectionProperty} > > > > scope=org.glassfish.jersey.process.internal.RequestScoped > qualifiers={} > > > > descriptorType=CLASS descriptorVisibility=NORMAL metadata= rank=0 > > > > loader=null proxiable=null proxyForSameScope=null analysisName=null > > id=137 > > > > locatorId=0 identityHashCode=1799286651 reified=false) at > > > > > > org.jvnet.hk2.internal.SystemDescriptor.reify(SystemDescriptor.java:684) > > > > ...* > > > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: dev-unsubscr...@solr.apache.org > > For additional commands, e-mail: dev-h...@solr.apache.org > > > > >
