This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.
Build: NetBeans IDE Dev (Build 110308-a13d1e436b6d) VM: Java HotSpot(TM) 64-Bit Server VM, 19.0-b09, Java(TM) SE Runtime Environment, 1.6.0_23-b05 OS: Linux User Comments: marfous: 1, created bean with method 2, wanted to generate JUnit tests 3, invoke on the file Tools->Generate Junit -> exception appears Stacktrace: java.lang.NoSuchMethodError: org.netbeans.modules.junit.Bundle.junitlib_confirm_text()Ljava/lang/String; at org.netbeans.modules.junit.DefaultPlugin.storeProjectSettingsJUnitVer(DefaultPlugin.java:1431) at org.netbeans.modules.junit.DefaultPlugin.createTestActionCalled(DefaultPlugin.java:1145) at org.netbeans.modules.junit.plugin.JUnitPlugin$1.createTestActionCalled(JUnitPlugin.java:77) at org.netbeans.modules.junit.CreateTestAction.performAction(CreateTestAction.java:296) at org.openide.util.actions.NodeAction$DelegateAction$1.run(NodeAction.java:586) at org.openide.util.actions.ActionInvoker$1.run(ActionInvoker.java:95)
Created attachment 106841 [details] stacktrace
Increasing priority - it's impossible to generate JUnits now. Definitely showstopper.
Created attachment 106842 [details] patch for fixing this
It seems that it could fix really easy patch because there is just unfinished HTML tag in the bundle. I'm proposing to mark the issue as 70_HR_FIX_CANDIDATE, do you agree?
Yes, all P2s in the "before cut-off" column on the bug dashboard should automatically be 70_HR_FIX_CANDIDATE.
Attaching a patch. Will be integrated as soon as core-main is open.
Created attachment 106853 [details] Patch
Patch doesn't make any sense to me. Perhaps the reporter was just using an incorrectly built module JAR?
Sorry, you are right, in the newest build works well also with the unfinished tag. Strange to me is how it knew new NbBundle.Messages API then, but as a probable reason sounds to me that I forgot after getting new HG changes rebuild so it was rebuild once I made changes. Sorry for invalid issue.
v/c
just happend to me, see my last submit. I am using IDE freshly built from core-main AFAIK i did: hg pull -u ant which should not generate any invalid JARs?! Or did i miss something?!
Copy-paste of my steps to reproduce from report: - custom core-main build - IDE did not contain JUnit, opened previously created project which contained reference to JUnit - ignored "Resolve reference problem", closed project - opened another one, without test - tried ti invoke on some class Tools|Create JUnit test --> entry point did not work, got NSME please note, that it is problem in this entry point, another entry point works. While creating of JUnit tests via Tools| Create JUnit failed with NSME, i was able to install JUnit via Resolve broken references later without any problems (after opening project containing references to tests).
(In reply to comment #11) > I am using IDE freshly built from core-main AFAIK i did: > > hg pull -u > ant In general you should use the sequence: ant clean hg pull -u (or hg fetch) ant Otherwise it is not "freshly built". That said, if you can find a specific test case which reproducibly causes the Bundle.class method to be missing after an incremental build, file a P3 for me in platform/Other blocking bug #192750.
you are right, unable to reproduce now. Closing.