Community
Participate
Working Groups
R3.3 (I20070625-1500). Sometime when the build path is not correct I get errors reported when I open the Java editor but there is/was no corresponding problem marker i.e. after building the project there is not error in the Problems view but when I open the Java editor I get errors reported inside the editor. A similar problem is currently discussed on the newsgroups: http://dev.eclipse.org/newslists/news.eclipse.tools.jdt/msg20813.html While I cannot reproduce exactly that I can reproduce at least additional errors being reported and some strange Problem marker locations: 1. start fresh workspace with R3.3 2. disable (deselect) all plug-ins on the 'Plug-in Development' > 'Target Platform' 3. load 'pre_R3_3' version of 'platform-text' module from CVS 4. import all required plug-ins as binary plug-ins (including fragments) ==> no errors 5. now import 'org.eclipse.core.commands' as binary plug-in from an R3.2 drop (replace the existing one) ==> - PDE BUG: it does not flag org.eclipse.jface and/or jface.text with an error which it should because the required 3.3 version of commands is missing - JDT Core correctly creates two build path problem markers (EventManager missing) - JDT Core BUG: it creates a marker on TaskRulerAction at a strange location (inside Javadoc) 6. open TaskRulerAction ==> besides the reported Problem marker two more temporary problems get reported about inconsistent hierarchies. I would expect that markers and temporary problems are the same as long as the file hasn't been changed.
Also, the message 'Project was not built...' is wrong as some files got error markers i.e. something obviously got built. The message should be changed to something like 'Project was not fully built....' or 'The project has an incomplete build path: ...'.
Might be related to bug 194432, where we don't insert working copies at the right place on the classpath during NameLookup.
Bug 194432 is now fixed. Dani can you reproduce with the latest I-build ?
I'd claim that my steps are simple enough for you guys to execute them yourselves - after all that's the idea of providing a test case, isn't it? Anyway, I did it once again and all reported issues are still there.
Sorry, the words "Sometime" and "While I cannot reproduce exactly" confused me into thinking that this bug was not easily reproducible. I'll try your steps.
>I'll try your steps. Thanks. Actually this should allow you to debug it via your dev workspace.
Running runtime workbench on a new workspace I see issue #5 but not issue #6. I'm going to try with the SDK on a new workspace.
Still cannot reproduce with I20070821-0800. TaskRulerAction has 1 annotation for "The type org.eclipse.core.commands.common.EventManager cannot be resolved. It is indirectly referenced from required .class files", but no other annotation.
Did you open the CU? The errors only appear as temporary problems. Should be visible via Overview ruler.
Created attachment 76899 [details] Screenshot of editor opened on TaskRulerAction.java Did I miss something ?
[Dani slaps his head on the desk] repeat many times ;-) >6. open TaskRulerAction Should read: open TemplatePreferencePage
When a missing .class file is detected, the Java builder assumes that there is a classpath problem (which is the case here) and removes all other problems (so that the user can concentrate on first fixing the classpath). We should also remove all other problems during reconcile. The Java builder code that removes the markers is in AbstractImageBuilder#storProblemsFor(...). Look for // handle missing classfile situation
This bug hasn't had any activity in quite some time. Maybe the problem got resolved, was a duplicate of something else, or became less pressing for some reason - or maybe it's still relevant but just hasn't been looked at yet. As such, we're closing this bug. If you have further information on the current state of the bug, please add it and reopen this bug. The information can be, for example, that the problem still occurs, that you still want the feature, that more information is needed, or that the bug is (for whatever reason) no longer relevant. -- The automated Eclipse Genie.
