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 200806160002) VM: Java HotSpot(TM) Client VM, 10.0-b22, Java(TM) SE Runtime Environment, 1.6.0_06-b02 OS: Linux, 2.6.22-14-generic, i386 User Comments: trying to open a groovy service from within a grails app
Created attachment 62951 [details] stacktrace
The stacktrace attachment only contains the word "NullPointerException", no actual stacktrace. I'm not sure what's going on here, but since it's related to Groovy and the hints provider API, I'm hoping maybe Matthias can look at this first and either figure out how to reproduce it, or examine the Groovy hints provider to make sure it's doing the right thing, before asssigning it back if the problem is in GSF.
Hi, after opening some 20 services from my grails projects i was not able to reproduce this error. Could you somehow share a testcase with us?
Created attachment 63338 [details] stacktrace
Hi "cfrias", Thanks for the stacktrace. This looks much better now. @Tor: I think we should put a null guard here (sfret/hints/infrastructure/SuggestionsTask.java): 99 HintsProvider provider = language.getHintsProvider(); 100 assert provider != null; // getHintsProviderLanguage will return null if there's no provider 101 GsfHintsManager manager = language.getHintsManager(); 102 RuleContext ruleContext = manager.createRuleContext(info, language, pos, -1, -1); Could you take a look why language.getHintsManager() returns null?
Hi Tor, could you take a look at this issue? Under some circumstances language.getHintsManager() returns null. Although I wasn't able to reproduce this case, i would place a guard against null in line # 102.
There is something wrong going on here. If you look at language.getHintsProvider(), it should not be possible to call language.getHintsManager() afterwards and get null back. Perhaps it's a race condition? I'm concerned that simply adding a null pointer there is going to push the problem downstream to something trickier to figure out.
I've tried to make initialization safer. Fixed in changeset 73e65900a47d.
reporter, can I ask you to verify this, please? Thanks.
v. Feel free to reopen if this appears again. Thanks.