| Re: [cdt-dev] false codan error |
|
Sorry Nate,
I'm an idiot. For the umpteenth time I've been fooled by the CUTE plugin. @Thomas that's something for you then.
bye Michi
On Friday 20 Sep 2013 16:07:51 Nathan Ridge wrote: > It sounds like a bug, but we need a testcase that reproduces it to fix it. The code snippet you posted doesn't reproduce the error as-is. > > Regards, > Nate > > From: scalpel4k@xxxxxxxxx > To: cdt-dev@xxxxxxxxxxx > Date: Fri, 20 Sep 2013 15:18:20 +0200 > Subject: [cdt-dev] false codan error > > > > Hi guys, > > I've come across some error in the codan checker, but I don't know how to describe the problem properly in a bug report. > > I have the following code excerpt that compiles/runs fine when I use it from my code (I know this example can't work, I simply tried to focus on the problem). > > template<typename _Handler_> > struct TestWrapper: > public _Handler_ { > > using H = _Handler_; > > typename _Handler_::Buffer _buf; // no error > > static void testHandler() { > > typename H::Buffer buf1(); // no error > > typename H::Buffer buf2; // error: No such constructor for type '#0::Buffer' > > } > > }; > > The error is shown even if there's nothing refering to this declaration, so even when there are no specializations/uses anywhere in the code. I think the parser should not be able to tell me anything about Buffer in the first place. And then I don't know where that error itself comes from; in the real code there is an appropriate ctor. > > Does anybody have an idea what is happening here? > > bye Michi > _______________________________________________ > cdt-dev mailing list > cdt-dev@xxxxxxxxxxx > https://dev.eclipse.org/mailman/listinfo/cdt-dev |