Have you run this warning on some large codebase (test codebases we've used in the past include LLVM itself, Chromium, Firefox, OpenOffice) and measured how many bugs and false positives it causes?
On Fri, Apr 25, 2014 at 3:24 AM, Anders Rönnholm <[email protected]> wrote: > Your right. I think it's a left over from when i first tried to cast it to an > arraysubscriptexpr. > > Here is a new patch with array type removed. > > //Anders > ________________________________________ > Från: Jordan Rose [[email protected]] > Skickat: den 24 april 2014 18:43 > Till: Anders Rönnholm > Cc: [email protected] > Ämne: Re: [PATCH] String plus char > > If we're not looking through implicit casts, when would there ever be a case > when the StringRefExpr is a constant array type? Shouldn't it always decay to > a pointer? > > Jordan > > > On Apr 17, 2014, at 5:19 , Anders Rönnholm <[email protected]> > wrote: > >> Hi, >> >> I have made some improvements on the string plus char warning. >> >> It can now also handle arrays, struct members and return values from >> function calls. >> >> e.q >> >> ab->a + 'b'; >> >> foo() + 'c'; >> >> char *strArr2[] = {"ac","dc"}; >> str = strArr2[0] + 'c'; >> >> //Anders >> <stringpluschar.diff>_______________________________________________ >> cfe-commits mailing list >> [email protected] >> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits > > > _______________________________________________ > cfe-commits mailing list > [email protected] > http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits > _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
