I will do as recommended by Richard. I'm aware of the bug, the comment in my 2nd patch references the PR. Actually, it's by reducing the faulty code that I found out it was due to the extern template instantiation.
But I see someone has submitted a patch addressing this PR (partially), so I will remove the check for the main file in my new patch. "Vane, Edwin" <[email protected]> writes: > I agree with Richard on this one. Just add the template case to > cpp11-migrate code. > > By the way, your patch actually fixes another known bug assigned to > someone else: http://llvm.org/bugs/show_bug.cgi?id=15827 > > From: [email protected] > [mailto:[email protected]] On Behalf Of Richard Smith > Sent: Sunday, April 28, 2013 5:09 PM > To: Guillaume Papin > Cc: cfe commits > Subject: Re: [PATCH] Add CXXMethodDecl::getInlineBody() and fix > cpp11-migrate segfault > > I'm not convinced this is the right approach: getInlineBody has weird > semantics and thus it doesn't seem like a good fit as a general > purpose AST interface. It would make a lot more sense to me to have > the fallback to the template in the cpp11-migrate code, where it can > also handle other consequences of this, such as classifying such a fix > as "risky" (because it might not be correct for every instantiation). > > On Sun, Apr 28, 2013 at 7:21 AM, Guillaume Papin > <[email protected]> wrote: > > Uh, embarrassing, of course I failed the attachments: > > -- > Guillaume Papin > > _______________________________________________ > cfe-commits mailing list > [email protected] > http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits > > -- Guillaume Papin _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
