> Unfortunately it looks like the canonical definition is not sufficient > for function templates. I don't fully understand our representation of > function templates, so it is possible this is just hiding a bug > somewhere. If not, the attached patch fixes the included testcase. Let > me know what do you think.
Douglas Gregor suggested using D->getMostRecentDecl()->getPreviousDecl(). An updated patch is attached. Cheers, Rafael
t.patch
Description: Binary data
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
