omtcyfz added inline comments.
================
Comment at: clang-rename/RenamingAction.cpp:74
@@ +73,3 @@
+ // FIXME: As for clang-rename, adding a replacement fails IFF either the
+ // AST node has been matched multiple times (which shouldn't happen in
+ // reality, need to fix that). For now, just ignore the error as it
----------------
alexshap wrote:
> omtcyfz wrote:
> > alexshap wrote:
> > > Let's consider the following example:
> > > src/include/Point.h:
> > > struct Point {};
> > > src/a.cpp:
> > > include <Point.h>
> > > src/b.cpp:
> > > include <Point.h>
> > > clang-rename -qualified-name Point -new-name Point2 srcs/a.cpp srcs/b.cpp
> > > Renaming failed in /Users/Alexshap/PlayRename/srcs/./include/Point.h! New
> > > replacement:
> > > /Users/Alexshap/PlayRename/srcs/./include/Point.h: 7:+5:"Point2"
> > > conflicts with existing replacement:
> > > /Users/Alexshap/PlayRename/srcs/./include/Point.h: 7:+5:"Point2"
> > >
> > > The thing is that clang-rename is trying to modify the same code twice
> > > (as in the example) and the return value (Error) of the method
> > > Replacements::add doesn't allow us to distinguish two cases: A. conflict
> > > (trying to apply different modifications to the same source code) B.
> > > (still conflict, but different) (trying to apply the same modification
> > > twice).
> > > In the past when Replacements was a typedef on std::set and clang-rename
> > > was using insert(...) the case B wasn't an issue.
> > > P.S. However (imo) the new (FIXME) comment seems to be a little bit
> > > misleading.
> > But the thing is that it's never the A. case. Unlike other refactorings,
> > clang-rename doesn't introduce any name conflicts, it will only sometimes
> > try to rename something multiple times.
> Yup
Hence, I'm not sure what your concern is. Can you please elaborate?
https://reviews.llvm.org/D24914
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
