rsmith added a comment. In https://reviews.llvm.org/D30946#754103, @schroedersi wrote:
> As an alternative to the current solution, the above-mentioned member > functions could each be extended by a "PrintingContext" argument, which then > contains the variable states. I'd be interested to see how much complexity that adds, if you're prepared to give it a try. > And a bit of clarification: [...] there is already some kind of temporary > suppress scope functionality, but it is called SuppressScope. Yes, I think that's right -- `SuppressScope` isn't really an externally useful "print without scopes" mechanism, it's an internal mechanism / implementation detail used for printing an "inner" AST node when we've already printed a scope for it based on outer sugar (eg, the entity is a type component of a nested name specifier, or the type inside an elaborated type specifier, or the class name in a destructor name). It's not really `SuppressScope`, it's `ScopeWasAlreadyPrintedForWhateverTopLevelThingIAskedYouToPrint` =) https://reviews.llvm.org/D30946 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
