On Sun, 2 Feb 2014, Marc Glisse wrote: > An alternative could be to have a helper function that does nothing in C and > calls default_conversion in C++. Or make the C version of default_conversion > return its argument unchanged instead of asserting when it sees an unexpected > tree.
Well, in principle I'm dubious about the cases where different implementations of functions with the same name are used in c-family code; I'd prefer an actual C/C++ set of langhooks, with clearly defined semantics based on what the c-family users need, where the hook used (c_family_lang_hooks.convert_attribute_argument or whatever) would indeed do nothing for C. But more than just default_conversion is involved there, and default_conversion is already used from c-family code, so this isn't an objection to this patch. -- Joseph S. Myers jos...@codesourcery.com
