Re: [C++ Patch] PR 51226

2013-05-08 Thread Jason Merrill
OK. Jason

Re: [C++ Patch] PR 51226

2013-05-07 Thread Paolo Carlini
Hi, On 05/07/2013 06:49 PM, Jason Merrill wrote: On 05/07/2013 06:24 AM, Paolo Carlini wrote: I'm not sure if it would be more correct to explicitly cp_parser_abort_tentative_parse before returning Well, there needs to be something to match the cp_parser_parse_tentatively before this function

Re: [C++ Patch] PR 51226

2013-05-07 Thread Jason Merrill
On 05/07/2013 06:24 AM, Paolo Carlini wrote: I'm not sure if it would be more correct to explicitly cp_parser_abort_tentative_parse before returning Well, there needs to be something to match the cp_parser_parse_tentatively before this function returns. But in this case the code is indeed an

[C++ Patch] PR 51226

2013-05-07 Thread Paolo Carlini
Hi, in this error recovery issue, we provide two correct error messages, then a bogus one, and then we ICE because much later in cp_parser_enum_specifier the is_ancestor helper cannot handle error_mark_node as nested_name_specifier: 51226.C:6:19: error: wrong number of template arguments (0,