[C++ Patch] PR 50080

2012-10-15 Thread Paolo Carlini
Hi, thus, if I understand correctly the resolution of Core/468 [CD1], we can simplify a bit the parser and just accept these 'template' outside templates. Tested x86_64-linux. Thanks, Paolo. /// /cp 2012-10-15 Paolo Carlini paolo.carl...@oracle.com PR

Re: [C++ Patch] PR 50080

2012-10-15 Thread Jason Merrill
OK. Jason

Re: [C++ Patch] PR 50080

2012-10-15 Thread Jason Merrill
Actually, let's keep the diagnostic when compiling with -pedantic in 98 mode. Jason

Re: [C++ Patch] PR 50080

2012-10-15 Thread Paolo Carlini
On 10/15/2012 07:30 PM, Jason Merrill wrote: Actually, let's keep the diagnostic when compiling with -pedantic in 98 mode. ... too late! ;) So I prepared the below, I'm finishing testing it. Thanks, Paolo. // /cp 2012-10-15 Paolo Carlini paolo.carl...@oracle.com PR

Re: [C++ Patch] PR 50080

2012-10-15 Thread Jason Merrill
On 10/15/2012 11:33 AM, Paolo Carlini wrote: + pedantic cxx_dialect == cxx98) + { + cp_token *token = cp_lexer_peek_token (parser-lexer); + error_at (token-location, + in C++98 %template% (as a disambiguator) is only + allowed