Re: c++/7939: [3.3/3.4 regression] [2003-05-03] ICE on function template specialization

Re: c++/7939: [3.3/3.4 regression] [2003-05-03] ICE on function template specialization

[Giovanni Bajo]

The following reply was made to PR c++/7939; it has been noted by GNATS.

From: "Giovanni Bajo" <giovannibajo@libero.it>
To: <gcc-gnats@gcc.gnu.org>,
	<kparz@iastate.edu>,
	<gcc-bugs@gcc.gnu.org>,
	<nobody@gcc.gnu.org>,
	<gcc-prs@gcc.gnu.org>
Cc:  
Subject: Re: c++/7939: [3.3/3.4 regression] [2003-05-03] ICE on function template specialization
Date: Mon, 5 May 2003 02:46:22 +0200

 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=7939
 
 This is a regression from 2.95. Currently, we get the following on mainline
 and 3.3:
 
 pr7939.cpp:2: error: `<anonymous>' has incomplete type
 pr7939.cpp:2: error: invalid use of `void'
 pr7939.cpp:2: error: partial specialization `foo<T, void>' of function
 template
 pr7939.cpp:2: internal compiler error: in comptypes, at cp/typeck.c:913
 Please submit a full bug report,
 
 while we used to get this on 2.95:
 
 pr7939.cpp:2: template-id `foo<T, void>' in declaration of primary template
 
 So, there is actually an improvement on the error message (the parser is
 able to understand that the code is trying to define a partial
 specialization of a function template, which is indeed a great diagnostic),
 but then it ICEs during error recovery. I think the error message could also
 be a little epurated (removing those other lines if possible, especially the
 one with <anonymous> which is really cryptic), and maybe specifying
 explicitally that "partial specialization of function template is not
 allowed by ISO C++".
 
 Giovanni Bajo