Sorry, thought I had responded to this. OK.
Jason
Hi,
On 04/07/2015 05:19 PM, Paolo Carlini wrote:
... thus I prepared the below. I tried to minimize the diagnostic
changes but a few more testcases needed tweaking anyway.
Is this version Ok for the trunk?
https://gcc.gnu.org/ml/gcc-patches/2015-04/msg00260.html
Thanks,
Paolo.
... thus I prepared the below. I tried to minimize the diagnostic
changes but a few more testcases needed tweaking anyway.
Thanks!
Paolo.
///
Index: cp/call.c
===
--- cp/call.c (revision 221871)
+++ cp/call.c
Hi,
On 04/06/2015 04:07 PM, Jason Merrill wrote:
On 04/04/2015 09:00 AM, Paolo Carlini wrote:
error ("%q+#D is inaccessible", diag_decl);
- error ("within this context");
+ inform (input_location, "within this context");
This is the wrong error to change to inform; we want
On 04/04/2015 09:00 AM, Paolo Carlini wrote:
error ("%q+#D is inaccessible", diag_decl);
- error ("within this context");
+ inform (input_location, "within this context");
This is the wrong error to change to inform; we want the error to be
associated with the point
Hi,
I noticed a while ago that for each access control issue we emit two
errors instead of an error + an inform, as we lately try to consistently
do. I also noticed that the testsuite reflects that in many testcases...
Anyway, the below - for next Stage 1? - implements the change, passes
test
