Thank you Marek,Appreciate your valuable feedback on the patch . Attached the latest ,please do let us know your thoughts.
~Umesh
On Thu, Nov 15, 2018 at 9:26 PM Marek Polacek <pola...@redhat.com> wrote:
>
> On Thu, Nov 15, 2018 at 02:26:24PM +0530, Umesh Kalappa wrote:
> > Thank you Marek for the inputs .
> > >>In the future, if using diff, please also use the -p option.
> > We are using svn diif and other comments are addressed .
>
> Thanks, but it doesn't seem like the -p option was used.
>
> > please let us know your take on the revised attached patch .
>
>
> Index: cp/ChangeLog
> ===================================================================
> --- cp/ChangeLog (revision 266026)
> +++ cp/ChangeLog (working copy)
> @@ -1,3 +1,9 @@
> +2018-11-14 Kamlesh Kumar <kamleshbha...@gmail.com>
> +
> + PR c++/52869
> + *parser.c () : restore the old current_class_{ptr,ref} by
> + inject_this_parameter().
> +
>
> This is still the same; can you adjust it according to my last suggestion?
>
> Index: cp/parser.c
> ===================================================================
> --- cp/parser.c (revision 266026)
> +++ cp/parser.c (working copy)
> @@ -24620,6 +24620,12 @@
> {
> matching_parens parens;
> parens.consume_open (parser);
> +
> + tree save_ccp = current_class_ptr;
> + tree save_ccr = current_class_ref;
> +
>
> Watch out for trailing whitespace in the blank lines.
>
> + if (current_class_type)
> + inject_this_parameter (current_class_type, TYPE_UNQUALIFIED);
>
> I think you can remove the if here.
>
> if (require_constexpr)
> {
> @@ -24640,6 +24646,9 @@
> }
>
> parens.require_close (parser);
> +
> + save_ccp = current_class_ptr = save_ccp;
> + save_ccr = current_class_ref = save_ccr;
>
> You don't need to set save_cc[pr] to itself here.
>
> Index: testsuite/ChangeLog
> ===================================================================
> --- testsuite/ChangeLog (revision 266026)
> +++ testsuite/ChangeLog (working copy)
> @@ -1,3 +1,8 @@
> +2018-11-14 Kamlesh Kumar <kamleshbha...@gmail.com>
> +
> + PR c++/52869
> + * g++.dg//DRs/dr52869.C: New.
> +
>
> So DR != PR. Please name the test dr1207.C
>
> Index: testsuite/g++.dg/DRs/dr52869.C
> ===================================================================
> --- testsuite/g++.dg/DRs/dr52869.C (nonexistent)
> +++ testsuite/g++.dg/DRs/dr52869.C (working copy)
> @@ -0,0 +1,22 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O0 -std=c++11" } */
>
> Instead of this, do:
>
> // { dg-do compile { target c++11 } }
>
> Also, I wrote a test that fails if current_class_{ptr,ref} aren't properly
> restored:
>
> // DR 1207
> // PR c++/52869
> // { dg-do compile { target c++11 } }
>
> void
> fn ()
> {
> struct S {
> bool operator!() noexcept(false);
> } s;
> S t = s;
> }
>
> So you can add that one too, e.g. testsuite/g++.dg/DRs/dr1207-2.C.
>
> Thanks,
> Marek
pr52869.patch
Description: Binary data
