public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jason Merrill <jason@redhat.com>
To: Marek Polacek <polacek@redhat.com>,
	GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH] c++: fix parsing with auto(x) at block scope [PR112482]
Date: Thu, 14 Dec 2023 16:07:59 -0500	[thread overview]
Message-ID: <b7a5e767-0990-4a35-a900-33071930fa18@redhat.com> (raw)
In-Reply-To: <20231214210227.1190177-1-polacek@redhat.com>

On 12/14/23 16:02, Marek Polacek wrote:
> Bootstrapped/regtested on x86_64-pc-linux-gnu, ok for trunk?

OK.

> -- >8 --
> This is sort of like r14-5514, but at block scope.  Consider
> 
>    struct A { A(int, int); };
>    void
>    g (int a)
>    {
>      A bar(auto(a), 42); // not a fn decl
>    }
> 
> where we emit error: 'auto' parameter not permitted in this context
> which is bogus -- bar doesn't declare a function, so the auto is OK,
> but we don't know it till we've seen the second argument.  The error
> comes from grokdeclarator invoked just after we've parsed the auto(a).
> 
> A possible approach seems to be to delay the auto parameter checking
> and only check once we know we are indeed dealing with a function
> declaration.  For tparms, we should still emit the error right away.
> 
> 	PR c++/112482
> 
> gcc/cp/ChangeLog:
> 
> 	* decl.cc (grokdeclarator): Do not issue the auto parameter error while
> 	tentatively parsing a function parameter.
> 	* parser.cc (cp_parser_parameter_declaration_clause): Check it here.
> 
> gcc/testsuite/ChangeLog:
> 
> 	* g++.dg/cpp23/auto-fncast15.C: New test.
> ---
>   gcc/cp/decl.cc                             | 13 +++++++++++--
>   gcc/cp/parser.cc                           | 17 +++++++++++++++--
>   gcc/testsuite/g++.dg/cpp23/auto-fncast15.C | 21 +++++++++++++++++++++
>   3 files changed, 47 insertions(+), 4 deletions(-)
>   create mode 100644 gcc/testsuite/g++.dg/cpp23/auto-fncast15.C
> 
> diff --git a/gcc/cp/decl.cc b/gcc/cp/decl.cc
> index 4d17ead123a..1ffe4c82748 100644
> --- a/gcc/cp/decl.cc
> +++ b/gcc/cp/decl.cc
> @@ -14203,6 +14203,7 @@ grokdeclarator (const cp_declarator *declarator,
>         tree auto_node = type_uses_auto (type);
>         if (auto_node && !(cxx_dialect >= cxx17 && template_parm_flag))
>   	{
> +	  bool err_p = true;
>   	  if (cxx_dialect >= cxx14)
>   	    {
>   	      if (decl_context == PARM && AUTO_IS_DECLTYPE (auto_node))
> @@ -14221,13 +14222,21 @@ grokdeclarator (const cp_declarator *declarator,
>   			    "abbreviated function template");
>   		  inform (DECL_SOURCE_LOCATION (c), "%qD declared here", c);
>   		}
> -	      else
> +	      else if (decl_context == CATCHPARM || template_parm_flag)
>   		error_at (typespec_loc,
>   			  "%<auto%> parameter not permitted in this context");
> +	      else
> +		/* Do not issue an error while tentatively parsing a function
> +		   parameter: for T t(auto(a), 42);, when we just saw the 1st
> +		   parameter, we don't know yet that this construct won't be
> +		   a function declaration.  Defer the checking to
> +		   cp_parser_parameter_declaration_clause.  */
> +		err_p = false;
>   	    }
>   	  else
>   	    error_at (typespec_loc, "parameter declared %<auto%>");
> -	  type = error_mark_node;
> +	  if (err_p)
> +	    type = error_mark_node;
>   	}
>   
>         /* A parameter declared as an array of T is really a pointer to T.
> diff --git a/gcc/cp/parser.cc b/gcc/cp/parser.cc
> index 58e910d64af..e4fbab1bab5 100644
> --- a/gcc/cp/parser.cc
> +++ b/gcc/cp/parser.cc
> @@ -25102,8 +25102,21 @@ cp_parser_parameter_declaration_clause (cp_parser* parser,
>        committed yet, nor should we.  Pushing here will detect the error
>        of redefining a parameter.  */
>     if (cp_lexer_next_token_is (parser->lexer, CPP_CLOSE_PAREN))
> -    for (tree p : pending_decls)
> -      pushdecl (p);
> +    {
> +      for (tree p : pending_decls)
> +	pushdecl (p);
> +
> +      /* Delayed checking of auto parameters.  */
> +      if (!parser->auto_is_implicit_function_template_parm_p
> +	  && cxx_dialect >= cxx14)
> +	for (tree p = parameters; p; p = TREE_CHAIN (p))
> +	  if (type_uses_auto (TREE_TYPE (TREE_VALUE (p))))
> +	    {
> +	      error_at (location_of (TREE_VALUE (p)),
> +			"%<auto%> parameter not permitted in this context");
> +	      TREE_TYPE (TREE_VALUE (p)) = error_mark_node;
> +	    }
> +    }
>   
>     /* Finish the parameter list.  */
>     if (!ellipsis_p)
> diff --git a/gcc/testsuite/g++.dg/cpp23/auto-fncast15.C b/gcc/testsuite/g++.dg/cpp23/auto-fncast15.C
> new file mode 100644
> index 00000000000..deb1efcc46c
> --- /dev/null
> +++ b/gcc/testsuite/g++.dg/cpp23/auto-fncast15.C
> @@ -0,0 +1,21 @@
> +// PR c++/112482
> +// { dg-do compile { target c++23 } }
> +// { dg-options "-Wno-vexing-parse" }
> +
> +void foo (auto i, auto j);
> +
> +struct A {
> +   A(int,int);
> +};
> +
> +void
> +g (int a)
> +{
> +  A b1(auto(42), auto(42));
> +  A b2(auto(a), auto(42));
> +  A b3(auto(42), auto(a));
> +  A b4(auto(a), // { dg-error "13:'auto' parameter" }
> +       auto(a2)); // { dg-error "13:'auto' parameter" }
> +  int v1(auto(42));
> +  int fn1(auto(a)); // { dg-error "16:'auto' parameter" }
> +}
> 
> base-commit: 767e2674875139ac8f354ceee655c1a9561b9779


      reply	other threads:[~2023-12-14 21:08 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-14 21:02 Marek Polacek
2023-12-14 21:07 ` Jason Merrill [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b7a5e767-0990-4a35-a900-33071930fa18@redhat.com \
    --to=jason@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=polacek@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).