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++: Improve diagnostic for class tmpl/class redecl [PR103749]
Date: Thu, 16 Dec 2021 20:13:01 -0500	[thread overview]
Message-ID: <86ae73a7-0df1-1082-7d67-f8aa89a3cd1e@redhat.com> (raw)
In-Reply-To: <20211216223659.820883-1-polacek@redhat.com>

On 12/16/21 17:36, Marek Polacek wrote:
> For code like
> 
>    template<typename>
>    struct bar;
> 
>    struct bar {
>      int baz;
>    };
> 
>    bar var;
> 
> we emit a fairly misleading and unwieldy diagnostic:
> 
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> $ g++ -c u.cc
> u.cc:6:8: error: template argument required for 'struct bar'
>      6 | struct bar {
>        |        ^~~
> u.cc:10:5: error: class template argument deduction failed:
>     10 | bar var;
>        |     ^~~
> u.cc:10:5: error: no matching function for call to 'bar()'
> u.cc:3:17: note: candidate: 'template<class> bar()-> bar< <template-parameter-1-1> >'
>      3 |   friend struct bar;
>        |                 ^~~
> u.cc:3:17: note:   template argument deduction/substitution failed:
> u.cc:10:5: note:   couldn't deduce template parameter '<template-parameter-1-1>'
>     10 | bar var;
>        |     ^~~
> u.cc:3:17: note: candidate: 'template<class> bar(bar< <template-parameter-1-1> >)-> bar< <template-parameter-1-1> >'
>      3 |   friend struct bar;
>        |                 ^~~
> u.cc:3:17: note:   template argument deduction/substitution failed:
> u.cc:10:5: note:   candidate expects 1 argument, 0 provided
>     10 | bar var;
>        |     ^~~
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 
> but with this patch we get:
> 
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> z.C:6:8: error: 'bar' was declared as template but no template header provided
>      6 | struct bar {
>        |        ^~~
> z.C:3:17: note: previous declaration here
>      3 |   friend struct bar;
>        |                 ^~~
> z.C:10:5: error: 'bar<...auto...> var' has incomplete type
>     10 | bar var;
>        |     ^~~
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 
> which is clearer about what the problem is.
> 
> I thought it'd be nice to avoid printing the messages about failed CTAD,
> too.  To that end, I'm using CLASSTYPE_ERRONEOUS to suppress CTAD.  Not
> sure if that's entirely kosher.

A bit dubious, but error-recovery often is.  Let's go with it.

> The other direction (first a non-template class declaration followed by
> a class template definition) we handle quite well:
> 
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> z.C:11:8: error: 'bar' is not a template
>     11 | struct bar {};
>        |        ^~~
> z.C:8:8: note: previous declaration here
>      8 | struct bar;
>        |        ^~~
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 
> Bootstrapped/regtested on x86_64-pc-linux-gnu, ok for trunk?
> 
> 	PR c++/103749
> 
> gcc/cp/ChangeLog:
> 
> 	* decl.c (lookup_and_check_tag): Give an error when a class was
> 	declared as template but no template header has been provided.
> 	* pt.c (do_class_deduction): Don't deduce CLASSTYPE_ERRONEOUS
> 	types.
> 
> gcc/testsuite/ChangeLog:
> 
> 	* g++.dg/template/redecl4.C: Adjust dg-error.
> 	* g++.dg/diagnostic/redeclaration-2.C: New test.
> ---
>   gcc/cp/decl.c                                 | 11 +++++++++++
>   gcc/cp/pt.c                                   |  7 +++++++
>   .../g++.dg/diagnostic/redeclaration-2.C       | 19 +++++++++++++++++++
>   gcc/testsuite/g++.dg/template/redecl4.C       |  2 +-
>   4 files changed, 38 insertions(+), 1 deletion(-)
>   create mode 100644 gcc/testsuite/g++.dg/diagnostic/redeclaration-2.C
> 
> diff --git a/gcc/cp/decl.c b/gcc/cp/decl.c
> index 7ca8770bd02..56fa5b3ffeb 100644
> --- a/gcc/cp/decl.c
> +++ b/gcc/cp/decl.c
> @@ -15438,6 +15438,17 @@ lookup_and_check_tag (enum tag_types tag_code, tree name,
>         return error_mark_node;
>       }
>   
> +  if (DECL_CLASS_TEMPLATE_P (decl)
> +      && !template_header_p
> +      && how == TAG_how::CURRENT_ONLY)
> +    {
> +      error ("%qD was declared as template but no template header provided",
> +	     name);

Maybe "class template %qD redeclared as non-template" ?

OK with that change.

> +      inform (location_of (decl), "previous declaration here");
> +      CLASSTYPE_ERRONEOUS (TREE_TYPE (decl)) = true;
> +      return error_mark_node;
> +    }
> +
>     if (DECL_CLASS_TEMPLATE_P (decl)
>         /* If scope is TAG_how::CURRENT_ONLY we're defining a class,
>   	 so ignore a template template parameter.  */
> diff --git a/gcc/cp/pt.c b/gcc/cp/pt.c
> index 18c6f118ae6..4f0ae6d5851 100644
> --- a/gcc/cp/pt.c
> +++ b/gcc/cp/pt.c
> @@ -29601,6 +29601,13 @@ do_class_deduction (tree ptype, tree tmpl, tree init,
>     if (DECL_TEMPLATE_TEMPLATE_PARM_P (tmpl))
>       return ptype;
>   
> +  /* If the class was erroneous, don't try to deduce, because that
> +     can generate a lot of diagnostic.  */
> +  if (TREE_TYPE (tmpl)
> +      && TYPE_LANG_SPECIFIC (TREE_TYPE (tmpl))
> +      && CLASSTYPE_ERRONEOUS (TREE_TYPE (tmpl)))
> +    return ptype;
> +
>     /* Wait until the enclosing scope is non-dependent.  */
>     if (DECL_CLASS_SCOPE_P (tmpl)
>         && dependent_type_p (DECL_CONTEXT (tmpl)))
> diff --git a/gcc/testsuite/g++.dg/diagnostic/redeclaration-2.C b/gcc/testsuite/g++.dg/diagnostic/redeclaration-2.C
> new file mode 100644
> index 00000000000..ac14c2c135f
> --- /dev/null
> +++ b/gcc/testsuite/g++.dg/diagnostic/redeclaration-2.C
> @@ -0,0 +1,19 @@
> +// PR c++/103749
> +
> +struct foo {
> +  template<typename>
> +  friend struct bar;
> +};
> +
> +struct bar { // { dg-error "declared as template" }
> +  int baz;
> +};
> +
> +template<typename>
> +struct T; // { dg-message "previous" }
> +
> +struct T { // { dg-error "declared as template" }
> +};
> +
> +bar var; // { dg-error "" }
> +T t; // { dg-error "" }
> diff --git a/gcc/testsuite/g++.dg/template/redecl4.C b/gcc/testsuite/g++.dg/template/redecl4.C
> index 5638bde413d..498ca83a0f4 100644
> --- a/gcc/testsuite/g++.dg/template/redecl4.C
> +++ b/gcc/testsuite/g++.dg/template/redecl4.C
> @@ -2,4 +2,4 @@
>   // { dg-do compile }
>   
>   template<int> union A;  // { dg-message "previous" }
> -struct A;               // { dg-error "non-template" }
> +struct A;               // { dg-error "declared as template" }
> 
> base-commit: f91814c22595e1db642140efe030caf2c092ab6f


  reply	other threads:[~2021-12-17  1:13 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-16 22:36 Marek Polacek
2021-12-17  1:13 ` Jason Merrill [this message]
2021-12-17 18:08   ` [PATCH v2] " Marek Polacek

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=86ae73a7-0df1-1082-7d67-f8aa89a3cd1e@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).