[Bug fortran/52325] New: unclear error: Unclassifiable statement

[Bug fortran/52325] New: unclear error: Unclassifiable statement

[kloedej at knmi dot nl]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52325

             Bug #: 52325
           Summary: unclear error: Unclassifiable statement
    Classification: Unclassified
           Product: gcc
           Version: 4.6.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: kloedej@knmi.nl


The given example code generates this error:

   unclassifiable.F90:9.4:

       cc%a = 1
       1
   Error: Unclassifiable statement at (1)


while in fact it would be much more helpfull if the error would clearly state
that the symbol is not yet defined (as the commented out lines do):

   unclassifiable.F90:16.5:
       d=1
        1
   Error: Symbol 'd' at (1) has no IMPLICIT type

Especially in larger modules/programs having a clear error in case of this kind
of variable naming msitakes/typos could save a lot of debugging time for
gfortran users.

Example code:

module test_unclassifiable
  implicit none
  type test
     integer :: a
  end type test
contains
  subroutine test_sub1(c)
    type(test), intent(out) :: c
    cc%a = 1
  end subroutine test_sub1
!  subroutine test_sub2(c)
!    integer :: c
!    d=1
!  end subroutine test_sub2
end module test_unclassifiable

[Bug fortran/52325] unclear error: Unclassifiable statement

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52325

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |diagnostic
                 CC|                            |burnus at gcc dot gnu.org

--- Comment #1 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-02-21 11:05:11 UTC ---
Untested patch:

--- primary.c   (revision 184431)
+++ primary.c   (working copy)
@@ -1910,6 +1910,19 @@ gfc_match_varspec (gfc_expr *primary, int equiv_fl
       && gfc_get_default_type (sym->name, sym->ns)->type == BT_DERIVED)
     gfc_set_default_type (sym, 0, sym->ns);

+  if (sym->ts.type == BT_UNKNOWN && gfc_match_char ('%') == MATCH_YES)
+    {
+      gfc_error ("Symbol %s at %C has no IMPLICIT type", sym->name);
+      return MATCH_ERROR;
+    }
+  else if ((sym->ts.type != BT_DERIVED && sym->ts.type != BT_CLASS)
+          && gfc_match_char ('%') == MATCH_YES)
+    {
+      gfc_error ("Unexpected %% for nonderived type variable %s at %C",
+                sym->name);
+      return MATCH_ERROR;
+    }
+
   if ((sym->ts.type != BT_DERIVED && sym->ts.type != BT_CLASS)
       || gfc_match_char ('%') != MATCH_YES)
     goto check_substring;

[Bug fortran/52325] unclear error: Unclassifiable statement

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52325

--- Comment #2 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-02-21 15:58:23 UTC ---
Submitted patch (pending review):
  http://gcc.gnu.org/ml/fortran/2012-02/msg00089.html

[Bug fortran/52325] unclear error: Unclassifiable statement

[Joost.VandeVondele at mat dot ethz.ch]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52325

--- Comment #3 from Joost VandeVondele <Joost.VandeVondele at mat dot ethz.ch> 2012-02-22 06:49:41 UTC ---
(In reply to comment #2)
> Submitted patch (pending review):
>   http://gcc.gnu.org/ml/fortran/2012-02/msg00089.html

OK ;-)

this would be a significant improvement. 

I think it is independent, but a better choice for the error message could be
'Symbol %s at %C has an undefined type'. The type could be implicitly or
explicitly defined, that doesn't matter so much. For consistency, I believe
your proposed message is fine.

[Bug fortran/52325] unclear error: Unclassifiable statement

[Joost.VandeVondele at mat dot ethz.ch]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52325

--- Comment #4 from Joost VandeVondele <Joost.VandeVondele at mat dot ethz.ch> 2012-02-22 06:53:09 UTC ---
(In reply to comment #2)
> Submitted patch (pending review):
>   http://gcc.gnu.org/ml/fortran/2012-02/msg00089.html

and a nitpick... it should be 'non-derived type' instead on 'nonderived type'
(unless I got this with the hyphens wrong again).

[Bug fortran/52325] unclear error: Unclassifiable statement

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52325

--- Comment #5 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-03-02 13:21:00 UTC ---
Author: burnus
Date: Fri Mar  2 13:20:52 2012
New Revision: 184785

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=184785
Log:
2012-03-02  Tobias Burnus  <burnus@net-b.de>

        PR fortran/52325
        * primary.c (gfc_match_varspec): Add diagnostic for % with
        nonderived types.

2012-03-02  Tobias Burnus  <burnus@net-b.de>

        PR fortran/52325
        * gfortran.dg/derived_comp_array_ref_8.f90: New.
        * gfortran.dg/nullify_2.f90: Update dg-error.
        * gfortran.dg/nullify_4.f90: Ditto.
        * gfortran.dg/pointer_init_6.f90: Ditto.


Added:
    trunk/gcc/testsuite/gfortran.dg/derived_comp_array_ref_8.f90
Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/primary.c
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/gfortran.dg/nullify_2.f90
    trunk/gcc/testsuite/gfortran.dg/nullify_4.f90
    trunk/gcc/testsuite/gfortran.dg/pointer_init_6.f90

[Bug fortran/52325] unclear error: Unclassifiable statement

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52325

Tobias Burnus <burnus at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |FIXED

--- Comment #6 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-03-02 13:36:35 UTC ---
FIXED on the trunk (i.e. GCC 4.8).

It had bothered me already a couple of times - your PR report finally nudged me
to fix it.

Thanks for the bug report.

[Bug fortran/52325] unclear error: Unclassifiable statement

[kloedej at knmi dot nl]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52325

--- Comment #7 from Jos de Kloe <kloedej at knmi dot nl> 2012-03-02 13:50:59 UTC ---
Thanks for your (really) fast response and fix.
I'll keep my eye open for other details that might improve gfortran.

[Bug fortran/52325] unclear error: Unclassifiable statement

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52325

--- Comment #8 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-03-02 14:59:05 UTC ---
Author: burnus
Date: Fri Mar  2 14:58:58 2012
New Revision: 184790

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=184790
Log:
2012-03-02  Tobias Burnus  <burnus@net-b.de>

        PR fortran/52325
        * primary.c (gfc_match_varspec): Add missing ;.


Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/primary.c