[Bug fortran/57530] New: [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[Bug fortran/57530] New: [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

            Bug ID: 57530
           Summary: [OOP] Wrongly rejects  type_pointer => class_target
                    (which have identical declared type)
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Keywords: rejects-valid
          Severity: normal
          Priority: P3
         Component: fortran
          Assignee: unassigned at gcc dot gnu.org
          Reporter: burnus at gcc dot gnu.org
                CC: janus at gcc dot gnu.org

Found in the examples to http://www.cambridge.org/rouson (->
ssdSource/chapter05/hermetic). According to comments, the original program is
accepted by xlf2003 and nagf95.

The following simplified code is accepted by Crayftn, ifort and pgf95, but
rejected by gfortran with:

    ptr => tgt  ! TYPE => CLASS
    1
Error: Different types in pointer assignment at (1); attempted assignment of
CLASS(t) to TYPE(t)


>From F2008:

"C714 (R733) If data-target is not unlimited polymorphic, data-pointer-object
shall be type compatible (4.3.1.3) with it and the corresponding kind type
parameters shall be equal."

There (4.3.1.3): "A nonpolymorphic entity is type compatible only with entities
of the same declared type. A polymorphic entity that is not an unlimited
polymorphic entity is type compatible with entities of the same declared type
or any of its extensions."


In the example below, TYPE(t) and CLASS(t) have same declared type.


module m
  type t
  end type t
contains
  subroutine sub (tgt)
    class(t), target :: tgt
    type(t), pointer :: ptr
    ptr => tgt  ! TYPE => CLASS
  end subroutine sub
end module m

[Bug fortran/57530] [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

--- Comment #1 from Tobias Burnus <burnus at gcc dot gnu.org> ---
Patch: http://gcc.gnu.org/ml/fortran/2013-06/msg00045.html

[Bug fortran/57530] [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

--- Comment #2 from Tobias Burnus <burnus at gcc dot gnu.org> ---
(In reply to Tobias Burnus from comment #1)
> Patch: http://gcc.gnu.org/ml/fortran/2013-06/msg00045.html

Sigh, the patch isn't sufficient as the trans*.c code associate the TYPE with
the CLASS container - not its content. Test case below.

! { dg-do run }
!
! PR fortran/57530
!
module m
  implicit none
  type t
    integer :: ii = 55
  end type t
contains
  subroutine sub (tgt, tgt2)
    class(t), target :: tgt, tgt2(:)
    type(t), pointer :: ptr, ptr2(:)

    if (tgt%ii /= 43) call abort()
    if (size (tgt2) /= 3) call abort()
    if (any (tgt2(:)%ii /= [11,22,33])) call abort()

    ptr => tgt  ! TYPE => CLASS
    ptr2 => tgt2  ! TYPE => CLASS

    if (.not. associated(ptr)) call abort()
    if (.not. associated(ptr2)) call abort()
    if (.not. associated(ptr,tgt)) call abort()
    if (.not. associated(ptr2,tgt2)) call abort()
    if (ptr%ii /= 43) call abort()
    if (size (ptr2) /= 3) call abort()
    if (any (ptr2(:)%ii /= [11,22,33])) call abort()
  end subroutine sub
end module m

use m
type(t), target :: x
type(t), target :: y(3)
x%ii = 43
y(:)%ii = [11,22,33]
call sub(x,y)
end

[Bug fortran/57530] [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

--- Comment #3 from Tobias Burnus <burnus at gcc dot gnu.org> ---
Created attachment 30262
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=30262&action=edit
Draft patch - does not fully work, yet

Currently fails for
  x = func1()
in assign_11.f90

[Bug fortran/57530] [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

--- Comment #4 from Tobias Burnus <burnus at gcc dot gnu.org> ---
(In reply to Tobias Burnus from comment #3)
The resolve patch has been approved (but not yet committed):
  http://gcc.gnu.org/ml/fortran/2013-06/msg00049.html

The trans*.c patch has still to be fixed and submitted. (Cf. comment 3)

[Bug fortran/57530] [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

--- Comment #5 from Tobias Burnus <burnus at gcc dot gnu.org> ---
Resolution patch (approved but not yet committed):
   http://gcc.gnu.org/ml/fortran/2013-06/msg00049.html

trans*.c patch for  TYPE => CLASS (submitted):
   http://gcc.gnu.org/ml/fortran/2013-07/msg00086.html

Outstanding/TODO: trans*c patch for TYPE = CLASS, where CLASS is a scalar or
array variable, function or array constructor.

[Bug fortran/57530] [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

--- Comment #7 from Tobias Burnus <burnus at gcc dot gnu.org> ---
Author: burnus
Date: Tue Jul 30 07:20:43 2013
New Revision: 201329

URL: http://gcc.gnu.org/viewcvs?rev=201329&root=gcc&view=rev
Log:
2013-07-30  Tobias Burnus  <burnus@net-b.de>

        PR fortran/57530
        * symbol.c (gfc_type_compatible): A type is type compatible with
        a class if both have the same declared type.
        * interface.c (compare_type): Reject CLASS/TYPE even if they
        are type compatible.


Modified:
    trunk/gcc/fortran/ChangeLog
    trunk/gcc/fortran/interface.c
    trunk/gcc/fortran/symbol.c

[Bug fortran/57530] [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[burnus at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

--- Comment #8 from Tobias Burnus <burnus at gcc dot gnu.org> ---
The original test case (cf. comment 0) is now solved.

However, only "type => class" is handled. Still missing is "type = class",
where CLASS is a (coarray) scalar or (coarray) array variable, function or an
array constructor. See also comment 3.

[Bug fortran/57530] [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING
   Last reconfirmed|                            |2014-09-06
     Ever confirmed|0                           |1

--- Comment #9 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
The codes in comments 0 and 2 compile with gfortran 4.9.1 and trunk (5.0).
Should not this PR resolved as FIXED and a new PR open for the remaining issue
in comment 8?

[Bug fortran/57530] [OOP] Wrongly rejects type_pointer => class_target (which have identical declared type)

[dominiq at lps dot ens.fr]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57530

Dominique d'Humieres <dominiq at lps dot ens.fr> changed:

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

--- Comment #10 from Dominique d'Humieres <dominiq at lps dot ens.fr> ---
> The codes in comments 0 and 2 compile with gfortran 4.9.1 and trunk (5.0).
> Should not this PR resolved as FIXED and a new PR open for the remaining
> issue in comment 8?

No feedback for over a year. Closing as FIXED. Please open new PR(s) with a
test case if there are remaining issue(s).