[Patch, fortran] PR96320 - gfortran 8-10 shape mismatch in assumed-length dummy argument character array

[Patch, fortran] PR96320 - gfortran 8-10 shape mismatch in assumed-length dummy argument character array

[Paul Richard Thomas]

[-- Attachment #1: Type: text/plain, Size: 1080 bytes --]

This patch fixes the problems in comments 23 and 24 of the PR.

Comment 23 is fixed by the chunk in expr.c. The chunks in decl.c and
resolve.c fix #24. To be quite honest, I am not sure why they were not
needed in the first place! However, the changes don't cause any problems.
Removing the interface bodies causes the expected error cascade.

Regtests on FC33/x86_64 - OK for master and, after a decent delay 9- and
10- branches?

Paul

Fortran: This patch fixes comments 23 and 24 of PR96320.

2021-01-06  Paul Thomas  <pault@gcc.gnu.org>

gcc/fortran
PR fortran/96320
* decl.c (gfc_match_modproc): It is not an error to find a
module procedure declaration within a contains block.
* expr.c (gfc_check_vardef_context): Pure procedure result is
assignable. Change 'own_scope' accordingly.
* resolve.c (resolve_typebound_procedure): A procedure that
has the module procedure attribute is almost certainly a
module procedure, whatever its interface.

gcc/testsuite/
PR fortran/96320
* gfortran.dg/module_procedure_5.f90 : New test.
* gfortran.dg/module_procedure_6.f90 : New test.

[-- Attachment #2: Change2.Logs --]
[-- Type: application/octet-stream, Size: 653 bytes --]

Fortran: This patch fixes comments 23 and 24 of PR96320.

2021-01-06  Paul Thomas  <pault@gcc.gnu.org>

gcc/fortran
	PR fortran/96320
	* decl.c (gfc_match_modproc): It is not an error to find a
	module procedure declaration within a contains block.
	* expr.c (gfc_check_vardef_context): Pure procedure result is
	assignable. Change 'own_scope' accordingly.
	* resolve.c (resolve_typebound_procedure): A procedure that
	has the module procedure attribute is almost certainly a
	module procedure, whatever its interface.

gcc/testsuite/
	PR fortran/96320
	* gfortran.dg/module_procedure_5.f90 : New test.
	* gfortran.dg/module_procedure_6.f90 : New test.

Re: [Patch, fortran] PR96320 - gfortran 8-10 shape mismatch in assumed-length dummy argument character array

[Paul Richard Thomas]

[-- Attachment #1: Type: text/plain, Size: 1269 bytes --]

Sorry, the testcases were left off.

Paul


On Wed, 6 Jan 2021 at 20:23, Paul Richard Thomas <
paul.richard.thomas@gmail.com> wrote:

> This patch fixes the problems in comments 23 and 24 of the PR.
>
> Comment 23 is fixed by the chunk in expr.c. The chunks in decl.c and
> resolve.c fix #24. To be quite honest, I am not sure why they were not
> needed in the first place! However, the changes don't cause any problems.
> Removing the interface bodies causes the expected error cascade.
>
> Regtests on FC33/x86_64 - OK for master and, after a decent delay 9- and
> 10- branches?
>
> Paul
>
> Fortran: This patch fixes comments 23 and 24 of PR96320.
>
> 2021-01-06  Paul Thomas  <pault@gcc.gnu.org>
>
> gcc/fortran
> PR fortran/96320
> * decl.c (gfc_match_modproc): It is not an error to find a
> module procedure declaration within a contains block.
> * expr.c (gfc_check_vardef_context): Pure procedure result is
> assignable. Change 'own_scope' accordingly.
> * resolve.c (resolve_typebound_procedure): A procedure that
> has the module procedure attribute is almost certainly a
> module procedure, whatever its interface.
>
> gcc/testsuite/
> PR fortran/96320
> * gfortran.dg/module_procedure_5.f90 : New test.
> * gfortran.dg/module_procedure_6.f90 : New test.
>

[-- Attachment #2: module_procedure_5.f90 --]
[-- Type: text/x-fortran, Size: 604 bytes --]

! { dg-do compile }
!
! Test the fix for the testcase in comment 23 of PR96320, which used to
! fail with the message: Variable ‘new_foo’ cannot appear in a variable
! definition context.
!
! Contributed by Damian Rouson  <damian@sourceryinstitute.org>
!
module foobar
  implicit none

  type foo
    integer bar
  end type

  interface
    pure module function create() result(new_foo)
      implicit none
      type(foo) new_foo
    end function
  end interface

contains
  module procedure create
    new_foo%bar = 1  ! Error here
  end procedure
end module

  use foobar
  print *, create ()
end

[-- Attachment #3: module_procedure_6.f90 --]
[-- Type: text/x-fortran, Size: 1421 bytes --]

! { dg-do run }
!
! Test the fix for the testcase in comment 24 of PR96320, which used to
! fail with the message: ‘set_user_defined’ must be a module procedure or
! an external procedure with an explicit interface at (1)
!
! Contributed by Damian Rouson  <damian@sourceryinstitute.org>
!
module hole_interface
  type hole_t
    integer :: user_defined
    real :: hole_diameter
  contains
    procedure set_user_defined
    procedure set_diameter
  end type

  interface
    module subroutine set_diameter (this, diameter)
      class(hole_t) :: this
      real :: diameter
    end subroutine

    module subroutine set_user_defined(this, user_defined)
      class(hole_t) :: this
      integer :: user_defined
    end subroutine
  end interface

contains
  module procedure set_user_defined
    this%user_defined = user_defined
  end procedure

  module procedure set_diameter
    this%hole_diameter = diameter
    if (this%user_defined .lt. 0) then
      call this%set_user_defined (0)
    end if
  end procedure
end module

  use hole_interface ! Error was here
  type (hole_t) :: ht = hole_t (-1, 0.0)
  call ht%set_diameter(1.0)
  if ((ht%user_defined .ne. 0) .and. (ht%hole_diameter .ne. 1.0)) stop 1
  call ht%set_user_defined (5)
  if ((ht%user_defined .ne. 5) .and. (ht%hole_diameter .ne. 1.0)) stop 2
  call ht%set_diameter(2.0)
  if ((ht%user_defined .ne. 5) .and. (ht%hole_diameter .ne. 2.0)) stop 3
end

Ping: [Patch, fortran] PR96320 - gfortran 8-10 shape mismatch in assumed-length dummy argument character array

[Paul Richard Thomas]

Ping!

On Wed, 6 Jan 2021 at 20:23, Paul Richard Thomas <
paul.richard.thomas@gmail.com> wrote:

> This patch fixes the problems in comments 23 and 24 of the PR.
>
> Comment 23 is fixed by the chunk in expr.c. The chunks in decl.c and
> resolve.c fix #24. To be quite honest, I am not sure why they were not
> needed in the first place! However, the changes don't cause any problems.
> Removing the interface bodies causes the expected error cascade.
>
> Regtests on FC33/x86_64 - OK for master and, after a decent delay 9- and
> 10- branches?
>
> Paul
>
> Fortran: This patch fixes comments 23 and 24 of PR96320.
>
> 2021-01-06  Paul Thomas  <pault@gcc.gnu.org>
>
> gcc/fortran
> PR fortran/96320
> * decl.c (gfc_match_modproc): It is not an error to find a
> module procedure declaration within a contains block.
> * expr.c (gfc_check_vardef_context): Pure procedure result is
> assignable. Change 'own_scope' accordingly.
> * resolve.c (resolve_typebound_procedure): A procedure that
> has the module procedure attribute is almost certainly a
> module procedure, whatever its interface.
>
> gcc/testsuite/
> PR fortran/96320
> * gfortran.dg/module_procedure_5.f90 : New test.
> * gfortran.dg/module_procedure_6.f90 : New test.
>


-- 
"If you can't explain it simply, you don't understand it well enough" -
Albert Einstein

Un-Ping: [Patch, fortran] PR96320 - gfortran 8-10 shape mismatch in assumed-length dummy argument character array

[Paul Richard Thomas]

[-- Attachment #1: Type: text/plain, Size: 1143 bytes --]

Thomas points out that the attachments went AWOL - apologies.

This patch fixes the problems in comments 23 and 24 of the PR.

Comment 23 is fixed by the chunk in expr.c. The chunks in decl.c and
resolve.c fix #24. To be quite honest, I am not sure why they were not
needed in the first place! However, the changes don't cause any problems.
Removing the interface bodies causes the expected error cascade.

Regtests on FC33/x86_64 - OK for master and, after a decent delay 9- and
10- branches?

Paul

Fortran: This patch fixes comments 23 and 24 of PR96320.

2021-01-15  Paul Thomas  <pault@gcc.gnu.org>

gcc/fortran
PR fortran/96320
* decl.c (gfc_match_modproc): It is not an error to find a
module procedure declaration within a contains block.
* expr.c (gfc_check_vardef_context): Pure procedure result is
assignable. Change 'own_scope' accordingly.
* resolve.c (resolve_typebound_procedure): A procedure that
has the module procedure attribute is almost certainly a
module procedure, whatever its interface.

gcc/testsuite/
PR fortran/96320
* gfortran.dg/module_procedure_5.f90 : New test.
* gfortran.dg/module_procedure_6.f90 : New test.

[-- Attachment #2: submit2.diff --]
[-- Type: text/x-patch, Size: 1732 bytes --]

diff --git a/gcc/fortran/decl.c b/gcc/fortran/decl.c
index 4771b591f1a..723915822f3 100644
--- a/gcc/fortran/decl.c
+++ b/gcc/fortran/decl.c
@@ -9856,7 +9856,8 @@ gfc_match_modproc (void)
   gfc_namespace *module_ns;
   gfc_interface *old_interface_head, *interface;
 
-  if (gfc_state_stack->state != COMP_INTERFACE
+  if ((gfc_state_stack->state != COMP_INTERFACE
+       && gfc_state_stack->state != COMP_CONTAINS)
       || gfc_state_stack->previous == NULL
       || current_interface.type == INTERFACE_NAMELESS
       || current_interface.type == INTERFACE_ABSTRACT)
diff --git a/gcc/fortran/expr.c b/gcc/fortran/expr.c
index 188e79669cb..4f456fc629a 100644
--- a/gcc/fortran/expr.c
+++ b/gcc/fortran/expr.c
@@ -6243,6 +6243,9 @@ gfc_check_vardef_context (gfc_expr* e, bool pointer, bool alloc_obj,
 
   /* Variable not assignable from a PURE procedure but appears in
      variable definition context.  */
+  own_scope = own_scope
+	      || (sym->attr.result && sym->ns->proc_name
+		  && sym == sym->ns->proc_name->result);
   if (!pointer && !own_scope && gfc_pure (NULL) && gfc_impure_variable (sym))
     {
       if (context)
diff --git a/gcc/fortran/resolve.c b/gcc/fortran/resolve.c
index fa6f756d285..2cb009f8cef 100644
--- a/gcc/fortran/resolve.c
+++ b/gcc/fortran/resolve.c
@@ -14016,7 +14016,8 @@ resolve_typebound_procedure (gfc_symtree* stree)
       /* Check for F08:C465.  */
       if ((!proc->attr.subroutine && !proc->attr.function)
 	  || (proc->attr.proc != PROC_MODULE
-	      && proc->attr.if_source != IFSRC_IFBODY)
+	      && proc->attr.if_source != IFSRC_IFBODY
+	      && !proc->attr.module_procedure)
 	  || proc->attr.abstract)
 	{
 	  gfc_error ("%qs must be a module procedure or an external "

[-- Attachment #3: module_procedure_6.f90 --]
[-- Type: text/x-fortran, Size: 1421 bytes --]

! { dg-do run }
!
! Test the fix for the testcase in comment 24 of PR96320, which used to
! fail with the message: ‘set_user_defined’ must be a module procedure or
! an external procedure with an explicit interface at (1)
!
! Contributed by Damian Rouson  <damian@sourceryinstitute.org>
!
module hole_interface
  type hole_t
    integer :: user_defined
    real :: hole_diameter
  contains
    procedure set_user_defined
    procedure set_diameter
  end type

  interface
    module subroutine set_diameter (this, diameter)
      class(hole_t) :: this
      real :: diameter
    end subroutine

    module subroutine set_user_defined(this, user_defined)
      class(hole_t) :: this
      integer :: user_defined
    end subroutine
  end interface

contains
  module procedure set_user_defined
    this%user_defined = user_defined
  end procedure

  module procedure set_diameter
    this%hole_diameter = diameter
    if (this%user_defined .lt. 0) then
      call this%set_user_defined (0)
    end if
  end procedure
end module

  use hole_interface ! Error was here
  type (hole_t) :: ht = hole_t (-1, 0.0)
  call ht%set_diameter(1.0)
  if ((ht%user_defined .ne. 0) .and. (ht%hole_diameter .ne. 1.0)) stop 1
  call ht%set_user_defined (5)
  if ((ht%user_defined .ne. 5) .and. (ht%hole_diameter .ne. 1.0)) stop 2
  call ht%set_diameter(2.0)
  if ((ht%user_defined .ne. 5) .and. (ht%hole_diameter .ne. 2.0)) stop 3
end

[-- Attachment #4: module_procedure_5.f90 --]
[-- Type: text/x-fortran, Size: 604 bytes --]

! { dg-do compile }
!
! Test the fix for the testcase in comment 23 of PR96320, which used to
! fail with the message: Variable ‘new_foo’ cannot appear in a variable
! definition context.
!
! Contributed by Damian Rouson  <damian@sourceryinstitute.org>
!
module foobar
  implicit none

  type foo
    integer bar
  end type

  interface
    pure module function create() result(new_foo)
      implicit none
      type(foo) new_foo
    end function
  end interface

contains
  module procedure create
    new_foo%bar = 1  ! Error here
  end procedure
end module

  use foobar
  print *, create ()
end

[-- Attachment #5: module_procedure_4.f90 --]
[-- Type: text/x-fortran, Size: 1460 bytes --]

! { dg-do run }
!
! Test the fix for PR96320 in which the assumed shape of 'arg' in the
! interface for 'bar' was mirrored by the 'arg' in the module procedure
! incorrectly have deferred shape.
!
! Contributed by Damian Rouson  <damian@sourceryinstitute.org>
!
module foobar
  type foo
  contains
    procedure, nopass :: bar1
    procedure, nopass :: bar2
    procedure, nopass :: bar3
  end type

  interface

    module subroutine bar1(arg)
      character(len=*) arg(:)
    end subroutine

    module subroutine bar2(arg)
      character(len=*) arg(3:)
    end subroutine

    module subroutine bar3(arg)
      character(len=*) arg(2)
    end subroutine

  end interface
contains

  module procedure bar1
    if (lbound(arg, 1) .ne. 1) stop 1
    if (arg(3) .ne. 'hijk') stop 2
  end procedure

! Make sure that the lower bound of an assumed shape array dummy,
! if defined, is passed to the module procedure.

  module procedure bar2
    if (lbound(arg, 1) .ne. 3) stop 3
    if (arg(3) .ne. 'abcd') stop 4
  end procedure

! This makes sure that an dummy with explicit shape has the upper
! bound correctly set in the module procedure.

  module procedure bar3
    if (lbound(arg, 1) .ne. 1) stop 5
    if (arg(3) .ne. 'hijk') stop 6       ! { dg-warning "is out of bounds" }
  end procedure

end module

  use foobar
  character(4) :: list(3) = ['abcd', 'efgh' , 'hijk']
  type(foo) :: f
  call f%bar1(list)
  call f%bar2(list)
  call f%bar3(list)
end