Re: [Patch, fortran] Parameterized Derived Types

[Janus Weil <janus@gcc.gnu.org>]

[-- Attachment #1: Type: text/plain, Size: 1721 bytes --]

2017-09-09 13:27 GMT+02:00 Damian Rouson <damian@sourceryinstitute.org>:
>
> Hooray!!!! Great work, Paul.

+1 :D


As promised, I have done some testing by now, and found several
problems (ICE-on-invalid, accepts-invalid and rejects-valid), see the
attached test cases. Nothing too severe I guess.



> One small step for gfortran.  One big step for the Fortran world.

I have never looked very closely into PDTs, but now that I do, it
seems like the step for the Fortran world is not quite as big as it
could be (or as I had hoped).

The restriction that KIND parameters cannot be assumed/deferred is a
pretty severe one, isn't it? In terms of "generic programming" it
makes PDTs far less powerful than e.g. C++ templates (I knew that they
are less powerful, but with this restriction it's worse than I
thought).

In particular, it seems that I can define a nice 'generic' type that
deals with arbitrary KIND values in theory, but then all the functions
/ TBPs that work on my type need to be implemented separately for each
KIND value (which means lots of cod duplication), right? Or am I
overlooking anything here? Is anyone aware if this restriction is
lifted in F08 or later, or is planned to be lifted?

Cheers,
Janus




> On September 9, 2017 at 4:15:37 AM, Paul Richard Thomas (paul.richard.thomas@gmail.com(mailto:paul.richard.thomas@gmail.com)) wrote:
>
>> Dear All,
>>
>> The patch has been committed as revision 251925.
>>
>> I look forward to "feedback" (aka PRs) in the coming weeks. I will
>> return to Parameterized Derived Types at the end of this month to
>> clear up some of the known deficiencies (see the notes attached to the
>> patch submission) and any PRs that arise.
>>
>> Regards
>>
>> Paul
>

[-- Attachment #2: pdt1.f90 --]
[-- Type: text/x-fortran, Size: 561 bytes --]

! ICE on invalid & accepts invalid

implicit none

type :: param_matrix(c,r)
  integer, len :: c,r
  real :: m(c,r)
end type

type real_array(k)
  integer, kind :: k
  real(kind=k), allocatable :: r(:)
end type

type(param_matrix(1)) :: m1       ! segfault in gfc_get_pdt_instance (decl.c)
type(param_matrix(1,2)) :: m2     ! ok
type(param_matrix(1,2,3)) :: m3   ! accepted, but invalid
type(param_matrix(1,2.5)) :: m4   ! accepted, but invalid

type(real_array(4)) :: a1        ! ok
type(real_array(5)) :: a2        ! segfault in tree_class_check (tree.h)
end

[-- Attachment #3: pdt2.f90 --]
[-- Type: text/x-fortran, Size: 259 bytes --]

! accepts invalid

implicit none
type :: t(k,i,a,x,y)
  integer, kind :: k
  integer :: i              ! KIND or LEN missing
  integer :: a(3)           ! must be scalar
  real :: x                 ! must be integer
  ! y is not declared at all
end type

end

[-- Attachment #4: pdt3.f90 --]
[-- Type: text/x-fortran, Size: 266 bytes --]

! rejects valid

implicit none

type :: param_matrix(k,c,r)
  integer, kind :: k
  integer, len :: c,r
  real(kind=k) :: m(c,r)
end type

type(param_matrix(8,3,2)) :: mat
real(kind=mat%k) :: m    ! Error: Parameter ‘mat’ at (1) has not been declared or ...

end

[-- Attachment #5: pdt4.f90 --]
[-- Type: text/x-fortran, Size: 696 bytes --]

! rejects valid

module my_matrix
  type :: param_matrix(k,c,r)
    integer, kind :: k
    integer, len :: r
    integer, len :: c
    real(kind=k) :: m(c,r)
  end type param_matrix

  interface process_matrix
    module procedure process_matrix_4
    module procedure process_matrix_8
  end interface

contains

  subroutine process_matrix_4(m)
    type(param_matrix(4,*,*)) :: m   ! Derived type ‘pdtparam_matrix_4’ at (1) is being used before it is defined
  end subroutine process_matrix_4

  subroutine process_matrix_8(m)
    type(param_matrix(8,*,*)) :: m   ! Derived type ‘pdtparam_matrix_8’ at (1) is being used before it is defined
  end subroutine process_matrix_8

end module