To: J3                                                     J3/24-149
From: Malcolm Cohen
Subject: Interp F23/015 on coindexed objects in structure constructors
Date: 2024-June-26
References: 24-007, 24-123

----------------------------------------------------------------------

NUMBER: F23/015
TITLE: Coindexed objects in structure constructors
KEYWORDS: Pointer component, coindexed object, structure constructor
DEFECT TYPE: Erratum
STATUS: J3 consideration in progress

QUESTION:

Consider

  TYPE realptrwrap
    REAL,POINTER :: p
  END TYPE
  TYPE t
    TYPE(realptrwrap) c
  END TYPE
  TYPE(realptrwrap) x[*]
  TYPE(t) y
  REAL,TARGET :: z[*]

  x = realptrwrap(z[2])     ! (A) harmful
  y%c = realptrwrap(z[2])   ! (B) harmful
  x = realptrwrap(z)        ! This assignment is valid and harmless.
  y = t(x[2])               ! (C) harmful

(Aside: "harmful" means "copies a pointer from one image to another".)

The structure constructors in the statements (A) and (B) are invalid,
by combination of
    R758 component-data-source  is  expr
                                or  data-target
                                or  proc-target
    C7109 (R758) A data-target shall correspond to a data pointer
          component; ...
    R1038 data-target  is  expr
    C1029 (R1038) A data-target shall not be a coindexed object.

However, the statement (C) has the same undesirable effect as (B), but
does not violate those constraints, and thus appears on the face of it
to be a valid statement.

Statement (C) would, however, be prohibited in a pure procedure, even
though it cannot cause side effects, merely undefined pointers.

This appears to be inconsistent. Is this deliberate?

ANSWER:

This apparent inconsistency was inadvertent.
Edits are provided to correct the issue.

EDITS to 24-007:

[93:21+] 7.5.10 Construction of derived-type values,
         after the penultimate constraint C7109, insert new
         constraint
    "C7109a (R758) If <expr> is a coindexed object, it shall not have
            a pointer component at any level of component selection."

[93:23-] Same subclause, after NOTE 1, insert new NOTE
    "NOTE 1a
     Although a coindexed object with a pointer subcomponent is not
     the only way for the structure constructor to produce a value
     with an undefined pointer subcomponent, copying a pointer from
     another image is particularly pernicious, and thus precluded in
     this context."

[34:1+] 4.3.3 Fortran 2018 compatibility, after paragraph 4,
        insert new paragraph
    "Fortran 2018 permitted a <component-data-source> in a structure
     constructor to be a coindexed object with a pointer subcomponent.
     This document does not permit such usage."

[35:8+] 4.3.4 Fortran 2008 compatibility, after paragraph 14,
        insert new paragraph
    "Fortran 2008 permitted a <component-data-source> in a structure
     constructor to be a coindexed object with a pointer subcomponent.
     This document does not permit such usage."

SUBMITTED BY: Malcolm Cohen

HISTORY: 24-nnn   m233  Submitted

----------------------------------------------------------------------