J3/10-125r1
To: J3/WG5
From: Malcolm Cohen
Subject: Edits for ISO Secretariat comments
Date: 2010 February 15


1. Introduction
---------------

The ISO Secretariat comments are in paper N1788.  One comment was also
addressed by 10-114 but the edit in that paper was insufficient.

Note that several ISO Secretariat comments had "etc." in the location.
I think I have caught all the subclauses that are affected, but it is
not impossible that I might have missed one or more.


2. Edits to 09-007r3 (aka WG5/N1791)
------------------------------------

[throughout] Change headers to "ISO/IEC 1539-1:2010 (E)",
             aligned right on odd pages and left on even pages.

[throughout] Change odd-page-numbered footers to
    "ISO/IEC copyright (c) ISO/IEC 2010 (All rights reserved)   pagenumber"
  and even-page-numbered footers to
    "pagenumber   ISO/IEC copyright (c) ISO/IEC 2010 (All rights reserved)"
  where "(c)" is a copyright symbol and "pagenumber" is the page number.

[3:9] 1.3.7 associating entity,
  "in ... association," -> "\termqual{in ... association}",
  where "\termqual{x}" expands to "$\langle$x$\rangle$".
{Notes: $\langle$ is supposed to be a left angle bracket, not a < sign.
        Yes, I do mean to swallow the comma.}
{NOTE TO EDITOR: Check for correct spacing using \termqual, if that is
                 problematic use the expansion inplace.}

[4:44] 1.3.19 C address,
  Delete semi-colon and turn everything after it into a proper sentence
  in a note.

[5:3-8] 1.3.21 characteristics,
  Replace entire definition as follows:
    "\termqual{procedure} the properties listed in 12.3.1

     \mterm{characteristics}
     \termqual{dummy argument} being ... indicator)

     \mterm{characteristics}
     \termqual{dummy data object} ...

     \mterm{characteristics}
     \termqual{dummy procedure or dummy procedure pointer} ... 12.3.2.3

     \mterm{characteristics}
     \termqual{function result} ... 12.3.3"
{Note: \mterm is a new macro for a multiply-defined term; it is nearly the
       same as \term but does not define a new hyperlink target.}

[5:45] 1.3.29.3 subcomponent,
  "of a structure," -> "\termqual{of a structure}".

[6:11] 1.3.31 conformable, replace with
  "of two data entities," -> "\termqual{of two data entities}",
  "shape, or" -> "shape or".

[6:33] 1.3.36 contiguous,
  "of a multi-part data object," -> "\termqual{a multi-part data object}",
  replace "; of an array,"
  with "\mterm{contiguous}
        \termqual{of an array}".
{Note: the subterm "simply contiguous" is under the second definition.}

[6:37] 1.3.36.1 simply contiguous,
  "of an array designator or variable,"
  ->"\termqual{of an array designator or variable}".

[7:21] 1.3.44 default-initialized,
  "of a subcomponent," -> "\termqual{subcomponent}".

[7:28-30] 1.3.46 defined,
  Change "either ... data object, the" to "\termqual{data object}",
  Change "or, ... pointer, the"
  to "
      \mterm{defined}
      \termqual{pointer}".

[8:3-5] 1.3.50 definition, replace entirely with
  "\termqual{derived type (4.5.2), enumeration (4.6), or procedure (12.6)}
   specification of the type, enumeration, or procedure

   \mterm{definition}
   \termqual{data object} process by which the data object becomes defined
   (16.6.5)".

[8:8] 1.3.51 descendant,
  "of ... submodule," -> "\termqual{module or submodule}".

[8:23-26] 1.3.53 disassociated,
  "either ... status of"
  -> "\termqual{pointer association} the pointer association status of",
  ", or ... status"
  -> "
      \mterm{disassociated}
      \termqual{pointer} having a pointer association status of
      disassociated".

[10:17-19] 1.3.66 finalizable,
  "either ... of a type," -> "\termqual{type}",
  ", or ... entity,"
  -> "
      \mterm{finalizable}
      \termqual{nonpointer data entity} of finalizable type".

[10:31] 1.3.70 host instance,
  "of an ..., or of a ... procedure,"
  -> "\termqual{..., or ... procedure";
  this results in
  "\termqual{internal procedure, or dummy procedure or procedure pointer
   associated with an internal procedure}".

[11:12] 1.3.78 inherit,
  "of an extended type," -> "\termqual{extended type}".

[11:20] 1.3.80 interface,
  "of a procedure," -> "\termqual{procedure}",

[12:3] 1.3.83 interoperable,
  "of a Fortran entity," -> "\termqual{Fortran entity}".

[12:11] 1.3.84.1 standard intrinsic,
  "of a procedure or module," -> "\termqual{procedure or module}".

[12:14] 1.3.84.2 nonstandard intrinsic,
  "of a procedure or module," -> "\termqual{procedure or module}".

[13:22-23] 1.3.97 operator, replace with
  "\si{intrinsic-operator}, \si{defined-unary-op}, or
   \si{defined-binary-op} (R309, R703, R723)"
OPTIONAL follow by a Note
  "Note 1.x
   An operator specifies a computation involving one data value (unary
   operator with prefix syntax) or two data values (binary operator with
   infix syntax)."
{Personally I don't think we need the note.}

[13:43] 1.3.101 polymorphic,
  "data entity" -> "\termqual{data entity}",
  Delete "declared with the CLASS keyword,".
{Either the first part before the comma is the definition, in which case
 the second part after the comma is extra info which must be given in a
 note (or omitted), or vice versa.  Since the reference goes straight to
 the syntax, I have chosen the "semantic" version of the definition.}

[14:3] 1.3.102 preconnected,
  "of a file or unit," -> "\termqual{file or unit}".

[15:33-36] 1.3.115 scoping unit, replace entirely with
  "derived-type definition, interface body, program unit, or subprogram,
   excluding all nested scoping units in it"
{Sure, there is no syntax to nest anything in a derived-type definition,
 but that doesn't matter.  We don't want this to be multiple separate
 definitions, and we do want the wording to be simple.}
{Interacts with other papers; the result should be to insert "block
 construct," at the beginning of the definition.}

[16:16] 1.3.120 size,
  "of an array," -> "\termqual{array}".

[17:6] 1.3.128 storage unit,
  Delete "unit of storage; a".

[18:43-44] 1.3.137.7 extension type, replace with
  "\termqual{of one type with respect to another} is the same type or is
   an extended type whose parent type is an extension type of the other
   type".
{Previous version was easier to read than this, but did not have the
 correct form required for a term definition.}

[19:12] 1.3.137.11 type compatible,
  Replace
    "of one entity with respect to another, compatibility of the types of
     the entities"
  with
    "compatibility of the type of one entity with respect to another".

[19:16-17] 1.3.137.12 type parameter,
  Delete ", further ... available".
{Useless waffle is forbidden in a term definition!}

[19:20-23] 1.3.137.12.1 assumed type parameter,
  Replace "which is, ... dummy argument"
  with "
        Note 1.x
        The other entity is
        ... dummy argument.".
{Turn the extra info into a note.  We could just delete it, but that might
 be slightly unfriendly here.}

[19:45-20:1] 1.3.139 undefined,
  "either ... object," -> "\termqual{data object}",
  "or ... pointer, the"
  -> "
      \mterm{undefined}
      \termqual{pointer}".

[20:9] 1.3.141 unlimited polymorphic,
  Delete "declared with CLASS(*),".
{Either the first part before the comma is the definition, in which case
 the second part after the comma is extra info which must be given in a
 note (or omitted), or vice versa.  Since the reference goes straight to
 the syntax, I have chosen the "semantic" version of the definition.}

[24:28-29] 1.6.4, first bullet point, replace entirely with
  "The result value of the intrinsic function SIGN when the second
   argument is negative real zero differs from that of Fortran 90."
{Note: It would be false to say that this "is changed by this standard",
       since in fact it was Fortran 95 that changed it.
       I also changed "behavior" to "result value", since it is the
       result value that has changed, not some other aspect of the
       behavior like whether it raises IEEE_INVALID.}
ALTERNATIVE
[24:27-33] Change first "." to ":" and replace the list contents with
  "- the result value of the intrinsic function SIGN (when the second
     argument is a negative real zero);
   - formatted output of negative real values (when the output value is
     zero);
   - whether an expression is a constant expression (thus whether a
     variable is considered to be automatic)."
{Line 27 already says these are differences between F90 and this F, why
 do we have to say it again in every list item?}

[233:Note 9.58] 9.9 "FLUSH statement", after p4,
  "this standard" -> "this part of ISO/IEC 1539".
{ISO-4: use name of standard instead of 'this standard'.}

===END===