(j3.2006) J3 Fortran interp letter ballot #14 - due 8-Oct-2007

[Jim Xia]

My votes for letter ballot #14:



The following Fortran interpretations are being balloted:
 
Yes   No    Number     Title
 
-Y-   ---   F03/0092   Procedure characteristics and unlimited
                        polymorphic
-C-   ---   F03/0093   Allocatable array on intrinsic assignment with
                        scalar expr
-Y-   ---   F03/0094   Final subroutine and VALUE attribute
-Y-   ---   F03/0095   Bounds remapped pointer assignment and
                        ASSOCIATED
-Y-   ---   F03/0097   Blanks as separators in NAMELIST input
-C-   ---   F03/0098   Does allocate with source= define subcomponents?
-N-   ---   F03/0099   Clause 16 does not account for volatile variable
-Y-   ---   F03/0100   Error in field width for special cases of signed
                        INFINITY output
-C-   ---   F03/0101   Is UDDTIO output suitable for namelist and
                        list-directed input

-----------------------------------------------------------------------
Comment on interp F03/0093 (Allocatable array on intrinsic assignment
with scalar expr)

In the first edit, the second half of the sentence could be improved
to say that <expr> shall be an array of the same rank as <variable>.
-----------------------------------------------------------------------
Comments on interp F03/0098 (Does allocate with source= define
subcomponents?)

The subclause numbers are wrong in edit 2 and 3:
[421:27-28] 16.5.6, ...

     should be

[421:27-28] 16.5.5, ...


[421:28+] 16.5.6, ...

     should be

[421:28+] 16.5.5, ...

-----------------------------------------------------------------------
NO vote on F03/0099  (Clause 16 does not account for volatile variable)

I agree partially with Van's No vote.  The type parameter of a non-
polymorphic entity that can be changed by means other than the program
itself must be deferred. In addition, the deferred type parameter does
not require the entity to be polymorphic.

In the case of a polymorphic entity with VOLATILE, however, there is a
possibility that the dynamic type of the entity may require type
parameters in addition to those specified in the type declaration itself.
Any of these additional type parameters may change due to either change
of the dynamic type or change in entity's association status.  This is
not covered in Van's edit.

So I suggest an edit as follows (it may need wordsmithery)

[415:27]  Add a new paragraph at the end of 16.4.2.1.4

"When a pointer object with the VOLATILE attribute is changed by a means
not specified by the program it may become defined or undefined.  In
addition, its array bounds, association status and deferred type
parameters may change.  If it is polymorphic, its dynamic type and any
additional type parameters not specified in its declaration may also
change."

-----------------------------------------------------------------------
Comments on F03/0101 (Is UDDTIO output suitable for namelist and
 list-directed input)

I agree with Bill's comments (a), (b) and (c).
-----------------------------------------------------------------------


Cheers,

Jim Xia

XL Fortran Compiler Testing
IBM Toronto Lab at 8200 Warden Ave.
Phone (905) 413-3444  Tie-line 969-3444
D2/NAH/8200 /MKM
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://j3-fortran.org/pipermail/j3/attachments/20070920/ded3baee/attachment.html