(j3.2006) (SC22WG5.4504) Ballot on N1866

[Rasmussen, Craig]

> On 07/11/2011 06:02 PM, John Reid wrote:
>
> Please answer the following question "Is N1866 ready for forwarding to
> SC22 as the PDTR?" in one of these ways.
>

It looks like I've missed the deadline for the vote (I would have
voted for choice 2).  I read N1866 carefully last week and have a few
minor suggestions for the revised document:

1. Pg 9, line 10: I believe there is extraneous space between "A" and "shall".

2. Table 5.3: Is there a reason all of the lines aren't right justified?

3. sizeof() usage:  The sizeof() usage in elem_len description seems
awkward to me (for example, pg 17, line 13):  Consider changing to,
for example, "elem_len shall be the number of characters in an element
of the object times the sizeof() function applied to a scalar of the
character type."

4. NOTE 5.7:  I believe the standard uses a comma before "and" in a
list, so the note should read "... to set the rank, attribute, type,
and element length."

5. NOTE 5.8: I believe there is a missing hyphen.  The end of the note
should read "... for a Fortran assumed-shape array."

6. Pg. 20 line 32 (for example): Maybe I missed it but it's not clear
to me from the specification of the functions in 5.3.5 that zero-based
indexing is to be used for the bounds arrays.  The examples for
CFI_section and CFI_select_part use array declarations with a lower
bounds of 1 in Fortran but this is the least complicated choice.  It
would be helpful if at least one of the examples were for an array
pointer with lower bounds other than 1.

Cheers,
Craig