(j3.2006) Namelist output

Robert Corbett Robert.Corbett
Fri Jan 16 03:50:59 EST 2009

Van Snyder wrote:
> On Thu, 2009-01-15 at 02:36 -0800, John Reid wrote:
>>The effect of this interpretation is that vendors are given the choice
>>of using either form. That seems to be what is happening in the
>>Are we all agreed on this, or should I submit a formal interp request?
> The standard is not inconsistent and not terribly unclear that namelist
> output either can or cannot include component names, at the vendors
> whim.

What is your rationale for that claim?  I agree that the standard is
reasonably clear about the form of namelist output, but it clearly
does not allow component names in namelist output.

Section of the Fortran 2003 standard states

     The name of each namelist group object list item is placed
     in the output record followed by an equals sign and a list
     of values of the namelist group object list item.

Section of CD 1539-1 contains the same statement.  That
statement does not say that the names of the subobjects of a
namelist group list item can be placed in an output record.  I
find nothing in the standard or in CD 1539-1 that allows that.

The statement in the standard that "the form of the output
produced is the same as that required for input" does not imply
that any form that can be used for input can be produced as

Would anyone argue that the value of a variable of type CHARACTER
could be written as a sequence of substrings?

Bob Corbett

More information about the J3 mailing list