Fortran 77 Interface

[Ray Osborn]

I've been trying to develop a more comprehensive test program to make sure
that the different language interfaces give consistent input/output,
particularly with regard to slabs of multi-dimensional datasets.  I've now
got around to the Fortran 77 interface which appears to have a number of
serious problems.  Freddie drew attention to this a year ago, but I'm afraid
that I was occupied with other matters and have only just realized how
serious they are.

The problem is that, as you probably know, Fortran passes character strings
by descriptor rather than reference, so that it is impossible to have a
single routine that accepts either numeric or character data in the same
argument.  Fortran 90 can cope with this by allowing multiple generic
routines that are invoked according to the type of argument in the calling
program.  Fortran 77 doesn't have this facility.

Unfortunately, the only way around this, I believe, is to have separate
NXputattr and NXputcharattr routines (and NXgetattr etc).  I now realize
that this applies to NXputdata and NXgetdata as well, since character string
SDS's are allowed in the standard.  This makes the F77 API different from
the others, but I can't see a more elegant or portable way.  I think that
Freddie came to the same conclusions last year.

I can make the necessary changes but I wanted to check if anyone has any
objections first, before I commit the files to the CVS server.  We will then
have to make the announcement to the NeXus mailing list.

Please reply by Monday if possible, because I want to get this out of the
way so that we can release a (bug-free?) v 1.1.1.  I've also not heard any
response to my last message concerning the HDF Browser.  Has anyone seen
linked data items properly displayed in it?  I would like to send Fortner a
bug report.

Regards,
Ray
--
Dr Ray Osborn                Tel: +1 (630) 252-9011
Materials Science Division   Fax: +1 (630) 252-7777
Argonne National Laboratory  E-mail: ROsborn at anl.gov
Argonne, IL 60439-4845