Aleksandar Donev <[log in to unmask]>
> Since we do not have variable-IO lists or anything like the C-printf, it
> is not possible for me to replace the WRITE statement with my own
> provided subroutine. Rather, I let the user use internal IO, and then I
> take care of actually writing that to a file, the screen, sending it
> via MPI, or whatever.
This is exactly the reason that I advocated, in J3 paper 97-114, to allow
the user to intercept read and write at the bottom level. If this were
allowed, it would be possible to define READ and WRITE on some units to
use MPI or whatever for communication. Because I cannot do this, I have
enormous PVM-read and PVM-write routines for each of my data structures.
For most of them, each of PVM-read and PVM-write require more code than
filling the data structure from the configuration file. So my code size
triples in this area. In the F03 era, one might say "Just use UDDTIO,"
but this just pushes the mess from one place to another.
--
Van Snyder | What fraction of Americans believe
[log in to unmask] | Wrestling is real and NASA is fake?
Any alleged opinions are my own and have not been approved or disapproved
by JPL, CalTech, NASA, Sean O'Keefe, George Bush, the Pope, or anybody else.
|