[netCDF #WVL-288144]: NetCDF Fortran 90 Interface

> 
> Hi, your NetCDF Fortran 90 Interface looks like it might be very useful
> to us, and we'd like to incorporate it into our code, but I'm a little
> bit concerned about the possibility of Fortran unit number conflicts.
> I assume that, on opening a new NetCDF file, a Fortran unit number is
> assigned, but you don't appear to have any control over which unit
> number that is. Since we also do non-NetCDF I/O in our code then this
> seems to open the possibility that a NetCDF will choose a unit number
> which will later be used (or perhaps is already being used) by another
> part of the code, leading to a conflict. Is this indeed a possibility,
> or am I underestimating the cleverness of your code? If it is a
> possibility then is their a way around it, or of mitigating the risk?
> Can we at least know what unit number has been chosen for the NetCDF
> file?
> 
> Any advice greatly appreciated,
> 
> Luke Jones.
> 
> --
> Dr Luke Jones  Mesoscale Development Scientist
> Met Office   FitzRoy Road   Exeter   EX1 3PB   United Kingdom
> Tel: +44 (0)1392 884109   Fax: +44 (0)1392 885681
> E-mail: address@hidden   http://www.metoffice.com
> 
> 

Howdy Luke Jones!

NetCDF files, even when accessed with the F90 API, are actually all opened and 
manipulated in C code, not in Fortran code. So no Fortran unit numbers are used.

The netCDF F90 functions call netCDF F77 functions. The netCDF F77 functions 
call netCDF C functions. The C functions are where all the file opens and 
closes take place.

So Robert has assumed his proper avuncular relationship!

Keep on netCDFing!

Ed Hartnett

Ticket Details
===================
Ticket ID: WVL-288144
Department: Support netCDF
Priority: Normal
Status: Closed