[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 20020328: NetCDF file status

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: 20020328: NetCDF file status
  • Date: Tue, 02 Apr 2002 12:52:54 -0700
>To: address@hidden
>cc: address@hidden
>From: Paul van Delst <address@hidden>
>Subject: NetCDF file status
>Organization: CIMSS @ NOAA/NCEP
>Keywords: 200203281510.g2SFAha03731

Hi Paul,

> I hope this is the right place to ask this - I didn't want to
> broadcast it to netcdfgroup.

Yes, sorry it took me so long to reply ...

> How does one determine if a netCDF file is already open? E.g. If I
> have a f90 routine:
> 
>   function read_netcdf_data( ncfile_id, ...other arguments, filename = 
> filename )
> 
> How can I use the ncfile_id to determine if the file is open. Or for
> that matter, what mode the file is in; data or define? e.g. I'd like
> to be able to do something like
> 
>   if ( .not. ncfile_is_open( ncfile_id ) ) then ...
>     open it using optional filename argument
> 
> and
> 
>   if ( ncfile_mode( ncfile_id ) /= 'DATA' ) then ... 
> or
>   if ( .NOT. ncfile_data_mode( ncfile_id ) ) then ... 
>     put it in data mode
> 
> any ideas? I think the open test can be handled with the
> nf90_inquire() function - the only *listed* cause of error is if the
> specified netCDF id does not refer to an open netCDF dataset.  But the
> second one seems a bit trickier. And the documentation always states
> "possible causes of errors include:" suggesting something other than
> an unopened dataset may trigger an error in the inquire function.

Your idea of using nf90_inquire() will work in that it will tell you
whether the ncfile_id is associated with an open file (but not which
file).  The only possible cause for nf90_inquire() to return an error
is if the specified id is not in the list of open IDs (as can be seen
in the C source in libsrc/nc.c for the NC_check_id function, which is
ultimately called by nf90_inquire()).  The documentation should be more
definitive about this; I think it states "possible causes of errors
include:" just to make it a little more maintainable in case there are
additional errors in the future (e.g. a network error for remote
access).

Once you have determined that the file is open, you can determine
which mode it is in by calling a function that returns an error if the
dataset was in define mode, for example nf90_redef().  This will
return NF90_NOERR if in data mode (in which case you can call
nf90_enddef() to return to data mode).  Otherwise you were already in
define mode and there was no change of state.

A function that would safely return the state would be a good addition
to the interface.

Please let us know if this doesn't work or you have other questions
about this.

--Russ

_____________________________________________________________________

Russ Rew                                         UCAR Unidata Program
address@hidden                     http://www.unidata.ucar.edu