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

Re: a persistence problem

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: a persistence problem
  • Date: Tue, 25 Apr 2000 11:16:09 -0600
>To: address@hidden
>From: Jacob Bensabat <address@hidden>
>Subject: Re: 20000425: a persistence problem
>Organization: 
>Keywords: C++ persistence STL

Hi Jacob,

> I have been using the NetCDF lib for a while and I have encountered a
> problem, that I will try to describe.
> I have built a c++ class that wraps the Netcdf functionality. One of its
> members is a netcdf file, others are access and write functions.
> Since I want to open more than one Netcdf file at a time, I store each
> instance of the class in an stl vector. The problem
> is that once I stored the class instance in the vector I loose the
> ability to retrive data from the file (I get a -33 error codes which
> means: invalid cdf file). I solved the problem by open and closing the
> netcdf file each time I retrieve a member from the stl vector. My
> question is wheter there is a better way to overcome this problem.

From the description above, I can't tell whether the problem is with
the netCDF library or your wrapper.  It's intended to be possible to
have multiple netCDF files open simultaneously with the netCDF C++
interface, and no one else has reported a problem with opening
multiple netCDF files at the same time from C++.  If you are declaring
an NcFile object that goes out of scope, its destructor will get
called, which closes that file.  Perhaps you just need to make sure
the ~NcFile(void) destructor doesn't get called.

Otherwise, if you could construct a small example that demonstrates
the problem, we could reproduce it here and diagnose where the problem
lies.

Also, it would be better to send netCDF problem reports to
"address@hidden" rather than my email address.  That way the
problem will get routed to someone who can deal with it even if I
happen to be on vacation or away from my email.

--Russ

_____________________________________________________________________

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