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

[no subject]

Steve,

>Date: 10 Jun 1997 15:34:21 -0500 
>From: "Steve Mauget" <address@hidden>
>Organization: USDA
>To: "Steve E" <address@hidden>
>Keywords: 199706021417.IAA27579

In the above message, you wrote:

>     Make ran ok so its not included here. But make test produced
> the same old set of problems associated with integer and integer*2
> arguments to subroutines..... 
>                                       :
>                                       :
>                                       :
> \/
> 
> Making `test' in directory /temp/netcdf-3.3.1/src/fortran
> 
>       /usr/local/bin/f90 -c -g ftest.f
> Warning: ftest.f, line 1446: Unused external reference NF_INQ_NDIMS
>          detected at END@<end-of-statement>
> Warning: ftest.f, line 1446: Unused external reference NF_GET_VARA_INT1
>          detected at END@<end-of-statement>
>                               :
>                             etc, etc.
>                               :
> Warning: ftest.f, line 18935: Unused external reference NF_PUT_VARA_INT2
>          detected at END@<end-of-statement>
> Warning: ftest.f, line 18935: Unused external reference NF_INQ_NVARS
>          detected at END@<end-of-statement>

These warnings may safely be ignored.

> [f90 continuing despite warning messages]
> Error: ftest.f, line 2786: Inconsistent datatype for arg 4 in call to NCAGT
> Error: ftest.f, line 2788: Inconsistent datatype for arg 4 in call to NCAGT
> Error: ftest.f, line 2790: Inconsistent datatype for arg 4 in call to NCAGT
>                                       :
>                                     etc, etc.
>                                       :
>                                       :
> Error: ftest.f, line 18923: Inconsistent datatype for arg 4 in call to NCVGT1
> Error: ftest.f, line 18924: Inconsistent datatype for arg 4 in call to NCVGT1
> Error: ftest.f, line 18925: Inconsistent datatype for arg 4 in call to NCVGT1
> Error: ftest.f, line 18926: Inconsistent datatype for arg 4 in call to NCVGT1
> [f90 error termination]
> *** Error code 2 (bu21)
> *** Error code 1 (bu21)
> *** Error code 1 (bu21)

Any word yet from NAG about this "error"?

> 
>                                       :
>                                       :
>                                       :
> \/
> 
> A related problem (I believe) shows up as a run-time error when I run 
> my wrnetcdf.f program. I tried explicitly defining ncid as an integer,
> but got the same results. Also wrote  ncid out in a formatted statement
> (which wrote without error). But for some reason my one variable was 
> evidently not defined.
> 
>                  **** stdrd out and err ****
> 
> >>>ncvardef: ncid 3: Invalid dimension id or name<<< (My arrows)
> OUTPUT =  PCP.NC
> NCID =     0
> IERR=  0
> IERR=  0
>                  **** program text ****
>        Vname='PCP'
>        OutFile=Vname//'.NC'
>        write(*,'(a10,a6)')'OUTPUT =  ',OutFile
>        write(*,'(a10,I2)')'NCID =    ',ncid
> C
> C      Enter define mode, create new file...
>         write(*,'(a6,i2)')'IERR= ',ierr
>       ncid = NCCRE(OutFile,ncnoclob,ierr)
>         write(*,'(a6,i2)')'IERR= ',ierr
> C      Define dimensions...
>       imon = NCDDEF(ncid,'month',nmon,ierr)
>                                       :
>                                       :
>                  **** stdrd out and err ****
>   NDIMS  2
>   NCID   3
>   NVARS  0   << My one variable evidently not defined
>                 because (?) ncid is an Invalid dimension id or name
>   NATTS  1
>   IREC   -1
>   STINDX  7
>   IERR   0
>   ::::::::::::::::::::::::::::::::::::::::::::::::::::
> Fortran-90 STOP

Unfortunately, there's not enough context in the above for me to
determine what the problem might be.  The error message, however,
indicates that the problem probably isn't with "ncid", which is the netCDF
dataset ID, but rather with whatever variables you're giving to "ncvdef"
as the dimension IDs.  The netCDF library seems to be telling you that
at least one of the dimension IDs is invalid -- not that the netCDF ID
is invalid.

--------
Steve Emmerson   <address@hidden>