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

951219: appending data

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: 951219: appending data
  • Date: Tue, 19 Dec 95 10:57:02 -0700
Maureen,

> To:       address@hidden
> From: Maureen Cribb <address@hidden>
> Organization: Dalhousie University
> Keywords: 199512191601.AA23714

In the above message you wrote:
 
> Enclosed are a couple of sample programs and small data sets that
> illustrate the problem I'm having in appending data.  
> 
> The program "first.f" creates a netcdf file containing the variable
> "a1" (with an unlimited dimension), reads in data from "first.dat"
> (ten elements) and writes this data to the netcdf file. This part
> works fine.
> 
> Now an additional five elements (from "second.dat") are to appended
> to the variable in this netcdf file.  The program "second.f" opens
> the newly created netcdf file, gets the id number for the "a1"
> variable, reads in data from "second.dat" and writes it to the netcdf
> file. This data is, indeed, appended but extra data also appears to be
> appended. 
> 
> I expected fifteen elements to appear for "a1" in the netcdf file but
> twenty-five elements actually appear.  Could you shed some light on this
> situation?
> 
> Thanks in advance for your help ...
> 
> M.Cribb
> 
> - ----------
> X-Sun-Data-Type: default
> X-Sun-Data-Name: first.f
> X-Sun-Content-Lines: 35
> 
>       program first
> 
>       include '/opt/netcdf/include/netcdf.inc'
> 
>       integer ncid,rcode,rec_dim,a1_id
>       real a1(10)
> 
> c open data file
>       open(1000, file='first.dat', status='old')
>       
> c create netcdf file
>       ncid=nccre('try.nc',ncnoclob,rcode)

I'd change the `ncnoclob' to `ncclob' in the above line to prevent reuse
of a previously existing netCDF dataset.

>       
> c define dimension and variable
>       rec_dim=ncddef(ncid,'index',ncunlim,rcode)
>       
>       a1_id=ncvdef(ncid,'a1',ncfloat,1,rec_dim,rcode)
> 
> c leave define mode
>       call ncendf(ncid,rcode)
> 
> c read in data and put into netcdf file
> 
>       do i=1,10
>           
>          read(1000,'(f5.2)') a1(i)
> 
>       end do
>       
>       call ncvpt(ncid,a1_id,1,10,a1,rcode)
> 
> c close netcdf file
>       call ncclos(ncid,rcode)
> 
>       end
> - ----------
> X-Sun-Data-Type: default
> X-Sun-Data-Name: second.f
> X-Sun-Content-Lines: 32
> 
>       program second
> 
>       include '/opt/netcdf/include/netcdf.inc'
> 
>       integer ncid,rcode,a1_id
>       real a1(5)
> 
> c open data file
>       open(1000, file='second.dat', status='old')
>       
> c open old netcdf file
>       ncid=ncopn('try.nc',ncwrite,rcode)
>       
> c get variable id
>       a1_id=ncvid(ncid,'a1',rcode)
> 
> c read in data and append to netcdf file
> 
>       do i=1,5
> 
>          read(1000,'(f5.2)') a1(i) 
> 
>       end do
>       
> 
>       call ncvpt(ncid,a1_id,11,15,a1,rcode)

The fourth argument above should be `5' rather than `15'.  The COUNTS
argument indicates how much data the `a1' variable contains.  In this
case, you're adding 5 additional values -- not 15.

> 
> 
> c close netcdf file
>       call ncclos(ncid,rcode)
> 
>       end

Please let me know if this helps.

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