[netCDF #RJZ-836571]: problem converting netcdf4 files to netCDF4-classic

Hello Cathy,

Thank you for providing the file; it certainly helped me debug this issue!

The problem appears to be that the uses a group, `Grid`.  There isn't an easy 
way to remove the group with `ncdump` or `nccopy`, although NCO may let you do 
this.  Alternatively, I was able to fix this with an intermediary step. 

I dumped the file you provided to a `.cdl` file as follows:

    $ ncdump precip.imerge.mon.202312.nc > tmp.cdl

I then edited the cdl file to remove the `group: Grid {` line near the start of 
the file, and removed the corresponding closing '}' at the end of the file. I 
was then able to generate a binary `.nc` file using `ncgen` as follows:

    $ ncgen -7 -o tmp.nc tmp.cdl

Running `ncdump -k tmp.nc` returns `netCDF-4 classic model`.

That at least should answer the question; I will look into if there is an 
easier way to do this, but at the moment I'm not certain there is.

I hope this helps!

-Ward

> Hi all
> 
> We have run into issues converting files from netCDF4 to netCDF4-classic.
> 
> The problem is this
> 
>   nccopy -k 4 precip.imerg.mon.202312.nc a.nc
> NetCDF: Attempting netcdf-4 operation on strict nc3 netcdf-4 file
> Location: file ../../ncdump/nccopy.c; line 657
> 
> and similarly for
> 
>   nccopy -k "netCDF-4 classic model"  a.nc
> 
> ncdump gives
> 
>    ncdump -k precip.imerg.mon.202312.nc
> netCDF-4
> 
> and
> 
>   ncdump -hs precip.imerg.mon.202312.nc
> 
> skip stuff
> 
>       float precip(time, lon, lat) ;
>           precip:DimensionNames = "time,lon,lat" ;
>           precip:Units = "mm/hr" ;
>           precip:units = "mm/hr" ;
>           precip:coordinates = "time lon lat" ;
>           precip:_FillValue = -9999.9f ;
>           precip:CodeMissingValue = "-9999.9" ;
>           precip:LongName = "Merged microwave-infrared-gauge
> precipitation estimate" ;
>           precip:_Storage = "chunked" ;
>           precip:_ChunkSizes = 1, 145, 1800 ;
>           precip:_DeflateLevel = 6 ;
>           precip:_Endianness = "little" ;
> 
> skip stuff
> 
>  // group attributes:
>           :GridHeader =
> "BinMethod=ARITHMETIC_MEAN;\nRegistration=CENTER;\nLatitudeResolution=0.1;\nLongitudeResolution=0.1;\nNorthBoundingCoordinate=90;\nSouthBoundingCoordinate=-90;\nEastBoundingCoordinate=180;\nWestBoundingCoordinate=-180;\nOrigin=SOUTHWEST;\n"
> ;
>   } // group Grid
> 
> GrADS won't plot the files
> 
> Scanning self-describing file:  precip.imerg.mon.202312.nc
> gadsdf: SDF file has no discernable X coordinate.
>   To open this file with GrADS, use a descriptor file with an XDEF entry.
>   Documentation is at
> http://cola.gmu.edu/grads/gadoc/SDFdescriptorfile.html
> 
> But of course the variable does have a good X coordinate
> 
> CDO gives a similar error to the NCO command for converting
> 
> cdo -f nc4c -O copy precip.imerg.mon.202312.nc
> <http://precip.imerg.mon.202312.nc> ~/a.nc <http://a.nc>
> Segmentation fault
> 
> Is there anyway to convert the files from the command line from netCDF4
> to netCDF4-classic? We had the same issue with some unrelated files
> produced from python code.
> 
> I've put one file into /Public/csmith/
> 
> precip.imerg.mon.202312.nc
> 
> ftp ftp.cdc.noaa.gov
> 
> Thanks for any advice!
> 
> Cathy Smith
> 
> --
> ----------------------------------------------------------------
> Cooperative Institute for Research in Environmental Sciences ,
> University of Colorado Boulder
> Working in the NOAA Physical Sciences Laboratory
> https://psl.noaa.gov/people/cathy.smith/
> https://cires.colorado.edu/people/cathy-smith
> 
> Emails about data/webpages may get quicker responses by emailing
> address@hidden
> 
> 


Ticket Details
===================
Ticket ID: RJZ-836571
Department: Support netCDF
Priority: Normal
Status: Closed
===================
NOTE: All email exchanges with NSF Unidata User Support are recorded in the 
Unidata inquiry tracking system and then made publicly available through the 
web.  If you do not want to have your interactions made available in this way, 
you must let us know in each email you send to us.