NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
Dear all,As I have found out, netCDF should be built with HDF5 support to be able to benefit from compound types. I installed HDF5 and recompiled netCDF with HDF5 support and now I can benefit from compound types. First, I create the file to be written with:
*nf90_create(FILE_NAME, NF90_HDF5, ncid)* Up till here everything is OK. Then I define the compound type with: *nf90_def_compound(ncid, sizeof(ep_data), "ecopath_data", typeid)*where it gives the error: "NetCDF: HDF error STOP Stopped". If I further add the below line which defines a field for the compound type; without resolving the former error:
*nf90_insert_compound(ncid, typeid, "biomass", 1, field_typeid)*The error changes to " NetCDF: Not a valid data type or _FillValue type mismatch. STOP Stopped." All the errors come as runtime errors and the code compiles flawlessly. I am wrestling through the documentation but many commands miss explanation, e.g. *NF90_FILL_REAL* (to fill _FillValue), *NF90_PUT_VARA* (to flush array into the defined compound field) etc.
I would be glad if someone can help me on the matter, Regards, Ekin On 02/23/2012 11:36 AM, Ekin Akoglu wrote:
Dear all,This time around I have come up with a problem concerning compound types. I scripted a FORTRAN subroutine to write data in compound form into a .nc file and it compiles perfectly, however, it gives exactly the error "Unknown error" when I run the model calling the ncwrite function after model run once is completed. My FORTRAN subroutine file is attached. It gives the error when it reaches the line with "nf90_def_compound" function. My netCDF version is 4.1.3 and fortran compiler is gfortran 4.4.5. I had compiled netCDF with --disable-netcdf-4 option. Could this be the cause of the problem?Regards, Ekin On 02/23/2012 10:58 AM, Ekin Akoglu wrote:Dear all,I have come up with an answer to my question, that is, user-defined "compound types" in netCDF. The compound types are not very well documented under FORTRAN API, however, section for C API gives also an example where one can develop on for FORTRAN. If you are also interested please see: http://www.unidata.ucar.edu/software/netcdf/docs/netcdf-f90/Compound-Types.html#Compound-TypesThank you... Ekin On 02/22/2012 10:03 PM, Ekin Akoglu wrote:Dear all,I am quite inexperienced in using netCDF FORTRAN API and netCDF format in storing data. Currently I am struggling on a problem. I have user derived data types in FORTRAN with multiple components, i.e. fields, such as:ep_data%biomass ep_data%pob .... ms_data%biomass ms_data%age ...Is there a way to envelope the fields (after % sign) under e.g. "ms_data" and "ep_data" defined types when writing in a single netcdf file so that the "biomass" field of "ms_data" does not interfere with the "biomass" field of "ep_data" while reading it from the .nc file? I would be glad to hear if you have suggestions on the matter.Regards, -- *Ekin Akoglu* Research Assistant Institute of Marine Sciences Middle East Technical University P.O. Box 28, 33731 Erdemli, Mersin Turkey Web: www.ims.metu.edu.tr Email: ekin@xxxxxxxxxxxxxxx <mailto:ekin@xxxxxxxxxxxxxxx> Phone: +90 324 521 34 34 GSM: +90 506 554 03 90 Fax: +90 324 521 23 27 _______________________________________________ netcdfgroup mailing list netcdfgroup@xxxxxxxxxxxxxxxxFor list information or to unsubscribe, visit:http://www.unidata.ucar.edu/mailing_lists/-- *Ekin Akoglu* Research Assistant Institute of Marine Sciences Middle East Technical University P.O. Box 28, 33731 Erdemli, Mersin Turkey Web: www.ims.metu.edu.tr Email: ekin@xxxxxxxxxxxxxxx <mailto:ekin@xxxxxxxxxxxxxxx> Phone: +90 324 521 34 34 GSM: +90 506 554 03 90 Fax: +90 324 521 23 27 _______________________________________________ netcdfgroup mailing list netcdfgroup@xxxxxxxxxxxxxxxxFor list information or to unsubscribe, visit:http://www.unidata.ucar.edu/mailing_lists/-- *Ekin Akoglu* Research Assistant Institute of Marine Sciences Middle East Technical University P.O. Box 28, 33731 Erdemli, Mersin Turkey Web: www.ims.metu.edu.tr Email: ekin@xxxxxxxxxxxxxxx <mailto:ekin@xxxxxxxxxxxxxxx> Phone: +90 324 521 34 34 GSM: +90 506 554 03 90 Fax: +90 324 521 23 27 _______________________________________________ netcdfgroup mailing list netcdfgroup@xxxxxxxxxxxxxxxx For list information or to unsubscribe, visit: http://www.unidata.ucar.edu/mailing_lists/
-- *Ekin Akoglu* Research Assistant Institute of Marine Sciences Middle East Technical University P.O. Box 28, 33731 Erdemli, Mersin Turkey Web: www.ims.metu.edu.tr Email: ekin@xxxxxxxxxxxxxxx <mailto:ekin@xxxxxxxxxxxxxxx> Phone: +90 324 521 34 34 GSM: +90 506 554 03 90 Fax: +90 324 521 23 27
netcdfgroup archives: