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

[netCDFJava #CPC-398807]: FW: question about large file support



Yes, a netcdf4 file is a valid hdf5 file. Howeverr, HDF5 doesnt understand the 
shared dimensions. But for many uses, it doesnt matter so much.

John

> Can netcdf4 files be read by hdf5 readers? because of the interoperability 
> between the two?
> Thanks
> Zhenping
> 
> Sent from my iPhone
> 
> > On Feb 11, 2015, at 12:52 PM, Unidata netCDF Java Support <address@hidden> 
> > wrote:
> >
> > Can you send the ncdump -h output?
> >
> > If dimension len > 2^31-1, we dont have a workaround presently.
> >
> > John
> >
> >> Here is the java print out the netcdf file (version 4) using the 
> >> toString() method in netcdf4 java library:
> >>
> >> netcdf 
> >> /home/zpli/MLS/G17/L0/Archive/IT_ABI-L0-M03_G17_s20142461945100_e20142462000180_c20142462000204.nc
> >>  {
> >> dimensions:
> >> number_of_packets = UNLIMITED;   // (6131690 currently
> >> number_of_data_bytes = UNLIMITED;   // (0 currently
> >> variables:
> >> int offset_to_packet(number_of_packets=6131690);
> >> :long_name = "index in the abi_space_packet_data variable of the start of 
> >> the packet";
> >> :_ChunkSize = 256; // int
> >>
> >> float percent_uncorrectable_L0_errors;
> >> :units = "percent";
> >> :long_name = "percent uncorrectable L0 errors";
> >>
> >> int size_of_packet(number_of_packets=6131690);
> >> :long_name = "size of each packet";
> >> :_ChunkSize = 256; // int
> >>
> >> byte abi_space_packet_data(number_of_data_bytes=0);
> >> :long_name = "GOES-R Advanced Baseline Imager (ABI) L0 CCSDS Space 
> >> Packets";
> >> :coverage_content_type = "physicalMeasurement";
> >> :_ChunkSize = 2048; // int
> >>
> >> // global attributes:
> >> :dataset_name = 
> >> "IT_ABI-L0-M03_G17_s20142461945100_e20142462000180_c20142462000204.nc";
> >> :naming_authority = "gov.noaa.goes-r";
> >> :iso_series_metadata_id = "a70be540-c38b-11e0-962b-0800200c9a66";
> >> :Conventions = "CF-1.6";
> >> :Metadata_Conventions = "Unidata Dataset Discovery v1.0, CF-1.6";
> >> :standard_name_vocabulary = "CF Standard Name Table (v18, 22 July 2011)";
> >> :title = "Advanced Baseline Imager (ABI) L0 CCSDS data packets";
> >> :summary = "Raw data reconstructed to unprocessed instrument data at full 
> >> space-time resolution with all available supplemental information to be 
> >> used in subsequent processing appended.";
> >> :keywords = "SPECTRAL/ENGINEERING > INFRARED WAVELENGTHS > SENSOR COUNTS, 
> >> SPECTRAL/ENGINEERING > VISIBLE WAVELENGTHS > SENSOR COUNTS";
> >> :keywords_vocabulary = "NASA Global Change Master Directory (GCMD) Earth 
> >> Science Keywords, Version 7.0.0.0.0";
> >> :license = "Unclassified data.  Access is restricted to approved users 
> >> only.";
> >> :institution = "DOC/NOAA/NESDIS > U.S. Department of Commerce, National 
> >> Oceanic and Atmospheric Administration, National Environmental Satellite, 
> >> Data, and Information Services";
> >> :date_created = "2014-09-03T20:00:20Z";
> >> :processing_level = "National Aeronautics and Space Administration (NASA) 
> >> L0";
> >> :time_coverage_start = "2014-09-03T19:45:10Z";
> >> :time_coverage_end = "2014-09-03T20:00:18Z";
> >> :timeline_id = "03";
> >> :orbital_slot = "GOES-West";
> >> :platform_ID = "G17";
> >> :instrument_type = "GOES R Series Advanced Baseline Imager";
> >> :instrument_id = "ABI-1";
> >> :production_data_source = "Test";
> >> :production_site = "WCDAS";
> >> :production_environment = "ITE";
> >> :id = "c4420190-4a71-4451-a2a2-b38c2db89296";
> >> }
> >>
> >> Notice the dimension number_of_data_bytes is unlimited but 0 currently. 
> >> The actual size is 3374823640 when using the ncdump. In netcdf Java API: 
> >> getLength() method for Dimension class is defined as int, the number 
> >> 3374823640 would return as negative for int. Do you have a sample code for 
> >> work around? Also, the variable offset_to_packet should be defined as 
> >> unsigned int, so that it could be larger than 0x7FFFFFFF. I am getting the 
> >> negative output values for this variable.
> >>
> >> Thanks for your help.
> >>
> >> Zhenping
> >>
> >> -----Original Message-----
> >> From: Unidata netCDF Java Support [mailto:address@hidden]
> >> Sent: Tuesday, February 10, 2015 3:42 PM
> >> To: address@hidden
> >> Cc: address@hidden; address@hidden
> >> Subject: [netCDFJava #CPC-398807]: FW: question about large file support
> >>
> >> Sorry for jumping in, I hope it is not too confusing.
> >> When you say "...when Java reads it, the dimension length is defined as 
> >> the (signed) integer..."
> >> do you mean the netcdf-java library or a program you are writing?
> >> If the latter, then it is possible to use the Java long type to properly 
> >> represent such a number.
> >> =Dennis Heimbigner
> >> Unidata
> >>
> >>
> >> Ticket Details
> >> ===================
> >> Ticket ID: CPC-398807
> >> Department: Support netCDF Java
> >> Priority: Normal
> >> Status: Closed
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: CPC-398807
> > Department: Support netCDF Java
> > Priority: Normal
> > Status: Closed
> >
> 
> 


Ticket Details
===================
Ticket ID: CPC-398807
Department: Support netCDF Java
Priority: Critical
Status: Closed