[netCDFJava #TJD-270946]: HDF5 reader

Hi Chris: if the XML file is in the user block, i can use it, so thats a 
reasonable solution for us. I'll have a look, thanks for the sample files, ill 
let you know. Do you have a pointer to any docs for these XML files?

> Hi John,
> The files that we have from NPOESS have the XML in the file before the actual 
> data.  Just in case you are interested,
> here is another of our files of sample data.  So for us, it looks like it 
> will all be in a single file.  The HDF_UserBlock
> element if the XML is supposed to be the "Meta-data" for the file, at least 
> that is what the NPOESS documentation calls
> it.  So far, we don't need to associate two files together.  I imagine we 
> will simply parse the XML
> separately, using netCDF for the HDF5 part, and just use a custom class to 
> retrieve the data from the XML.  I guess that
> is really not a big help to you.  I also know that NPOESS puts some of the 
> data for the file in the file name, such as the
> orbit number.
> 
> But for our purposes, the netCDF java interface is just fine for retrieving 
> the data.  In some instances I think it does
> better than the libraries provided by the HDF5 group.  I will let you know if 
> I find anything new.
> 
> v/r
> 
> Chris Webb
> Software Engineer, WDAC
> Harris Corporation
> (402) 294-4358
> 
> 
> 
> -----Original Message-----
> From: Unidata netCDF Java Support [mailto:address@hidden]
> Sent: Friday, October 02, 2009 10:10 AM
> To: Webb, Christopher A CTR USAF AFWA AFWA/WDAC
> Cc: address@hidden; Webb, Christopher A CTR USAF AFWA AFWA/WDAC
> Subject: [netCDFJava #TJD-270946]: HDF5 reader
> 
> Hi Chris, thanks for that information. Im looking at a random file example i 
> have, happends to be VIIRS. It seems like all
> the interesting info is available, even without the region references. Im 
> wondering if for your purposes you need those,
> or if the way the netcdf-java interface presents the info is adequate?
> 
> Im also seeing that they have important metadata in external XML files, 
> though im not sure how to associate a data file
> with an XML file. Do you know anything about that?
> 
> John
> 
> > Hello,
> > I finally got a response from the NPOESS people about the data that netCDF 
> > doesn't currently read.  As near as I could
> > tell, netCDF is capable of reading all the data from the files with the 
> > exception of the
> > "<CollectionShortName>_Gran_<n>" data sets.  I figured you may be 
> > interested in what NPOESS wrote to me, so here it is.
> >
> > "Chris,
> >
> > The datasets CollectionShortName_Gran_n, where n is {Agg, 0..# of granules 
> > in package}, are basically storage bins of
> > metadata. The dataset portion is a hyperslab that points to datasets in the 
> > "AllData" group. The All_Data group contains
> > the raw data.  The reason for the "_0", "_Agg" items is that if the user 
> > requested aggregated granules, then there are
> > multiple sets of metadata - sets for each granule, and 1 set for the entire 
> > aggregated data.
> >
> > If you want to look at only granule-level (not aggregation-level) metadata, 
> > then you can ignore the "_Agg" items.
> >
> > Thanks,
> > Bill"
> >
> > The HDF5 libraries decode the data in these data sets to looks something 
> > like this:
> > [0:7144 {(0)-(11)}, 0:9840 {(0)-(11)}, 0:12208 {(0,0)-(11,95)}, etc...]
> >
> > So, if you really wanted too, you could just probably return the data as a 
> > String, and let the users determine what they
> > want to do with it.  For me, it looks like our project will not need the 
> > data found in that dataset:).
> >
> >
> > v/r
> >
> > Chris Webb
> > Software Engineer, WDAC
> > Harris Corporation
> > (402) 294-4358
> >
> >
> > -----Original Message-----
> > From: Unidata netCDF Java Support [mailto:address@hidden]
> > Sent: Thursday, July 30, 2009 8:04 PM
> > To: Webb, Christopher A CTR USAF AFWA AFWA/WDAC
> > Subject: [netCDFJava #TJD-270946]: HDF5 reader
> >
> > Hi Chris:
> >
> >
> > Release 4.0.48.5 disables HDF5 region references (npoess "granules") for 
> > now, until i can figure out how these work.
> > This allows the rest of the dataset to be used. Please let me know if you 
> > see any problems. If you understand what use
> > case granules serve, i'd love to know that also.
> >
> > thanks!
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: TJD-270946
> > Department: Support netCDF Java
> > Priority: Normal
> > Status: Open
> >
> >
> 
> 
> Ticket Details
> ===================
> Ticket ID: TJD-270946
> Department: Support netCDF Java
> Priority: Normal
> Status: Open
> 
> 
> 


Ticket Details
===================
Ticket ID: TJD-270946
Department: Support netCDF Java
Priority: Normal
Status: Open