NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.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.

Re: [netcdfgroup] NetCDF 4.1-beta and OPeNDAP/coordinate variable question

Hi Dennis,
I agree with Steven that hiding the coordinate variables seems inexplicably weird. If people do not want access to these data they are free to ignore them, but much client software depends on being able to read the coordinate data as variables in their own right. I also wonder if it is really correct to say that the original nc-dap code did not provide access to coordinate variables. 

"
Yes, If you look at the underlying DAP, you will
see that coordinate arrays are represented by a Grid
construct which has a "maps" section.
You can see this by looking at, for example:
    http://test.opendap.org/opendap/data/nc/123.nc.dds
"
That may be, but traditionally in earlier versions of OPeNDAP, I can certainly find examples on-line where the coordinate arrays are presented as variables in their own right. Here's one (of numerous possibilities) for example: 

http://tds.hycom.org/thredds/dodsC/glb_current_analysis.html
Why should NetCDF 4 semantics be required to get information that is perfectly presentable using NetCDF 3? 

 -Dave Brown
 NCL development team




------------------------------

Message: 2
Date: Mon, 01 Feb 2010 15:01:41 -0700
From: Dennis Heimbigner <dmh@xxxxxxxxxxxxxxxx>
To: Mary Haley <haley@xxxxxxxx>
Cc: netcdfgroup@xxxxxxxxxxxxxxxx
Subject: Re: [netcdfgroup] NetCDF 4.1-beta and OPeNDAP/coordinate
        variable question
Message-ID: <4B674F45.5090307@xxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
This seems to be setting "--enable-dap" internally, which is what I want. 
That is correct; it is on by default.


For example, if I do an "ncdump -h" on both the "served" file, and a
downloaded version of the file, I get the following:

When you say "downloaded version; how are you downloading it?
In any case, this is, I believe, consistent with the behavior
of the original nc-dap C++ code, namely that map variables are not
included in the output. Originally, I did include them, but there were objections because this behavior was different from the nc-dap version. 

=Dennis Heimbigner
 Unidata



------------------------------

Message: 6
Date: Tue, 2 Feb 2010 14:34:22 -0000
From: <stephen.pascoe@xxxxxxxxxx>
To: <netcdfgroup@xxxxxxxxxxxxxxxx>
Subject: Re: [netcdfgroup] NetCDF 4.1-beta and OPeNDAP/coordinate
        variable        question
Message-ID:
        <EB1E7CB92F5B35459E0B926D2A614DB60B5BC296@xxxxxxxxxxxxxxxxxxxxxxxxxx>
Content-Type: text/plain;       charset="US-ASCII"



In any case, this is, I believe, consistent with the behavior of the

original nc-dap C++ code,
namely that map variables are not included in the output. Originally, 
I did include them, but

there were objections because this behavior was different from the

nc-dap version.

This sounds bizarre.  Is it ncdump that hides  coordinate variables in
OPeNDAP datasets or are they not visible in the NetCDF-C API either? I 
can't see how we can use OPeNDAP as a transparent "on the wire" NetCDF
if it misses out coordinate variables.

S.

---
Stephen Pascoe  +44 (0)1235 445980
British Atmospheric Data Centre
Rutherford Appleton Laboratory
  • 2010 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the netcdfgroup archives: