netcdf files served by pydap.

[Matt Savoie]

Hi there, 

I'm new to the group and pydap. In playing with the server (Pydap-3.1.RC1), I'm finding differences between the behavior of a raw netcdf file and that same file served through pydap. 

For example the dimensions look incorrect: (using netCDF4==1.0.4) 

Using the pydap server, projection and maxStrlen64 show up as dimensions, which doesn't seem right. 

>>> y = Dataset('http://localhost:8001/seaice_conc_daily_nh_f13_20070123_v01r00.nc') 

>>> y.dimensions 

OrderedDict([(u'maxStrlen64', ), (u'projection', ), (u'time', ), (u'xgrid', ), (u'ygrid', )]) 

>>> len(y.dimensions) 

5 

>>> z = Dataset('/sharehome/savoie/projects/pydap_new/myserver/data/seaice_conc_daily_nh_f13_20070123_v01r00.nc') 

>>> z.dimensions 

OrderedDict([(u'xgrid', ), (u'ygrid', ), (u'time', )]) 

>>> len(z.dimensions) 

3 

As a result of this, none of the CF aware clients, panoply, idv, etc. are able to plot the data from these files. Is this likely a bug with the netCDF handlers, or is it expected behavior and I will need to write separate handlers to provide the data in a CF compliant manner?

Thanks

Matt Savoie

NSIDC

[Matt Savoie]

Forgot to mention, if someone wants to look at the file I'm playing with, you can grab one via ftp:

ftp://sidads.colorado.edu/DATASETS/NOAA/G02202/north/daily/2007/


Thanks

Matt

[Roberto De Almeida]

Matt, can you send me the file so I can test it? If it's too big just an ncdump of the header (ncdump -h seaice_conc_daily_nh_f13_20070123_v01r00.nc) is enough.

Thanks,

--Rob

Matt Savoie

NSIDC

--
You received this message because you are subscribed to the Google Groups "pydap" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pydap+un...@googlegroups.com.
To post to this group, send email to py...@googlegroups.com.
Visit this group at http://groups.google.com/group/pydap?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--

Roberto De Almeida, PhD

http://dealmeida.net/
:wq

[Roberto De Almeida]

Hi, Matt!

I'll download a file and test it, thanks!

--Rob