Hi Ethan,
thanks for taking a look into this!
A couple more questions about the netCDF Subset service:
- Is it possible or will be to set a limit to the size of data that users can
download from the NCSS?
- For grids as point data, is the multiple variables with different vertical
coordinates issue to be solved in a next NCSS version or release?
Cheers!
On April 21, 2011 at 10:25 PM Ethan Davis <edavis@xxxxxxxxxxxxxxxx> wrote:
> Hi Marcos, all,
>
> Thanks for all the information on this. I am able to reproduce the
> problem now. I will dig into the problem and let you know when we have a
> fix.
>
> Thanks,
>
> Ethan
>
> On 4/15/2011 4:44 PM, Marcos Hermida wrote:
> > Hi all,
> >
> > here are some tests we did regarding the NetCDF Subset Service for
> > curvilinear grid as point data issue and could be helpful:
> >
> > If you sent this opendap request:
> >
> > http://oos.soest.hawaii.edu/thredds/dodsC/hioos/roms_forec/hiog/runs/HI-ROMS_Forecast_Model_Run_Collection_RUN_2011-04-14T00:00:00Z.ascii?lon_rho[46][60],lat_rho[46][60]
> >
> >
> > you qet these lon/lat values: -157.495325318087, 21.292617472685
> >
> > if you use them in a NCSS request like that:
> > http://oos.soest.hawaii.edu/thredds/ncss/grid/hioos/roms_forec/hiog/runs/HI-ROMS_Forecast_Model_Run_Collection_RUN_2011-04-14T00:00:00Z?var=temp&latitude=21.292617472685&longitude=-157.495325318087&time_start=2011-04-14T00%3A00%3A00Z&time_end=2011-04-21T00%3A00%3A00Z&temporal=point&time=2011-04-14T00%3A00%3A00Z&vertCoord=-0.983333333333333&accept=xml&point=true
> >
> > the lon/lat you get is: -157.622435719315, 21.1940855899557
> > so now, if we swap the lon_rho and lat_rho indices and send this request:
> > http://oos.soest.hawaii.edu/thredds/dodsC/hioos/roms_forec/hiog/runs/HI-ROMS_Forecast_Model_Run_Collection_RUN_2011-04-14T00:00:00Z.ascii?lon_rho[60][46],lat_rho[60][46]
> >
> > we get...-157.622435719315, 21.1940855899557 (as in the NCSS request!!)
> > And the last one, eta_rho is a 170 length array and xi_rho is an 70
> > length array, so let's choose a point with an index bigger than 69 for
> > the eta_rho and send a request:
> > http://oos.soest.hawaii.edu/thredds/dodsC/hioos/roms_forec/hiog/runs/HI-ROMS_Forecast_Model_Run_Collection_RUN_2011-04-14T00:00:00Z.ascii?lon_rho[70][60],lat_rho[70][60]
> >
> > we get: -157.708182949237, 21.3033444376878 and once again if we put
> > that on the NCSS request:
> >
> > http://oos.soest.hawaii.edu/thredds/ncss/grid/hioos/roms_forec/hiog/runs/HI-ROMS_Forecast_Model_Run_Collection_RUN_2011-04-14T00:00:00Z?var=temp&latitude=21.3033444376878&longitude=-157.708182949237&time_start=2011-04-14T00%3A00%3A00Z&time_end=2011-04-21T00%3A00%3A00Z&temporal=point&time=2011-04-14T00%3A00%3A00Z&vertCoord=-0.983333333333333&accept=xml&point=true
> >
> >
> >
> > what I get here is a 500 - Internal server error, and I guess there will
> > be an ArrayIndexOutOfBoundsException in the threddsServlet.log
> >
> > So I think it only returns data for requests where the lat/lon matches
> > points where, in this case, we have both indices for eta_rho, and xi_rho
> > lower than 70, and it fails any other case.
> >
> > Cheers!
>
> _______________________________________________
> thredds mailing list
> thredds@xxxxxxxxxxxxxxxx
> For list information or to unsubscribe, visit:
> http://www.unidata.ucar.edu/mailing_lists/
