Hi John, thanks for your help.
Our actual tiles are 70 megs each so not super easy to send, but i think
its easy to reproduce..
if there are 4 tiles like:
<netcdf location="file:P:/1.nc" section="0:10,0:10"/>
<netcdf location="file:P:/2.nc" section="0:10,11:20"/>
<!-- <netcdf location="file:P:/3.nc" section="11:20,11:20"/> -->
<netcdf location="file:P:/4.nc" section="11:20,0:10"/>
commenting or uncommenting 3.nc will give you a big 'zero' patch, not a
null (transparent) patch in the wms response.
we have done a workaround for this (and the problem of naming tiles
individually) by writing a script to generate the thredds config and to
identify/generate tiles with null values for the appropriate lat/longs,
so everything is working fine, but its not ideal ongoingly.
cheers,
-i
Le 4/05/2011 7:26 PM, John Caron a écrit :
On 5/3/2011 9:23 AM, Ivan PRICE wrote:
Another question regarding tiled aggregation in case anyone knows
anything about this..
our data is in a regular grid, but the grid is not always completely
rectangular, e.g. for instance when following a coastline there maybe
missing tiles where the land is when plotting data for the sea.
currently if they are left out of the thredds config they are
coloured according to the 'zero' value, and not the 'null' vaue. is
this configurable or do we need to ensure the tile coverage is always
completely rectangular ?
hmm, i would have thought you get missing values. do you have a test
case i can reproduce?
john
_______________________________________________
thredds mailing list
thredds@xxxxxxxxxxxxxxxx
For list information or to unsubscribe, visit:
http://www.unidata.ucar.edu/mailing_lists/
