NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.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.
NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.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.
On 6/2/2010 4:20 AM, Rich Signell wrote: > Jon, > >> The colour scale range is set per-variable, not per-dataset. >> The only sensible way I can see for a sysadmin to configure >> this accurately would be within the TDS catalog in NcML, for >> example: >> >> <variable name="temp"> >> <attribute name="defaltColorScaleRange" value="-5.0 35.0"/> >> </variable> >> >> (If we agreed on a syntax for this it would be an easy change >> in the WMS code I think.) > > So this would be an addition to the CF conventions, right? > It could be supplied in the original dataset or introduced via > something like NcML. It might be called simply "defaultRange" > as it could be used for other uses in addition to color scales > (e.g. default axis limits for line plots). > >> It might also be possible to provide a global set of defaults >> per CF standard name in a configuration file for variables >> that don't define a range explicitly. > > I like this idea as well, because it would allow much simpler > handling of this common case: there are a bunch of models > forecasting temperature in a region with a significant annual > cycle, and you want the default color range on the WMS to > change every month so that you can see spatial gradients > regardless of the time of year. > > This is why I like the current ability in ncWMS to edit the > defaults on a dataset-by-dataset, variable-by-variable basis. > > Rich