On Wed, 14 Aug 2019 at 11:02, Andrea Aime <[email protected]> wrote:
> On Wed, Aug 14, 2019 at 7:49 PM Jody Garnett <[email protected]> > wrote: > >> This proposal is targeted towards 2.16? We are days away from the RC >> deadline ... >> > Andrea did you have any feedback on this? I would expect hold off until after freeze, wait a month and back port? Same goes for the existing TIME and ELEVATION support for vectors, they > could be done via CQL_FILTER. But CQL_FILTER cannot be discovered by a > generic OGC WMS client, custom dimensions are instead part of the WMS > standard. > That makes perfect sense, a real gap in the standards not having a good overview of available values for attributes. - Do you have an example of WMS 1.1 and WMS 1.3 snippet advertising custom >> dimensions, an identical approach as for time >> > > It's really the same as the exiting support for raster custom dimensions > (WMS does not tell you if a particular layer is vector or raster). > Here is an example taken from a GeoServer serving a raster data set with > dimensions time, elevation and REFERENCE_TIME: > Thanks. > > >> - there is a risk of conflict between dimension parameters >> (population=100 in the demo) and other keys, perhaps something similar to >> vendor options should be considered? Imagine if I had a dimension named >> format, angle or env? >> > > The WMS specification handle this by asking the client to add a "DIM_" > prefix in the request, so if the dimension is called "REFERENCE_TIME" the > client will add a parameter &DIM_REFERENCE_TIME=value in the KVP request. > Okay there is a small mistake in the proposal where &poluation=10000 was used. I am +1 on the proposal, +0 on including it in the 2.16-RC timeframe...
_______________________________________________ Geoserver-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/geoserver-devel
