[postgis-devel] [raster] Regular blocking refinement

[Mateusz Loskot]

*bump*

On 01/12/10 12:53, Mateusz Loskot wrote:
> Hi,
> 
> At the FOSS4G 2010 in Barcelona, I talked with Pierre about
> refinement of the idea of regular blocking configuration.
> 
> It seems to be still unclear if the regular blocking should define a
> complete grid of tiles with no tiles missing. The picture attached
> presents two tile coverages:
> - blue is regular thus complete
> - red is irregular (or rather incomplete regular coverage?, see below)
> 
> I understand Pierre's reasoning and I tend to agree with Pierre
> that both cases should be supported.
> 
> I tried to understand how the popular tiling schemes deal with similar
> configurations and I posted some questions to the tiling ml:
> 
> http://lists.osgeo.org/pipermail/tiling/2010-September/000008.html
> 
> The very first question is if both, the blue and red configurations,
> are/should be kosher.
> 
> It seems we do allow it in the specification:
> 
> "It is permissible to for regular_blocking rasters to omit some
> blocks/tiles (sparse rasters) in which case the missing tiles
> should be assumed to be all NODATA or zero valued."
> 
> However, what the "assumed to be all NODATA or zero valued" means?
> Does it mean we actually want to store those tiles in database
> or generate them on fly?
> Or, is this just an indication for clients: if a tile is missing in
> query result, cope with it on your own by generating NODATA tiles?
> 
> IMHO, this matter is unclear and it is important to specify behaviour
> on SQL level first. For example, SQL query against in coverage
> configured as regular blocking, returns no tuples for missing tiles.
> 
> It is important to clarify the "regular blocking" refers to virtual grid
> used as frame reference to place tiles, but empty cells in this grid are
> allowed.
> 
> I think the "no tuples for missing tiles" approach is better,
> because it does not imply any particular raster value: NODATA tile,
> zero'ed tile or transparent tile or anything else.
> Clients can cope with missing tiles as they like.
> 
> Comments are welcome.
> 
> Best regards,
> 
> 
> 
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-devel


-- 
Mateusz Loskot, http://mateusz.loskot.net
Charter Member of OSGeo, http://osgeo.org
Member of ACCU, http://accu.org