Hey all,
I'm running into a bit of API consistency hunting wrt handling null
return values for GridCoverageReader.read in general and its use in
CatalogBuilder in particular.
API wise, it'd be enough for AbstractGridCoverage2DReader.read to state
null is a valid return value. So far it only says "@return a {...@linkplain
GridCoverage grid coverage} from the input source."
That's alright, all over the place the norm is whenever you call read,
you might get null. What is not that clear is in which situations that
could happen. If the requested envelope is outside the coverage
envelope, then returning null is alright. But what if you have a mosaic
with sparser data and hence a requested envelope, even if inside the
coverage bounds, overlaps no raster in the mosaic? I would say it's
equally fine to return null instead of a fake coverage?
Now, my problem with CatalogBuilder, is that its buildCoverage method
reads a test coverage accounting for the 4x4 bottom left pixels or so,
and if the reader returns null it throws an exception. Problem with that
is that it doesn't let me configure a raster mosaic (from ArcSDE) where
that corner has no raster (the mosaic is irregular, does not form a
rectangular shape).
So question is what to do in that situation. We could either require
coverage readers to return something, even a fake coverage, iif the
requested envelope overlaps the coverage's original envelope, or we
could fall back in CatalogBuilder to read the whole coverage if that
little optimization returned null. Opinions?
Cheers,
Gabriel
--
Gabriel Roldan
OpenGeo - http://opengeo.org
Expert service straight from the developers.
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Geoserver-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geoserver-devel
