[gdal-dev] FWD: OGC seeks comment on revised GML in JPEG 2000

[Even Rouault]

Hi,

I'm also trying to read that - beeeeeeeep - too long document and I guess I
should also read GMLCOV to make sense of it....

One thing among others that is not clear to me is if the coordinates of GML
features that can be embedded in the GMLJP2CoverageCollection are expressed in
pixel coordinates or in georeferenced coordinates. I may have miss something in
the doc... The example at page 47-48 would lead me to think this is pixel
coordinates because of the bounding box of the coverage collection.

At first sight, I'd say that the current GMLJP2 reader in GDAL should be able to
cope with GMLJP2 2.0 (limited to retrieving SRS + geotransformation matrix) with
few, if any, changes (metadata or features left apart). The write part would
need changes of course. "feature 67 in the reader requirements box " seems to be
an extra annoyance to handle. I'm wondering who will really implement such a
chek in the reading part, so it suse is probably just an opportunity to burn a
few extra trees by people printing the spec.

Oh well...

Even

> Mateusz Łoskot <mateusz <at> loskot.net> writes:
>
> > Hi,
> >
> > I think this may be of interested within the GDAL community:
> >
> > http://www.opengeospatial.org/pressroom/pressreleases/1965
>
> Hi,
>
> First time ever I tried to send feedback to OGC. Let's see if my mail went
> through. I commented two issues:
>
> 1. It looks like OGC has defined that origin in georeferencing means the
> centre of pixel in change number 6:
> Clarification of the use of CRS and Rectified grid coverage / image
> georeference with origin point (at pixel center). Is this interpretation
> right and could it be expressed in some fool proof way, like
> "Origin of the RectifiedGrid is placed at the centre point of the corner
> pixel"
>
> 2. Throughout the paper srsNames are given in two different ways.
> - srsName="http://www.opengis.net/def/crs/EPSG/0/4326"
> - srsName="urn:ogc:def:crs:EPSG::4326"
> In one example there is even this.
> - srsName="EPSG:4326"
>
> I believe that it means that both http and urn versions can be used but
> there was not enough room in the 91 pages long document to tell that.
> Perhaps the meaning is to be explicit and therefore there is a reference to
> another document with number OGC 07-092r3
> http://portal.opengeospatial.org/files/?artifact_id=30575
> However, reading OGC 07-092r3 does not help, that document does not mention
> at all that srsName="http://www.opengis.net/def/crs/EPSG/0/4326" could be
> used but it knows only the urn syntax.
>
> The same thing in with units of measures. There is a table 3 "URIs for
> units-of-measure" that gives examples of how to refer to UoM and they are
> like
> - http://www.opengis.net/def/uom/OGC/1.0/metre or
> - http://www.opengis.net/def/uom/EPSG/6.3/9001
>
> But in XML examples UoMs are expressed mostly as
> - uom="urn:ogc:def:uom:EPSG::9001"
>
> There seems to be also at least one copy-paste error in the examples:
> <Envelope srsName="http://www.opengis.net/def/crs/EPSG/0/4326"
> axisLabels="Lat Long" uomLabels="deg deg" srsDimension="2">
> <lowerCorner>270379.500 3942462.000</lowerCorner>
> <upperCorner>518842.500 3942462.000</upperCorner>
> </Envelope>
>
> -Jukka Rahkonen-
>
>
> _______________________________________________
> gdal-dev mailing list
> gdal-dev at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/gdal-dev