[wfs-dev] SRS of LatLongBoundingBox

[Aleksandar Milanovic]

It seems that you discovered a bug in the CITE tests. It may be that this
bug is benign. For example, if these coordinates are not actually used for
anything in the tests, then the fix will be easy to do.

Alex

-----Original Message-----
From: wfs-dev-bounces at opengis.org [mailto:wfs-dev-bounces at opengis.org] On
Behalf Of David McCullough
Sent: August 31, 2004 7:18 PM
To: Wfs-Dev (E-mail)
Subject: RE: [wfs-dev] SRS of LatLongBoundingBox


Alex,

Yeah, I have my server coded up to return EPSG:4326 (even though our
features are in a projected coordinate system). However, what got me
questioning things again was looking at the GetCapabilities response in the
WfsCapabilities.xml file that is included wfsTestData1.0.0.zip test data
from OGC CITE. This test file has XML like this:

...
      <wfs:SRS>EPSG:32615</wfs:SRS>
      <wfs:LatLongBoundingBox minx="500000" miny="500000" maxx="501000"
maxy="501000"/> ...

Clearly, these coordinates are not lat/long degree values, so is this sample
data incorrect?

Thanks.

-----Original Message-----
From: Aleksandar Milanovic [mailto:risjak at telus.net]
Sent: Tuesday, August 31, 2004 6:45 PM
To: David McCullough; 'Wfs-Dev (E-mail)'
Subject: RE: [wfs-dev] SRS of LatLongBoundingBox


Hi Dave,

LatLongBoundingBox coordinates are *not* bound to the SRS of the feature
type declared in the Capabilities document. Rather, they are in EPSG:4326
with decimal degrees as units, if I remember well :).

Cheers,
Alex

-----Original Message-----
From: wfs-dev-bounces at opengis.org [mailto:wfs-dev-bounces at opengis.org]
On
Behalf Of David McCullough
Sent: August 31, 2004 5:59 PM
To: Wfs-Dev (E-mail)
Subject: [wfs-dev] SRS of LatLongBoundingBox


This question concerns the optional <LatLongBoundingBox>[s] that may be
returned in the <FeatureType> elements of a GetCapabilities response. I'm
having some confusion about what SRS the coordinates should be relative to. 

The WFS-capabilities.xsd schema file says that a <LatLongBoundingBox> should
indicate the edges of the enclosing rectangle in the SRS of the associated
feature. That seems clear, but then why is the element called
LatLongBoundingBox? That seems to me to imply that the coordinates are
geodetic, regardless of the associated feature's SRS. Which is correct?

Thanks.

-Dave


_______________________________________________
wfs-dev mailing list
wfs-dev at opengis.org http://mail.opengis.org/mailman/listinfo/wfs-dev



_______________________________________________
wfs-dev mailing list
wfs-dev at opengis.org http://mail.opengis.org/mailman/listinfo/wfs-dev