As an FYI, Jody is obliquely referring to a decision taken by OGC
recently, which is that due to the confusion of what SRS "means"
regarding ordinate order, the new "URI" form of SRS coding will
*always* adhere to the "order as defined by the authority", no
excuses allowed, while the old form of coding will have some kind of
legacy behavior allowed. I will have to look up the pronouncement to
be sure of what that legacy behavior is, if it is defined at all.
P.
On Sep 29, 2005, at 4:31 PM, Jody Garnett wrote:
Martin Desruisseaux wrote:
Jody Garnett a écrit :
Martin - could epsg-hsql support this format right now? It does
produce a CRS with the ordinates in the correct order ... or does
this break our assumptions about "Authority".
I'm not sure to understand. Does epsg.xml#4326 have axis in the
(EAST,NORTH) axis order instead of the "official" (NORTH,EAST)?
The one that looks like a URI follows the "offical" use. The
traditional AUTHORY one does not.
http://www.opengis.net/gml/srs/epsg.xml#4326
-(NORTH,EAST)
EPSG:4326 aka "SRS AUTHORITY"
-(EAST,NORTH)
Even if the epsg.xml#4326 file does not exists, I guess that it is
defined in some document? Does the document said something about
axis order?
Note that the files does not have to exist - this is simply an URI.
Anyway, to come back on the issue of axis order with epsg-hsql, I
realize that some users will wants (EAST,NORTH) order anyway (no
matter what the EPSG database said), and that some mechanism may
be needed in order to give this control to users. I also suspect
that axis order is not the only issue. EPSG sometime uses some
particular units like gradians instead of degrees. I guess that
some users will want to force usage of degrees for all angular
measurements.
One possible approach is to create a wrapper around authority
factories. This is the subject of http://jira.codehaus.org/browse/
GEOT-694
Interesting idea, however the interpritation of http://
www.opengis.net/gml/srs/epsg.xml#4326 is clear and could be
supported right away. It is the traditional use that is
illspecified, and the best fix seems to be define something else
clearly.
Evil idea - consider it the "http:" Authority ....
Yes, we can do that. Not sure what the name should be, but I think
that considering a new authority may be a good solution.
Right now I'm a little bit stuck with my attempt to create a
release. I try to create a release with both Maven 1 and Maven 2
(the later being an experimental release), because I really want
transitive dependencies. Maven 2 work well except for the jjtree
and javacc plugins (required for building module/main) which are
giving me a lot of trouble. Once the release is out, I will look
at this axis order issue.
Okay, thanks for working on the release BTW. It is good to share
some of this work around.
Jody
-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads,
discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel
-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel
