On mardi 21 mai 2019 17:02:24 CEST Nyall Dawson wrote:
> Hi list,
> I see quite a lot of "IAU2000" CRS definitions on
> https://www.spatialreference.org/ref/?page=4&search=iau which are not
> available in the proj database.
> Is there any plans to integrate these systems into proj?
Anyway, paths forward for IAU integration into GDAL. All of which I could use help with.
(1) validate major update to "create_IAU2000.py" (by J-Christophe) and re-release WKT projection files to GitHub. This is a script to take the IAU published radius values and create the WTK projection files (per IAU-code) and initFile for proj4 can be created. We had to clarify the use of Reference_Meridian = 0.0 for both WKT1 and WKT2. You can wade through this confusing issue here: https://github.com/USGS-Astrogeology/GDAL_scripts/issues/34 . That should be all resolved in the latest version of the Python script and we can now move forward.
(2) Update script to create for what GDAL needs. Here is what is required (email from Even),
(3. longer-term). Lastly, I'm not sure what to do with spatialreference.org. I keep hearing it has been deprecated, so no updates, but it keeps hanging around. :-). If we need a new registry we can help stand something up (or maybe just use Github?). But, I would still like our codes to head over to the OGC CRS resolver (not sure the status for that effort though) - see: http://external.opengeospatial.org/twiki_public/CRSdefinitionResolver But ... by forcing the date in the name (e.g. IAU2000, IAU2015), I sort-of messed up RESTful "versioning" which should look more like (for Mars 49900 using the 2000 report would be something like "/def/crs/IAU/2000/49900" not "/def/crs/IAU2000/49900"). I guess for the resolver website we could just split the IAU and year...?
I would love to see progress made on this front - so again, I would welcome the help!
Also, as a reminder, J-Christophe, also has a fork of Mapserver (on his github site) with IAU projection code support. We need to test that too (as have used Mapserver for years and embarrassingly default to EPSG:4326! Fortunately degrees are degrees but not a good precedence for us to use all these years).