[gdal-dev] BAG CRS

Fri Apr 22 09:51:46 PDT 2022

Yeah, I agree with your assessment. I hadn’t realized that you could set the option externally when I started the thread. I agree with you that the user likely set it themselves given what you have shown thus far, so I’ll sort it out with them. Apologies for taking some of your time.

Thanks again,
André

From: Even Rouault <even.rouault at spatialys.com>
Sent: April 22, 2022 13:47
To: Vautour, André (INT) <Andre.Vautour at Teledyne.com>; gdal-dev at lists.osgeo.org
Subject: Re: [gdal-dev] BAG CRS

External Email

André,

hum ok. it might perhaps be possible that WKT2 could come when extract the vertical part but I still can't reproduce that trying:

$ gdal_translate autotest/gcore/data/byte.tif byte.bag -a_srs 'COMPOUNDCRS["foo",PROJCRS["NAD27 / UTM zone 11N",BASEGEOGCRS["NAD27",DATUM["North American Datum 1927",ELLIPSOID["Clarke 1866",6378206.4,294.978698213898,LENGTHUNIT["metre",1]]],PRIMEM["Greenwich",0,ANGLEUNIT["degree",0.0174532925199433]],ID["EPSG",4267]],CONVERSION["UTM zone 11N",METHOD["Transverse Mercator",ID["EPSG",9807]],PARAMETER["Latitude of natural origin",0,ANGLEUNIT["degree",0.0174532925199433],ID["EPSG",8801]],PARAMETER["Longitude of natural origin",-117,ANGLEUNIT["degree",0.0174532925199433],ID["EPSG",8802]],PARAMETER["Scale factor at natural origin",0.9996,SCALEUNIT["unity",1],ID["EPSG",8805]],PARAMETER["False easting",500000,LENGTHUNIT["metre",1],ID["EPSG",8806]],PARAMETER["False northing",0,LENGTHUNIT["metre",1],ID["EPSG",8807]]],CS[Cartesian,2],AXIS["(E)",east,ORDER[1],LENGTHUNIT["metre",1]],AXIS["(N)",north,ORDER[2],LENGTHUNIT["metre",1]],USAGE[SCOPE["Engineering survey, topographic mapping."],AREA["North America - between 120°W and 114°W - onshore. Canada - Alberta; British Columbia; Northwest Territories; Nunavut. Mexico. United States (USA) - California; Idaho; Nevada; Oregon; Washington."],BBOX[26.93,-120,78.13,-114]],ID["EPSG",26711]],VERTCRS["ellipse",VDATUM["NAD83(2011) / UTM zone 10N + ellipse"],CS[vertical,1],AXIS["ellipsoid height (h)",up,LENGTHUNIT["metre",1]]]]'

$ gdalinfo byte.bag -mdd xml:BAG | grep VERT_CS
            <gco:CharacterString>VERT_CS["ellipse",VERT_DATUM["NAD83(2011) / UTM zone 10N + ellipse",2005],UNIT["metre",1],AXIS["Ellipsoid height",UP]]</gco:CharacterString>

(the ,2005 here isn't really appropriate. it should rather be ,2002,  but such formulations of CompoundCRS with a VerticalCRS that expressed an ellipsoidal height are not ISO-19111 compliant)

Anyway if you can find a way to reproduce the issue with a CRS provided through the GetProjection()/SetProjection() API of GDAL, a fix would be welcome. If it is the user who provides WKT2 with a manual VAR_VERT_WKT option, then I'd say it is their responsibility to provide the right value.

Even
Le 22/04/2022 à 18:35, Vautour, André (INT) a écrit :
Sorry Even,

I should have done more research on my end up front before sending this. A client told us his WKT 2.0 BAG was created with GDAL. Looking at it some more, it was the vertical CRS that was WKT 2.0.

Their vertical CRS looked as follows:
<gco:CharacterString>VERTCRS["ellipse",VDATUM["NAD83(2011) / UTM zone 10N + ellipse"],CS[vertical,1],AXIS["ellipsoid height (h)",up,LENGTHUNIT["metre",1]]]</gco:CharacterString>

Cheers,
André

From: Even Rouault <even.rouault at spatialys.com><mailto:even.rouault at spatialys.com>
Sent: April 22, 2022 13:22
To: Vautour, André (INT) <Andre.Vautour at Teledyne.com><mailto:Andre.Vautour at Teledyne.com>; gdal-dev at lists.osgeo.org<mailto:gdal-dev at lists.osgeo.org>
Subject: Re: [gdal-dev] BAG CRS

External Email

André,

I don't confirm this trying:

$ gdal_translate autotest/gcore/data/byte.tif byte.bag

$ gdalinfo byte.bag -mdd xml:BAG | grep PROJCS
            <gco:CharacterString>PROJCS["NAD27 / UTM zone 11N",GEOGCS["NAD27",DATUM["North_American_Datum_1927",SPHEROID["Clarke 1866",6378206.4,294.978698213898,AUTHORITY["EPSG","7008"]],AUTHORITY["EPSG","6267"]],PRIMEM["Greenwich",0,AUTHORITY["EPSG","8901"]],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]],AUTHORITY["EPSG","4267"]],PROJECTION["Transverse_Mercator"],PARAMETER["latitude_of_origin",0],PARAMETER["central_meridian",-117],PARAMETER["scale_factor",0.9996],PARAMETER["false_easting",500000],PARAMETER["false_northing",0],UNIT["metre",1,AUTHORITY["EPSG","9001"]],AXIS["Easting",EAST],AXIS["Northing",NORTH],AUTHORITY["EPSG","26711"]]</gco:CharacterString>

This is WKT1

I'm wondering if the confusion might come from looking at the CRS reported by gdalinfo, which will always be WKT2 in recent GDAL versions whatever the encoding in the source dataset itself ?

Or perhaps you try to write a Geographic 3D CRS ? in which case WKT2 will be used since there's no WKT1 representation for this

$ gdalwarp -overwrite autotest/gcore/data/byte.tif byte.bag -t_srs EPSG:4979 -ot float32

$ gdalinfo byte.bag -mdd xml:BAG | grep GEOGCRS
            <gco:CharacterString>GEOGCRS["WGS 84",ENSEMBLE["World Geodetic System 1984 ensemble",MEMBER["World Geodetic System 1984 (Transit)"],MEMBER["World Geodetic System 1984 (G730)"],MEMBER["World Geodetic System 1984 (G873)"],MEMBER["World Geodetic System 1984 (G1150)"],MEMBER["World Geodetic System 1984 (G1674)"],MEMBER["World Geodetic System 1984 (G1762)"],MEMBER["World Geodetic System 1984 (G2139)"],ELLIPSOID["WGS 84",6378137,298.257223563,LENGTHUNIT["metre",1]],ENSEMBLEACCURACY[2.0]],PRIMEM["Greenwich",0,ANGLEUNIT["degree",0.0174532925199433]],CS[ellipsoidal,3],AXIS["geodetic latitude (Lat)",north,ORDER[1],ANGLEUNIT["degree",0.0174532925199433]],AXIS["geodetic longitude (Lon)",east,ORDER[2],ANGLEUNIT["degree",0.0174532925199433]],AXIS["ellipsoidal height (h)",up,ORDER[3],LENGTHUNIT["metre",1]], [... snip ....] ,ID["EPSG",4979]]</gco:CharacterString>

(I see that gdalinfo also issues a "ERROR 1: PROJ: proj_create_compound_crs: components of the compound CRS do not belong to one of the allowed combinations of http://docs.opengeospatial.org/as/18-005r4/18-005r4.html#34" when reading this, which I suspect comes from the fact that it tries to create a CompoundCRS from this 3D GeogCRS and the "VERT_CS["unknown", VERT_DATUM["unknown", 2000]]", so writing 3D GeogCRS isn't really appropriate for BAG)

Even
Le 22/04/2022 à 18:01, Vautour, André (INT) a écrit :
Hi all,

It has just come to my attention that the BAG driver in GDAL is writing the CRS with a WKT codeSpace and a WKT 2.0 string as the code. While I am fairly sure we started that incorrect practice here at CARIS with a WKT 1.0 string, this is the first time I’ve seen a WKT 2.0 string written in there, and I am therefore concerned about interoperability.

I think we can all agree that what makes the most sense is to write an EPSG codebase with the EPSG code if we have it, and I think falling back to a WKT 1.0 string if it is not EPSG is a reasonable fallback if it is not EPSG, given how it has been used in the past, but I am worried that older software won’t recognize the WKT 2.0 string.

Would you be willing to take a pull request that does that? Worst-case, would you be fine with changing that WKT 2.0 string to a WKT 1.0 string?

Regards,
André

_______________________________________________

gdal-dev mailing list

gdal-dev at lists.osgeo.org<mailto:gdal-dev at lists.osgeo.org>

https://lists.osgeo.org/mailman/listinfo/gdal-dev

--

http://www.spatialys.com

My software is free, but my time generally not.

--

http://www.spatialys.com

My software is free, but my time generally not.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/gdal-dev/attachments/20220422/7aa5454b/attachment-0001.html>