[PROJ] Vertical CRS transformations and vertical grid shift

[Even Rouault]

> Well, when providing that coordinate, I got them from a sample Dataset in
> EPSG:31255 so they should be valid.
> 
> Weird thing:
> cs2cs "EPSG:31255" "EPSG:4326"
> 88000 273000
> 
> 45d52'39.123"N 16d50'57.28"E 0.000 (as you said)
> 
> However:
> gdaltransform -s_srs EPSG:31255 -t_srs EPSG:4326
> 88000 273000
> 
> 14.5025684938377 47.5898013589131 46.4711159374565
> 
> And same thing if I draw a point in QGIS: SRID=31255;POINT(88000 273000)
> and I switch the map projection to 4326, that rendered point on the map has
> the same coordinates reported by gdaltransform.
> 
> What do you think about it?

Ah, the reason is axis order. EPSG:31255 uses Northing,Easting, and cs2cs honours it. 
Whereas gdaltransform uses the "traditional GIS axis order", ie easting, northing.

So with PROJ master:

$ echo 273000 88000 600 | PROJ_NETWORK=ON src/cs2cs "EPSG:31255+5778" "EPSG:
3035+9274"
2729881.62	4659643.84 599.75

With EPSG:5621 "EVRF2007 height" instead of EPSG:5778 "EVRF2000 height", I don't get 
any change in Z, because there's no registered transformation between EPSG:5778 "GHA 
height" and it.

Even

-- 
Spatialys - Geospatial professional services
http://www.spatialys.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20200416/d7e13ee0/attachment-0001.html>