[Proj] Understanding 2D Helmert

[Kristian Evers]

Ah yes, I missed that proj was used in the example.
The problem is the same, just located in proj.c instead.

I have pushed a fix here: https://github.com/OSGeo/proj.4/pull/1162

While proj is mainly meant to be used to project geodetic
coordinates it doesn’t hurt to treat other input correctly
when it is possible. It is a bit closer to doing that now. I
am sure there are some exceptions still that I haven’t
found (the code in proj.c structured quite poorly)

/Kristian

On 26 Oct 2018, at 17:37, Even Rouault <even.rouault at spatialys.com<mailto:even.rouault at spatialys.com>> wrote:

On vendredi 26 octobre 2018 14:40:17 CEST Kristian Evers wrote:
If that is the case it should just be a matter of changing the
type of the output units internally in the operation. If you
submit a new ticket on the github page with a short description
I'll take a look at it within the next couple of days.

Things are fine. This is just that using 'proj' for such operation is
inapproprate. You should use the new 'cct' binary.

'proj' is just for plain-old projection methods that take long, lat as input,
whereas here input coordinates are in the projected space

echo 1000 0 0 | cct +proj=helmert +convention=coordinate_frame +theta=-162000

outputs
    707.1068       707.1068        0.0000           inf

(162000 arc-second = 45 deg)

(due to a current oddity of cct, you need to specify 3D coordinates, or add "-
z 0" in the parameters )

--
Spatialys - Geospatial professional services
http://www.spatialys.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20181026/1ad8b657/attachment.html>