Re: [eigen] Using a Transformation as mapped expression type

[Christoph Hertzberg <chtz@xxxxxxxxxxxxxxxxxxxxxxxx>]

Yes, this would be possible, but nobody cared to implement it yet.
See this bug-entry:
http://eigen.tuxfamily.org/bz/show_bug.cgi?id=192

If you (or someone) knows a bit about the internals, this might be a 
nice possibility for contributing a patch (have a look at how 
Quaternion, QuaternionBase, Map<Quaternion> are implemented).

Also see the linked bug 1129, which suggest introducing

  id.asAffine()

as alternative to `Map<Affine3f>(id.data())` (similar to `id.array()`)

Cheers,
Christoph

On 02/08/2019 15.10, Janos Meny wrote:
> Hey,
>
> Why is it not possible to do something like this:
>
> Matrix4f id = Matrix4f::Identity();
> Map<Affine3f> map(id.data());
>
> Apparently the Transformation class does not model the dense public
> interface, so it is not "mappable". But isnt the Transformation class just
> a wrapper around a static dense matrix? Couldnt it therefore just pass
> through the stride, size etc. from the underlying dense matrix type?
>
> Best regards
>
> Janos

--
 Dr.-Ing. Christoph Hertzberg

 Besuchsadresse der Nebengeschäftsstelle:
 DFKI GmbH
 Robotics Innovation Center
 Robert-Hooke-Straße 5
 28359 Bremen, Germany

 Postadresse der Hauptgeschäftsstelle Standort Bremen:
 DFKI GmbH
 Robotics Innovation Center
 Robert-Hooke-Straße 1
 28359 Bremen, Germany

 Tel.:     +49 421 178 45-4021
 Zentrale: +49 421 178 45-0
 E-Mail:   christoph.hertzberg@xxxxxxx

 Weitere Informationen: http://www.dfki.de/robotik
  -------------------------------------------------------------
  Deutsches Forschungszentrum für Künstliche Intelligenz GmbH
  Trippstadter Strasse 122, D-67663 Kaiserslautern, Germany

  Geschäftsführung:
  Prof. Dr. Jana Koehler (Vorsitzende)
  Dr. Walter Olthoff

  Vorsitzender des Aufsichtsrats:
  Prof. Dr. h.c. Hans A. Aukes
  Amtsgericht Kaiserslautern, HRB 2313
  -------------------------------------------------------------