Re: [eigen] Mapping array of scalars into quaternions

[Benoit Jacob <jacob.benoit.1@xxxxxxxxx>]

Hi,

Thanks for the patch. There's something strange about it: it seems to
undo Gael's last commit? Could it be that you coded against the
Quaternion.h file of before Gael's commit, and then generated a patch
against the latest eigen2 ? If yes you need to merge that; how you do
that depends on how you manage your code, but for example if you
worked on a local clone my_eigen2, then basically all you need to do
is to make sure your changes are committed locally, then do "hg pull
-u" to pull the latest changes into your clone, which will create a
new head, and then "hg merge" to merge, edit Quaternion.h to resolve
conflicts, and "hg resolve -m /path/to/Quaternion.h". See help here:
http://eigen.tuxfamily.org/index.php?title=Developer%27s_Corner#Mercurial_workflow

Also, it would help to read some rationalizations:
* why the EIGEN_STRONG_INLINE in front of cross product? Did you
benchmark an improvement here?
* why the a.QuaternionType::operator* syntax, since a is already of
QuaternionType type? Was this useful for compilation?

Thanks
Benoit

2009/11/9 Mathieu Gautier <mathieu.gautier@xxxxxx>:
> Hi,
>
> I add some unit test based on the Map class's ones.
>
> Here the changeset :
>
> * add Map<Quaternion> test based on Map from test/map.cpp
> * replace implicit constructor AngleAxis(QuaternionBase&) by an explicit
> one, it seems ambiguous for the compiler
> * replace explicit constructor with conversion type Quaternion(Quaternion&)
> with a convert function: conflict between constructor.
>
> --
> Mathieu Gautier
>
>
>