On Thu, Dec 3, 2009 at 12:41 AM, Maxime Tournier
<maxime.tournier@xxxxxxxxxxxx> wrote:
After trying to implement this, it seems we don't need both a concept map and
seems fine. It's quite convenient to use so far.
Also I came across a (minor) problem with this approach: what if you need to
modify the wrapped element ?
Isn't that already possible?
Here we should be able to modify...
template<G>
G foo(G& _g, int n)
{
LieWrapper<G> g(_g);
...
}
and here not....
template<G>
G foo(const G& _g, int n)
{
LieWrapper<const G> g(_g); // or even const LieWrapper<const G> g(_g);
...
}
Going more conform with Eigen::Map, could we call this wrapper/mapper LieMap? What do you think? To me it makes more clear that the object is life-time dependant on what it is initialized with.
- Hauke