State of matrix concatenation, and workarounds

[johan-overbye]

What is the state of the matrix concatenation discussion/implementation? I've searched the issues and there seem to be multiple approaches in play at the moment.

As a user of nalgebra I'm missing this feature a lot. My best workaround at the moment seems to be:

• Construct a new fixed-size Matrix, which fits both of the matrices I want to concatenate, with arbitrary values. (All my matrices are currently fixed-size.)
• Call target_matrix.fixed_view_mut::<n, m>(x, y).clone_from(&source_matrix) for each source Matrix. Note that n and m can't be inferred, which makes it pretty clunky.

Any insight into when there might be a feature for this as well as better workarounds is much appreciated! Thanks a lot to the nalgebra contributors for this beautiful crate.

EDIT: My workaround above actually only seems to work if the source matrix has one column. (Haven't tied with one row.) Attempting to clone_from a 3x3 matrix into the top of a 4x3 matrix, I get:

error[E0599]: the method `clone_from` exists for struct `Matrix<f32, Const<3>, Const<3>, ViewStorageMut<'_, f32, Const<3>, Const<3>, Const<1>, Const<4>>>`, but its trait bounds were not satisfied

    = note: the following trait bounds were not satisfied:
            `ViewStorageMut<'_, f32, Const<3>, Const<3>, Const<1>, Const<4>>: Clone`
            which is required by `Matrix<f32, Const<3>, Const<3>, ViewStorageMut<'_, f32, Const<3>, Const<3>, Const<1>, Const<4>>>: Clone`     

EDIT 2: copy_from works though. (T is f32.)