SetRelativeLayout doesn't handled with positive/negative location offset.

[BDisp]

Actually the LayoutSubview is passing GetBoundsOffset () location in the parameter but this hasn't no impact on the SetRelativeLayout because this that not handles the location offset (positive or negative). Thus, for now, until the content offset is implemented, is more appropriate to pass an Point.Empty.

[tig]

I discovered this as part of #3269.

What I figured out is SetRelativeLayout has no need to know the location, it just needs the size. Now, in #3269 it looks like this:

    /// <summary>
    ///     Adjusts <see cref="Frame"/> given the SuperView's ContentSize (nominally the same as
    ///     <c>this.SuperView.ContentSize</c>)
    ///     and the position (<see cref="X"/>, <see cref="Y"/>) and dimension (<see cref="Width"/>, and
    ///     <see cref="Height"/>).
    /// </summary>
    /// <remarks>
    /// <para>
    /// If <see cref="X"/>, <see cref="Y"/>, <see cref="Width"/>, or <see cref="Height"/> are
    /// absolute, they will be updated to reflect the new size and position of the view. Otherwise, they
    /// are left unchanged.
    /// </para>
    /// </remarks>
    /// <param name="superviewContentSize">
    ///     The size of the SuperView's content (nominally the same as <c>this.SuperView.ContentSize</c>).
    /// </param>
    internal void SetRelativeLayout (Size superviewContentSize)

...

    private void LayoutSubview (View v, Size contentSize)
    {
        v.SetRelativeLayout (contentSize);
        v.LayoutSubviews ();
        v.LayoutNeeded = false;
    }

This simplifies and clarifies things substantially. I've added a ton of additional unit tests in #3269 that give confidence it's all working correctly.

I don't recommend we merge your change, and instead focus on ensuring #3269 is ready. It is close!

[BDisp]

No worries. I'll close all my PR then. The #3269 was done as is because @dodexahedron saw that the location wasn't used at all and he done well. But now I'm proving that the location can be handled, and must, in the SetRelativeLayout. I give up then,

[BDisp]

I'm reopen this because I'm convinced that the SetRelativeLayout should handled the location offset, although until now it wasn't never used.

[BDisp]

Closing this because it is really not necessary for SetRelativeLayout to deal with negative locations and it is more than proven with @tig's excellent work in #3323. Thanks.