Bypasses the usual restrictions on combining sets of subcomponents from two separate objects with the same geometrical structure.
A context in ICE is usually a component type and a geometry, for example, "Edges of Sphere1". Most nodes in ICE enforce that their inputs and outputs have matching contexts. For example, you can't add the positions of points of one object with the positions of the points of another object. This is because, the number and order of the points of each object is generally different. However, there are scenarios where you know that two objects have the same geometry structure (same number and order of points, edges, polygons) and you want to bypass this constraint and operate on the components of both objects. For example, you want to blend the position of two objects of the same geometric structure. To do this connect the data coming from one of the objects into the Switch Context node. The data output the Switch Context node is no longer constrained to the input context.
The Switch Context node does not work if the input and output contexts have a different number of elements. Keep in mind that there are only a few ways of creating objects with the same geometric structure. For example, you could duplicate an existing object which usually keeps the geometric structure. The geometric structure is maintained when deforming the object. However, if you change the topology of the geometry (for example, adding and removing edges), the structure is most likely not compatible even if you end up with the same number of points, edges, and polygons. This is because the component order will have changed.