What does visualintersect in GetFeatures do?

[TimWhiteaker]

According to the docs, the visualintersect argument of the GetFeatures method Indicates whether only features that visibly intersect the geometry should be returned.

What does "visibly intersect" mean? Is there a difference between visibly intersecting something and actually intersecting something?

[TimWhiteaker · Accepted solution]

A visualintersect value of true means the symbol used for the feature is taken into account. So for example if you represented a city (point feature class) with a point symbol consisting of a large circle that is 100 pixels across in screen units, then if the search geometry intersects that circle (as if the circle were a polygon feature in the map), then that city feature will be returned even if the search geometry doesn't actually intersect the city's latitude and longitude point coordinates.  At least that's what I have determined based on experimentation.

[ThomasEmge]

It becomes even more evident if you think about it in a 3D context. Imagine that you have building footprints that are extruded by an attribute and you do the selection in screen coordinates. The building geometry itself is only 2D on the ground and only the layer visual representation is in 3D. This combination can make the selection tricky as well and hence the argument.