I need a simple understanding of nontopological geometry.

I am completely new in this area so please correct me if I am wrong.

As far as I understand in topological geometry the object(say a line) can be bend but in nontopological geometry object can be only be moved but can not be changed in a new shape.

I came to know about nontopological geometry while reading a documentation about afrigis mapserver. Here is a little from the documentation where I found the term nontopological which may have its source the ESRI Shapefile Technical Description (page 1).

A shapefile stores nontopological geometry and attribute information for the spatial features in a data set. The geometry for a feature is stored as a shape comprising a set of vector coordinates. Because shapefiles do not have the processing overhead of a topological data structure, they have advantages over other data sources such as faster drawing speed and edit ability


I would not apply the term "non-topological" to a geometry. Basically, the shape of a geometry can be modified without affecting its topology, but this does not imply that you do not modify the shape of a geometry when you modify its topology.

I would use it for non-topological data structure or non topological geoprocessing. You have to look if the tool/data explicitly handles to topology internally. For example, does moving the point at the intersection of to lines also moves both lines ? Does moving the boundary between two polygons affect both polygons.

For instance, a shapefile is a non topological data structure. Each geometry is unaware of its neighbours. It can be modified in any way (moving or reshaping) and this will not affect the other geometries. Whereas in a GRASS feature class (for example), if you cannot modify the shared boundary of two polygons without affecting both.

In a non topological data structure, a shared boundary is stored once for each polygons

In a topological data structure, a shared boundaries is stored once, with attributes about its right and left objects.

  • Can you please explain a little more in the context of edited portion, I mean from the documentation point of view? – Farsan Rashid Sep 18 '15 at 9:57

Back when shapefiles were created, the predominate data format for Esri software was the Arc/Info coverage. Coverages used topological relationships of lines to define polygons (e.g., arc 2 forward, arc 7 backward, arc 5 forward,...).

This structure was very efficient from a data storage standpoint, but very inefficient for rendering polygons, since it required reading the polygon arc list (PAL) file, then scanning the ARC file twice, once to determine the number of vertices in each segment, and a second pass to actually add the vertices to an allocated array to hand off to the rendering engine.

Shapefiles were designed to be as efficient as possible for rendering, so a non-topological "whole feature" storage technique was used. This does not mean that the polygons in shapefiles do not have topology (they do!), just that topology is intrinsic to each shape, not derived from edges.

The drawback to the "whole shape" storage mechanism (which is shared with enterprise geodatabase, file geodatabase, and SDC files), is that editing boundaries shared in common is a lot more work (since the vertices are stored twice, aka "double-digitized'). This ArcUser article has more details.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.