draw a vector map

the result is an Encapsulated PostScript image

mapvec [-f float] [-m float] [-n float] [-o filename] configfile differencefile

-f float

limit fraction

-m float

required minimal difference as fraction of maximum

-n float

neighbourhood size

-o filename

output file

This programs draws a map with in each location a vector pointing in the direction of other locations (within a certain neighbourhood) that have the smallest difference with the current location.

See Map configuration file for the syntax of the configfile.

A vector v_i is calculated with the following formula:

Each vector depends on the dialect differences with locations in the neighbourhood. The size of the neighbourhood is set with the value of the option -n, which must be between 0 and 1. The maximum neighbourhood size 1 is equal to the geographic distance between the two locations that are farthest apart. The neighbourhood is limited to those locations that are not farther away than the given fraction of the maximum distance.

The algorithm needs all dialect differences to be larger than 0. If there are differences that are below a certain minimum as a fraction of the maximum difference, then all differences are increased to raise the smallest difference to the smallest allowed fraction. This fraction can be changed with the option -f.

Any vectors that are shorter than a certain fraction of the average vector length are omitted. The limit can be changed with the option -m.