[R-sig-Geo] maptools_0.4-1 (and spdep_0.2-17) released

[Roger Bivand]

I would like to announce the release of new maptools and spdep revisions. 

I would also like to suggest that users consider thinking through 
upgrading before actually doing so - the internal representation of 
polylist objects changes in this release of maptools, ramping up towards 
the introduction of new-style classes later this year. 

The release tries to work out how to handle plot order and holes in
relation to the many shapefiles "in the wild" that do not follow ESRI
recommendations about holes having counterclockwise ring direction. This
means that both Map2poly() and plot.polylist() may give warnings that were
not there before. On the other hand, a major weakness of plot.polylist()
has been removed - it could overplot internal polygons without warning,
giving the wrong colour in choropleth maps - this should no longer happen
in polylists created using the new Map2poly from shapefiles.

Please contact me directly or via this list if things go wrong (or at
useR! in Vienna), I have tried to test the maptools release on difficult
shapefiles, but I am sure there are stranger shapefiles around than I've
seen yet.

spdep code for constructing neighbour objects from polylists has been 
rewritten in C, and should be faster and less memory-intensive.

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
e-mail: Roger.Bivand at nhh.no