# Re: [R] polygon shapefile from line edge coordinate list

From: Roger Bivand <Roger.Bivand_at_nhh.no>
Date: Sat, 08 Mar 2008 12:51:08 +0000 (UTC)

Murray Richardson <murray.richardson <at> utoronto.ca> writes:

>
> Hello,
>
> I am looking for advice on a task I am trying to complete.
>
> I have a 4 column dataframe defining the start and end coordinates of
> line edges (from a CGAL alpha shapes function to define concave hulls
> from point clusters). I would like to create polygon shapefiles from
> these line edges, presumably creating lines first and then polygons.

How many polygons in each data.frame? If more than one, how are they separated? You note below that there are multiples, are they flagged by for example an NA row, or is there a jump from (endX endY) to the next (startX startY)?

>
> e.g. columns are:
>
> startX startY endX endY
>
> where each row represents start and end coordinates of a line segment.

Are the line segments ordered in sequence. If they are, something like:

xy0 <- data.frame(sX=c(1,2,2,1,3,4,4,3), sY=c(1,1,2,2,3,3,4,4),   eX=c(2,2,1,1,4,4,3,3), eY=c(1,2,2,1,3,4,4,3))

brks <- logical(nrow(xy0))
for (i in 2:nrow(xy0)) brks[i] <- (xy0\$sX[i] != xy0\$eX[i-1]) &   (xy0\$sY[i] != xy0\$eY[i-1])
cbrks <- cumsum(brks)+1
# find the separate polygons

xy1 <- split(xy0, cbrks)

library(sp)
Plist <- vector(mode="list", length=length(xy1)) for (i in seq(along=Plist)) {
crds <- rbind(cbind(xy1[[i]]\$sX, xy1[[i]]\$sY),     c(xy1[[i]]\$eX[nrow(xy1[[i]])], xy1[[i]]\$eY[nrow(xy1[[i]])])) # close the polygon
Plist[[i]] <- Polygons(list(Polygon(crds)), ID=names(xy1[i])) }
# make a list of Polygons objects with IDs

Plist1 <- SpatialPolygons(Plist)
plot(Plist1, axes=TRUE)
SPDF <- SpatialPolygonsDataFrame(Plist1, data=data.frame(i=names(xy1),   row.names=names(xy1)))
# convert this to a SpatialPolygonsDataFrame object for export

library(maptools)
writePolyShape(SPDF, "chulls")

If the segments are not in "join up the dots" sequence, it will be (even) more messy. Please also consider whether the R-sig-geo list might not be more relevant.

Roger

>
> I am new to R spatial packages and I am not sure which one is best
> suited to this task. There seems to be a lot of options, which is great
> but hard to know which one to start with. Any suggestions on the best
> way to proceed with this? One further challenge is that the list
> includes line segments defining multiple polygons.
>
>
> Murray
>
> ______________________________________________
> R-help <at> r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help