getNbrs | R Documentation |
Lists the indices (or identifiers if these are provided) of the Delaunay neighbours of each point in the set of points being triangulated/tessellated.
getNbrs(object, interior = NULL)
object |
An object of class |
interior |
Either a rectangle, i.e. a numeric vector of length 4,
If |
If interior
is specified then Delaunay neighbours are listed
only for those points which lie in interior
. Note that
these neighbours need not themselves lie in interior
.
Note also that it is possible for points i
and j
to
be neighbours even though the “clipped” versions of the tiles
are discontiguous.
A (named) list of length equal to the number points in the set
being triangulated/tessellated, or to the number of such points that
lie in interior
if that argument was specified. The names
of the list are the identifiers of the points as specified by
id
if id
was specified in the call to deldir()
that produced object
. If id
was not specified, then
the names are the indices of the points, coerced to character mode.
The entries of this list are vectors of the identifiers or indices of the Delaunay neighbours of the point corresponding to that entry.
Be careful about addressing the entries of the list returned
by this function. If id
was not specified in the call
to deldir()
that produced object
, then the names
of this list are the point indices coerced to character mode.
If interior
was specified then the name of i
-th entry
of the list will not in general be i
. E.g. given that
xxx
is the list returned by this function, xxx[[14]]
will not in general give the Delaunay neighbours of point 14.
Instead, specify xxx[["14"]]
or xxx[[id[14]]]
where
id
is the vector of identifiers supplied in the call to
deldir()
.
See deldir()
for references.
set.seed(42)
x <- runif(60)
y <- runif(60)
dxy <- deldir(x,y,rw=c(0,1,0,1))
nbrs <- getNbrs(dxy,interior=c(0.2,0.8,0.2,0.8))
nbrs[["14"]] # Correct.
nbrs[[14]] # Incorrect.
names(nbrs)[14] # The result is 42.
# Thus nbrs[[14]] actually gives the Delaunay neighbours of point 42.
# Demonstrate that neighbours can have discontiguous clipped tiles.
if(require(polyclip)) {
x <- c(0.38,0.44,0.04,0.97,0.43,0.96,0.89,0.64,0.97,0.62,0.33,0.35,
0.40,0.78,0.04,0.75,0.68,0.17,0.26,0.51)
y <- c(0.68,0.98,0.76,0.57,0.85,0.19,0.27,0.83,0.69,0.24,0.04,0.14,
0.22,0.48,0.20,0.72,0.01,0.38,0.51,0.00)
CP <- list(x=c(0.72,0.93,0.76,0.61,-0.03,-0.04,0.41),
y=c(0.46,0.76,0.94,1.03,1.01,0.37,0.31))
dxy <- deldir(x,y,rw=c(0,1,0,1))
TL <- tile.list(dxy)
plot(TL,labelPts=TRUE)
plot(TL[16],clipp=CP,fillcol="orange",labelPts=TRUE,add=TRUE)
polygon(CP,border="red")
nbrs <- getNbrs(dxy,interior=CP) # Tiles are clipped to CP.
# Note that point 14 is a neighbour of point 16, even though their
# clipped tiles do not meet.
}
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.