data frame object to use as as.linnet object

Question

I have a data.frame object which I can easily convert to a spatialpointdataframe then convert that to a spatiallinesdataframe but then when I tried to cover to a as.linnet it does not read marks

         X       Y roadID
1 177321.3 3378163      1
2 177321.4 3378168      1
3 177321.4 3378168      1
4 177321.5 3378177      1
5 177321.5 3378186      1
6 177321.5 3378195      1

then I make this data.frame to a SpatialPointsDataFrame

coordinates(roaDF1) <- c("X","Y")
proj4string(roaDF1)=proj4string(trtrtt)

class       : SpatialPointsDataFrame 
features    : 100412 
extent      : 143516.4, 213981, 3353367, 3399153  (xmin, xmax, ymin, ymax)
crs         : +proj=utm +zone=17 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs 
variables   : 1
names       : roadID 
min values  :      1 
max values  :    347

then using SpatialPointsDataFrame to convert a SpatialLinesDataFrame ( found this code online )

LineXX <- lapply(split(roaDF1, roaDF1$roadID), function(x) Lines(list(Line(coordinates(x))), x$roadID[1L]))

linesXY <- SpatialLines(LineXX)
data <- data.frame(roadID = unique(roaDF1$roadID))
rownames(data) <- data$roadID
lxy <- SpatialLinesDataFrame(linesXY, data)
proj4string(lxy)=proj4string(trtrtt)

now lxy looks like this

> lxy
class       : SpatialLinesDataFrame 
features    : 347 
extent      : 143516.4, 213981, 3353367, 3399153  (xmin, xmax, ymin, ymax)
crs         : +proj=utm +zone=17 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs 
variables   : 1
names       : roadID 
min values  :      1 
max values  :    347

finally I want to concert this lxy to a linnet object

> W3.orig=as.linnet(lxy)
Warning messages:
1: In spatstat::linnet(vertices = V, edges = edges, sparse = TRUE) :
  edge list should not join a vertex to itself; ignored
2: In as.linnet.SpatialLines(lxy) :
  Internal error: could not map data frame to lines

it is not because roaDF1 has 100421 features when I make it 20K it still gives the same error.

Any help ?

Please upload some example data that demonstrate this problem. — Adrian Baddeley, Apr 19 '20 at 01:30
I created a csv file https://www.dropbox.com/s/9pcr6tlinit1c3d/SE.csv?dl=0 — iHermes, Apr 19 '20 at 03:09

Adrian Baddeley · Accepted Answer · 2020-04-20T02:25:15.443

This is occurring because the line coordinate data include repeated vertices (i.e. two successive rows of coordinate data are identical).

There is a warning message about 'joining a vertex to itself' which is an important hint.

Although the message says that this was 'ignored' (i.e. repeated vertices were deleted), the code for handling marks has failed to handle this, and it gives the second warning and gives up.

I will write a fix for as.linnet.SpatialLines to handle repeated vertices. You will still get the warning!

PS: a fix for maptools::as.linnet.SpatialLines has been submitted to the maintainers of maptools and I expect it will be included in the next release.

Dr Baddeley, thank you! I wrote a for-loop and removed the repeated vertices and successfully created as.linnet.SpatialLines — iHermes, Apr 20 '20 at 00:48

data frame object to use as as.linnet object

1 Answers1

Linked