Replace geometries from a list in sf

问题

Suppose I have the following sf data frame:

library(sf)

nrows <- 10
geometry = st_sfc(lapply(1:nrows, function(x) st_geometrycollection()))
df <- st_sf(id = 1:nrows, geometry = geometry)

And I also have the following list:

mylist = list('2'=st_point(c(-73,42)), '3'=NA)

I want to replace the geometry from the second observation with the point in the list. I had thought about doing the following:

st_geometry(df[names(mylist),]) <- st_sfc(mylist)

But this throws an error:

"Error in vapply(lst, class, rep(NA_character_, 3)) : values must be length 3, but FUN(X[[2]]) result is length 1"

I found the following workaround by first eliminating the NA values:

condition <- mylist[!is.na(mylist)]
st_geometry(df[names(condition),]) <- st_sfc(condition)

Is there a better way to do this? Can I force the NA elements in mylist to be empty points?

回答1:

That's not working because in.

mylist = list('2'=st_point(c(-73,42)), '3'= NA)

3 is not a POINT but a logical, which (probably) can not be "forced" in any way into a sf object.

You can circumvent this by substituting your the NA elements of mylyst to empty POINTS beforehand. For example:

mylist[[which(is.na(mylist))]]  <- st_point()
st_geometry(df[names(mylist),]) <- st_sfc(mylist)

, giving:

> df
Simple feature collection with 10 features and 1 field (with 10 geometries empty)
geometry type:  GEOMETRY
dimension:      XY
bbox:           xmin: -73 ymin: 42 xmax: -73 ymax: 42
epsg (SRID):    NA
proj4string:    NA
   id                 geometry
1   1 GEOMETRYCOLLECTION EMPTY
2   2           POINT (-73 42)
3   3              POINT EMPTY
4   4 GEOMETRYCOLLECTION EMPTY
5   5 GEOMETRYCOLLECTION EMPTY
6   6 GEOMETRYCOLLECTION EMPTY
7   7 GEOMETRYCOLLECTION EMPTY
8   8 GEOMETRYCOLLECTION EMPTY
9   9 GEOMETRYCOLLECTION EMPTY
10 10 GEOMETRYCOLLECTION EMPTY

HTH.

来源：https://stackoverflow.com/questions/46385301/replace-geometries-from-a-list-in-sf

标签

list