Bug using tmap to plot world map

[MichaelChaoLi-cpu]

Dear maintainer,

When we use tmap plot a world map in R 4.1.2, there is a mistake.

library(tmap)
library("rnaturalearth")
world <- ne_countries(scale = "medium", returnclass = "sp")
test <- tm_shape(world) +
  tm_polygons(col = "scalerank") 

tmap_save(test, filename = "test.jpg")

The warning message is as follows:

Warning message:
The shape world is invalid. See sf::st_is_valid 

However, when I use 4.0.3, no such mistake or warning message appears.

Here is the version info and package version.

> packageVersion('tmap')
[1] ‘3.3.3’
> version
               _                           
platform       x86_64-w64-mingw32          
arch           x86_64                      
os             mingw32                     
system         x86_64, mingw32             
status                                     
major          4                           
minor          1.2                         
year           2021                        
month          11                          
day            01                          
svn rev        81115                       
language       R                           
version.string R version 4.1.2 (2021-11-01)
nickname       Bird Hippie  

Thank you.

[mtennekes]

Probably this has something to do with the underlying libraries or the spatial drivers (proj, geos and gdal) which may have been updated along with R itself.

You can try to repair the object with sf::st_make_valid.

[Robinlovelace]

Closing this issue as it seems not to be related to {tmap}. Keep us posted if you're still seeing this issue @MichaelChaoLi-cpu, do you get it with the plot() function also?

[MichaelChaoLi-cpu]

plot() is ok. However, the tmap does not work, even run sf::st_make_valid.

library(tmap)
library("rnaturalearth")
library(sp)

world <- ne_countries(scale = "medium", returnclass = "sp")
test <- tm_shape(world) +
  tm_polygons(col = "scalerank") 
tmap_save(test, filename = "test.jpg")

world <- ne_countries(scale = "medium", returnclass = "sf")
world2 <- sf::st_make_valid(world)
(test2 <- tm_shape(world2) +
  tm_polygons(col = "scalerank") )
tmap_save(test2, filename = "test2.jpg")

world <- ne_countries(scale = "medium", returnclass = "sp")
world@data <- world@data %>% dplyr::select('level')
plot(world)
jpeg(file="test3.jpeg", width = 297, height = 210, units = "mm", quality = 300, res = 300)
plot(world)
dev.off()

test2

test3

So I think that might be a bug.

Thanks.

[Robinlovelace]

Aha reopening now... Can anyone else reproduce this?

[Nowosad]

Try disabling S2 at the beginning of your code -- sf_use_s2(FALSE).

[MichaelChaoLi-cpu]

sf_use_s2(FALSE) works. Thank you @Nowosad @Robinlovelace @mtennekes