cities map not working with `plotly`

[GitHunter0]

Hi folks, for some reason the cities map is not working with plotly (while others like regions map are working flawlessly). Any idea what is happening?

library(plotly)
library(geobr)
map_regions <- geobr::read_region(year=2010) |> sf::st_cast("MULTIPOLYGON")
map_cities <- geobr::read_municipality(code_muni="all", year=2010) |> sf::st_cast("MULTIPOLYGON")

While this works:

plot_ly() %>% add_sf(data = map_regions) 

This returns Error in `[.data.frame`(x, i) : undefined columns selected:

plot_ly() %>% add_sf(data = map_cities) 

R version 4.1.1 (2021-08-10)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19044)

Matrix products: default

locale:
[1] LC_COLLATE=English_United States.1252 
[2] LC_CTYPE=English_United States.1252   
[3] LC_MONETARY=English_United States.1252
[4] LC_NUMERIC=C                          
[5] LC_TIME=English_United States.1252    
system code page: 65001

attached base packages:
[1] stats     graphics  grDevices utils    
[5] datasets  methods   base     

other attached packages:
[1] geobr_1.6.6   plotly_4.10.0 ggplot2_3.3.5

Thank you

[rafapereirabr]

Hi @GitHunter0 . Thanks for opening this issue. There could be some kind of conflict with data.table, but I'm not sure why.

library(plotly)
library(magrittr)
library(sf)
library(geobr)

# note the object has multiple classes
map_cities <- geobr::read_municipality(code_muni="all", year=2010)
class(map_cities)
> [1] "sf"         "data.table" "data.frame"

If if try plot_ly(), we get the error you pointed out.

plot_ly() %>% add_sf(data = map_cities) 

> Error in `[.data.frame`(x, i) : undefined columns selected

Now, when we remove the data.table class, then plot_ly() works.

map_cities <- as.data.frame(map_cities) %>% st_sf()
class(map_cities)
> [1] "sf"         "data.frame"

plot_ly() %>% add_sf(data = map_cities) 

This looks a bit strange, to be honest. It looks like a bug on plotly, but I cannot really grasp why it happens.

[GitHunter0]

Hey @rafapereirabr , it makes some sense I believe, AFAIK data.table has compatibility issues with sf.

Probably map_cities should not have data.table class as default, since other maps from geobr like map_regions (from the example above) do not have it.

But I'm glad you found a workaround, thank you.

[rafapereirabr]

Yes, I should proabably remove the data.table class from the data. I'll keep this issue open and will add this to my task list. Thanks,

[rafapereirabr]

Fixed in dev version with 45b99a20 .