CODEWITHSUNDEEP
rfacetterratmapr-stars
UseR10085
UseR10085

Reputation: 8198

How to reorder facets in tmap r?

I want to reorder the facets of tmap raster plot. I am using the following code

library(tmap)
library(stars)
library(terra)

#Read some data
tif = system.file("tif/L7_ETMs.tif", package = "stars")
x = rast(tif)

#Assign some fake names to the bands
names(x) <- c("RF_04.Apr.2022_LL", "RF_13.Dec.2021_LL", "RF_15.Feb.2022_LL", "RF_17.Dec.2020_LL", "RF_18.Nov.2021_LL", "RF_19.Feb.2021_LL")

#Plot it using tmap r package
tm_shape(x) +
  tm_raster(style="quantile")+
  tm_facets(nrow = 3) +
  tm_layout(panel.labels = names(x), 
            legend.outside=T, legend.outside.position = "right",
            legend.position= c("center", "center"),
            legend.text.size = 1,
            legend.format = list(digits = 2, text.separator = "-"))+
  tm_compass(position = c("LEFT", "BOTTOM"))+
  tm_scale_bar(breaks = c(0, 0.5), text.size = 1, position = c("LEFT", "BOTTOM"))

which returns me the following output enter image description here

But I want to have the facets in the following sequence (i.e. in chronological order)

"RF_17.Dec.2020_LL", "RF_19.Feb.2021_LL", "RF_18.Nov.2021_LL",
"RF_13.Dec.2021_LL", "RF_15.Feb.2022_LL", "RF_04.Apr.2022_LL"

How can I do it?

Upvotes: 0

Views: 315

Answers (1)

margusl
margusl

Reputation: 17564

There seems to be some kind of an issue with layer names and tmap (v3) facets when changing SpatRaster layer order though subsetting. Here's a partial answer that might provide a workaround by first sorting filenames by encoded dates so layers in the resulting SpatRaster would be ordered chronologically, presumably fixing facet order:

library(tidyverse)

file_list <- c("RF_04.Apr.2022_LL", "RF_13.Dec.2021_LL", "RF_15.Feb.2022_LL", "RF_17.Dec.2020_LL", "RF_18.Nov.2021_LL", "RF_19.Feb.2021_LL")

# extract dates, use file_list vector for names
date_list <- 
  file_list |> 
  stringr::str_extract("(?<=_).*(?=_)") |> 
  lubridate::dmy() |>
  setNames(file_list)
date_list
#> RF_04.Apr.2022_LL RF_13.Dec.2021_LL RF_15.Feb.2022_LL RF_17.Dec.2020_LL 
#>      "2022-04-04"      "2021-12-13"      "2022-02-15"      "2020-12-17" 
#> RF_18.Nov.2021_LL RF_19.Feb.2021_LL 
#>      "2021-11-18"      "2021-02-19"

# sort by dates, extract names
(file_list_sorted <- sort(date_list) |> names())
#> [1] "RF_17.Dec.2020_LL" "RF_19.Feb.2021_LL" "RF_18.Nov.2021_LL"
#> [4] "RF_13.Dec.2021_LL" "RF_15.Feb.2022_LL" "RF_04.Apr.2022_LL"

# import rasters
x = rast(file_list_sorted)

# alter names to only keep date values, part between _ & _
names(x) <- names(x) |> stringr::str_extract("(?<=_).*(?=_)")
names(x)
#> [1] "17.Dec.2020" "19.Feb.2021" "18.Nov.2021" "13.Dec.2021" "15.Feb.2022" "04.Apr.2022"


# continue with tmap

Upvotes: 1

Related Questions