Is there a (non-deprecated) way to determine the coordinates within the polygons of this shape file which I imported to R?

I have downloaded a shape file of Canadian federal districts. When I load the file into the namespace, it is a SpatialPolygonsDataFrame. Examining an individual polygon, I can see the following:

I can see that these are in the UTM system. I would like to be able to convert these into the normal longitude/latitude system using decimal units. The SPDF object has attributes: data, polygons, plotorder, bounding box and proj4string.

I have tried, for example:

sf::st_as_sf(MAP_HERE, coords = c('lon', 'lat'), crs = 4326)

In the resulting sf dataframe, I cannot find a way to access the coordinates of the new object in long/lat pairs. When I run st_coordinates() on the resulting dataframe, I can see that these have not been changed.

Ideally, I would like to access the coordinates of the bounding boxes in the original SPDF, and use those to calculate centroids for individual polygons.

Is there a way to do this?

You should read the data into an sf object from the shapefile and skip the SpatialPolygonsDataFrame step:

library(sf)
map = st_read("./path/to/mymapdata.shp")

then you use st_transform to convert to lat-long, which is epsg code 4326:

map_trans = st_transform(map, 4326)

then if you want to see the coordinates in full:

st_coordinates(map_trans)

Gets you something like this:

             X        Y L1 L2 L3
[1,] -81.47276 36.23436  1  1  1
[2,] -81.54084 36.27251  1  1  1
[3,] -81.56198 36.27359  1  1  1
[4,] -81.63306 36.34069  1  1  1
[5,] -81.74107 36.39178  1  1  1
[6,] -81.69828 36.47178  1  1  1

where L1 groups each feature (row) in the data, and L2 and L3 encode for multiple polygons and holes in these features.