htmlrgisr-sfr-leaflet

Import data from a leaflet map as an sf object

I want to import data as an sf object in R from a leaflet map. The map is this site: https://erickgn.github.io/mapafc/ I also have the HTML from the map as following:https://raw.githubusercontent.com/erickgn/mapafc/main/index.html.

Solution

  • I also have the HTML from the map as following:https://raw.githubusercontent.com/erickgn/mapafc/main/index.html.

    Then you have everything. Either save the page locally, either scrape it using xml2 package. If you look on the page source, you can find something like:

    geo_json_b75320e180b34bb88a8a9025dff8675e_add({"bbox": [-44.447264,
 -23.03329, -41.6957233, -22.2949485],[...]

    as seems that are your features, you can read it with sf::st_read or jsonlite package.

    A small example for the first JSON:

    library(rvest)
url <- "https://raw.githubusercontent.com/erickgn/mapafc/main/index.html"
text <- html_text(read_html(url))

    Now we have to locate two strings which are before and after json and take the part between. Please note +1, -22 -- the first one is quite obvious, the second one is a bit of trying to remove unnecessary new lines etc.

    library(stringi)

st <- stri_locate_first_fixed(text, "geo_json_b75320e180b34bb88a8a9025dff8675e_add(")[2]+1
fi <- stri_locate_first_fixed(text, "geo_json_b75320e180b34bb88a8a9025dff8675e.bindTooltip(")[1]-22

json <- substring(text, st, fi)

    And finally let's convert json to R objects:

    jsonlite::fromJSON(json)
#> $bbox
#> [1] -44.44726 -23.03329 -41.69572 -22.29495
#> 
#> $features
#>                                           bbox
#> 1   -43.59792, -22.82906, -43.58869, -22.82160
#> 2   -43.38023, -22.96123, -43.37173, -22.95453
#> 3   -43.50182, -23.03329, -43.49279, -23.02227
#> 4   -43.29931, -22.99099, -43.29163, -22.98606
[...]

    You can repeat similar steps for next json(s).

    And reading with sf package:

    library(sf)
a <- st_read(json)
#> Reading layer `OGRGeoJSON' from data source 
#> [...]
#>   using driver `GeoJSON'
#> Simple feature collection with 249 features and 16 fields
#> Geometry type: POLYGON
#> Dimension:     XY
#> Bounding box:  xmin: -44.44726 ymin: -23.03329 xmax: -41.69572 ymax: -22.29495
#> Geodetic CRS:  WGS 84
plot(a$geometry)

    Regards, Grzegorz