The goal of sfext is to extend existing functions from the {sf} package and offer a range of additional options for working with simple feature objects, bounding boxes, and data frame objects with coordinates or other spatial information.
Installation
You can install the development version of sfext like so:
# pak::pkg_install("elipousson/sfext")Usage
Extending existing {sf} functions
sfext is built around existing sf functions but designed to offer greater flexibility around both inputs and outputs. For example, read_sf_ext() is a wrapper for sf::read_sf() and offers a similar functionality:
nc <- read_sf_ext(system.file("shape/nc.shp", package = "sf"))However, read_sf_ext() also supports URLs for Google Sheets, FeatureLayers, data included with an installed package, and a variety of other sources. The function also supports an optional bounding box filter.
read_sf_ext(
"https://carto.nationalmap.gov/arcgis/rest/services/govunits/MapServer/29",
bbox = as_bbox(nc)
)
#> ── Downloading "National Park" from <https://carto.nationalmap.gov/arcgis/rest/s
#> Layer type: "Feature Layer"
#> Geometry type: "esriGeometryPolygon"
#> Service CRS: "EPSG:3857"
#> Output CRS: "EPSG:3857"
#>
#> Simple feature collection with 16 features and 19 fields
#> Geometry type: MULTIPOLYGON
#> Dimension: XY
#> Bounding box: xmin: -9402550 ymin: 4005420 xmax: -8400179 ymax: 4584128
#> Projected CRS: WGS 84 / Pseudo-Mercator
#> First 10 features:
#> boundarytype objectid permanent_identifier
#> 1 2 53213 {9fb6cd5d-441e-4c03-8426-3d4a803b07ef}
#> 2 2 53214 {865a3fe2-32c2-49d5-b68a-369f275b2c99}
#> 3 2 53242 {13e0e432-4a08-4be9-9a72-d50cb0928ed1}
#> 4 2 53309 {fdc693d6-c52b-489b-9acc-c4d29e3c6f46}
#> 5 2 53314 {4b27e664-6774-4cfb-9178-b7c76e3ca08b}
#> 6 2 53316 {d24ebeca-8f82-4b89-a46c-26fb0d76562e}
#> 7 2 53349 {9ca2f545-de4e-4c0d-a224-82035fd2a7bd}
#> 8 2 53350 {c351c3b1-85ee-46c5-880a-1e6965336690}
#> 9 2 53413 {a420e645-7448-4b7a-b3ff-213cef9cb543}
#> 10 2 53440 {3a21fb09-637b-4068-82ef-0ced22096e54}
#> source_featureid source_datasetid source_datadesc
#> 1 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> 2 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> 3 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> 4 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> 5 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> 6 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> 7 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> 8 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> 9 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> 10 NA {ee9f836e-d0af-4231-9111-e0026212f8c6} PAD-US 4.1
#> source_originator data_security distribution_policy loaddate
#> 1 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> 2 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> 3 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> 4 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> 5 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> 6 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> 7 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> 8 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> 9 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> 10 U.S. Geological Survey (USGS) 5 E4 1.76899e+12
#> name areasqkm ftype fcode
#> 1 Cumberland Gap National Historical Park 45.7183508 674 67400
#> 2 Cumberland Gap National Historical Park 21.8506152 674 67400
#> 3 Manhattan Project National Historical Park 0.2315734 674 67400
#> 4 Wright Brothers National Memorial 1.7347268 674 67400
#> 5 Guilford Courthouse National Military Park 0.9179618 674 67400
#> 6 Kings Mountain National Military Park 15.9040317 674 67400
#> 7 Great Smoky Mountains National Park 1120.8767616 674 67400
#> 8 Great Smoky Mountains National Park 975.5980353 674 67400
#> 9 Chattahoochee River National Recreation Area 48.0655438 674 67400
#> 10 Appalachian National Scenic Trail 37.3091713 674 67400
#> admintype ownerormanagingagency globalid
#> 1 1 13 {3F1C3525-CD9F-4D69-AEEC-BBEDFAE4D4D1}
#> 2 1 13 {CE87DABE-96B0-4258-B6BA-A94C3F92E79B}
#> 3 1 13 {94B8C8E9-AEE6-4C42-81F3-19997D374D46}
#> 4 1 13 {A5E2ACC1-E00A-41E9-88CD-E614D585408C}
#> 5 1 13 {52054BAF-CC96-4DAE-A467-F4F7B4A0E149}
#> 6 1 13 {8B638689-4E1B-4993-A793-8424784CEF38}
#> 7 1 13 {E8FF82C4-1169-482A-828B-AFAB64E5F156}
#> 8 1 13 {515DF85C-0CAA-4073-8D47-A43B08F2C7AB}
#> 9 1 13 {BFB5D087-A446-4AFC-89AE-4454783ADDBE}
#> 10 1 13 {0207A41A-C33C-466A-9A0F-E10338C4DF07}
#> shape_Length shape_Area geoms
#> 1 113682.810 71152073.2 MULTIPOLYGON (((-9288700 43...
#> 2 66922.761 33944322.3 MULTIPOLYGON (((-9313282 43...
#> 3 5857.886 353997.3 MULTIPOLYGON (((-9394763 42...
#> 4 9434.390 2657056.5 MULTIPOLYGON (((-8423490 43...
#> 5 10228.171 1410140.5 MULTIPOLYGON (((-8887125 43...
#> 6 22120.439 23836721.3 MULTIPOLYGON (((-9057033 41...
#> 7 433109.079 1697706425.8 MULTIPOLYGON (((-9337458 42...
#> 8 629078.132 1480396847.6 MULTIPOLYGON (((-9267885 42...
#> 9 326394.162 70122898.7 MULTIPOLYGON (((-9358817 40...
#> 10 281858.190 55412524.1 MULTIPOLYGON (((-9305056 41...st_union_ext() is nearly identical to sf::st_union() but optionally preserve a name column (collapsing the values of that column into a single string):
random_id <- sample(nrow(nc), size = 8)
nc_union <- st_union_ext(nc[random_id, ], name_col = "NAME")
plot(
nc_union
)
st_buffer_ext() wraps sf::st_buffer() but accepts bounding box objects as an input, allows you to set the units for the buffer distance using a character string (automatically converts the buffer distance units to match the units of the input object):
# Apply a 20 mile buffer to the unioned geometry
plot(
st_buffer_ext(nc_union, dist = 20, unit = "mi")
)
st_make_grid_ext() wraps sf::st_make_grid() but makes it easy to set the dimensions of the grid using rows, columns, and an overall aspect ratio:
# Make a 5 by 5 grid with a 8.5 by 11 aspect ratio filtered to x
plot(
st_make_grid_ext(
x = nc,
asp = 11 / 8.5,
ncol = 5,
nrow = 5,
filter = TRUE
)
)
Most functions that include a crs parameter can convert the coordinate reference system of the output (using transform_sf() or sf_bbox_transform()). The crs parameter also supports sf, sfc, or bbox inputs. Functions that include a class parameter can convert the class of an object using as_sf_class(). There are a set of functions for class conversion that typically wrap multiple sf functions with more limited input options.
nc_bbox <- as_bbox(nc, crs = 4326)
nc_bbox
#> xmin ymin xmax ymax
#> -84.32379 33.88209 -75.45656 36.58980
as_sfc(nc[1, ], crs = 3857)
#> Geometry set for 1 feature
#> Geometry type: MULTIPOLYGON
#> Dimension: XY
#> Bounding box: xmin: -9099356 ymin: 4332934 xmax: -9043562 ymax: 4382079
#> Projected CRS: WGS 84 / Pseudo-Mercator
#> MULTIPOLYGON (((-9069486 4332934, -9077066 4338...
as_sf(nc_bbox, crs = nc)
#> Simple feature collection with 1 feature and 0 fields
#> Geometry type: POLYGON
#> Dimension: XY
#> Bounding box: xmin: -84.32386 ymin: 33.88191 xmax: -75.45694 ymax: 36.58972
#> Geodetic CRS: NAD27
#> # A tibble: 1 × 1
#> geometry
#> * <POLYGON [°]>
#> 1 ((-84.32386 33.88199, -75.45698 33.88191, -75.45694 36.58965, -84.32385 36.58…Please note this flexibility make sfext easy to use (especially in an interactive context) but likely not appropriate for reproducible research. The package is being actively developed and the API may change.
Additional helper functions for sf objects
sfext also includes several helper functions that add new features by combining functions from the {sf} package, using affine geometry, or adding features from other packages.
For example, st_edge() combines sf::st_buffer() and sf::st_difference() to get the “edges” of any geometry.
st_nudge() allows you to shift the position of an sf object to a new location:
nc_nudge <- st_nudge(nc, to = nc[1, ])
plot(
st_union_ext(
nc,
nc_nudge
),
max.plot = 1
)
st_donut() allows you to create donuts around existing features:
plot(
st_donut(nc[c(1, 2, 3), ])
)
#> Warning: plotting the first 9 out of 15 attributes; use max.plot = 15 to plot
#> all
Lastly, the package has a whole group of helper functions for bbox objects. For example, sf_bbox_corners() creates a sf object with POINT geometry based on the corners of a bounding box:
plot(
sf_bbox_corners(
as_bbox(nc)
)
)
Related projects
sfext depends on two other development packages:
- {papersize}: A collection of convenience functions extending grid, ggplot2, and patchwork to help in sizing plots and files for printing to paper, postcards, playing cards, and other physical media.
- {filenamr}: A package to help create and modify file names and paths (that also supports reading and writing EXIF metadata).
It is also used extensively by two other development packages:
- {getdata}: A package to make the experience of getting location data easier and more consistent across a wide variety of sources.
- {maplayer}: A consistent set of functions for creating map layers for {ggplot2} using simple feature data.
There are many packages that build on sf for a variety of specialized use cases. A few worth noting include:
- {sfdep}: A sf and tidyverse friendly interface to the spdep package for spatial dependence.
- {sfnetworks}: Tidy Geospatial Networks in R.
- {sfhotspot}: A set of functions to identify and understand clusters of points (typically representing the locations of places or events).
- {sfx}: Extra ‘sf’ Simple Features manipulations.