Tips and tricks for using the package

This vignette presents a collection of useful tips and tricks we’ve gathered over the years for effectively using this package to read, download, and filter OpenStreetMap (OSM) extracts. First of all, let’s load the relevant packages:

library(osmextract)

How can I get OSM objects by node/way id number?

The example below demonstrates how to select a set of ways from an OSM extract, assuming you already know their OSM IDs:

osm_id <- c("4419868", "6966733", "7989989", "15333726", "31705837")

out <- oe_get(
  place = "ITS Leeds",
  query = paste0(
    "SELECT * FROM lines WHERE osm_id IN (", paste0(osm_id, collapse = ","), ")"
  ), 
  quiet = TRUE
)
print(out, n = 0L)
#> Simple feature collection with 5 features and 10 fields
#> Geometry type: LINESTRING
#> Dimension:     XY
#> Bounding box:  xmin: -1.5609 ymin: 53.8063 xmax: -1.549451 ymax: 53.81044
#> Geodetic CRS:  WGS 84

How can I convert the segments downloaded from OSM into a street network?

Starting from version 0.7, the package includes a set of functions to automatically convert the output of oe_get_network() (and similar functions) into sfnetwork or dodgr objects.

For example, the following command returns a sfnetwork object representing the walking network extracted from the toy ITS data included in the package:

sfn_walking = oe_get_sfnetwork(
  place = "ITS Leeds",
  mode = "walking", 
  quiet = TRUE
)
#> Warning: to_spatial_subdivision assumes attributes are constant over geometries

More precisely, the function runs the following operations:

  1. It extracts the walkable streets from the toy ITS Leeds data included in the package;
  2. It applies a series of preprocessing steps to standardise the values included in the oneway column and simplies the highway tag removing the "_link" suffix. See the clean_output argument in oe_get_network for more details.
  3. It converts the data into sfnetwork class and applies two spatial morphers, namely to_spatial_subdivision and to_spatial_smooth to fix possible inconsistencies in the geometries and simplify the geometry structure. See also ?net_2_sfnet_undirected;
  4. If requested, converts the output into a directed network after duplicating bidirectional edges with reversed geometries.

The following command runs similar operations (except for the spatial morphers) and returns a dodgr_streenet object:

dodgr_walking = oe_get_dodgrnetwork(
  place = "ITS Leeds",
  mode = "walking", 
  quiet = TRUE
)
#> Warning in oe_get_dodgrnetwork(place = "ITS Leeds", mode = "walking", quiet = TRUE): The 'oneway'
#> column is missing. All edges will be assumed to be bidirectional!

See the help pages of the corresponding functions for more details.