ms_erase: Remove features or portions of features that fall inside a...

View source: R/clip_erase.R

ms_eraseR Documentation

Remove features or portions of features that fall inside a specified area

Description

Removes portions of the target layer that fall inside the erasing layer or bounding box.

Usage

ms_erase(target, erase = NULL, bbox = NULL, remove_slivers = FALSE, ...)

Arguments

target

the target layer from which to remove portions. One of:

  • geo_json or character points, lines, or polygons;

  • SpatialPolygons, SpatialLines, SpatialPoints

erase

the erase layer (polygon). One of:

  • geo_json or character polygons;

  • SpatialPolygons*

bbox

supply a bounding box instead of an erasing layer to remove from the target layer. Supply as a numeric vector: c(minX, minY, maxX, maxY).

remove_slivers

Remove tiny sliver polygons created by erasing. (Default FALSE)

...

Arguments passed on to apply_mapshaper_commands

force_FC

should the output be forced to be a FeatureCollection (or sf object or Spatial*DataFrame) even if there are no attributes? Default TRUE. If FALSE and there are no attributes associated with the geometries, a GeometryCollection (or Spatial object with no dataframe, or sfc) will be output.

sys

Should the system mapshaper be used instead of the bundled mapshaper? Gives better performance on large files. Requires the mapshaper node package to be installed and on the PATH.

sys_mem

How much memory (in GB) should be allocated if using the system mapshaper (sys = TRUE)? Default 8. Ignored if sys = FALSE. This can also be set globally with the option "mapshaper.sys_mem"

quiet

If sys = TRUE, should the mapshaper messages be silenced? Default FALSE. This can also be set globally with the option "mapshaper.sys_quiet"

Value

erased target in the same format as the input target

Examples

if (rmapshaper:::check_v8_major_version() >= 6L) {
  library(geojsonsf, quietly = TRUE)
  library(sf)

  points <- structure("{\"type\":\"FeatureCollection\",
    \"features\":[{\"type\":\"Feature\",\"properties\":{},
    \"geometry\":{\"type\":\"Point\",\"coordinates\":
    [52.8658,-44.7219]}},{\"type\":\"Feature\",\"properties\":{},
    \"geometry\":{\"type\":\"Point\",\"coordinates\":
    [53.7702,-40.4873]}},{\"type\":\"Feature\",\"properties\":{},
    \"geometry\":{\"type\":\"Point\",\"coordinates\":[55.3204,-37.5579]}},
    {\"type\":\"Feature\",\"properties\":{},\"geometry\":
    {\"type\":\"Point\",\"coordinates\":[56.2757,-37.917]}},
    {\"type\":\"Feature\",\"properties\":{},\"geometry\":
    {\"type\":\"Point\",\"coordinates\":[56.184,-40.6443]}},
    {\"type\":\"Feature\",\"properties\":{},\"geometry\":
    {\"type\":\"Point\",\"coordinates\":[61.0835,-40.7529]}},
    {\"type\":\"Feature\",\"properties\":{},\"geometry\":
    {\"type\":\"Point\",\"coordinates\":[58.0202,-43.634]}}]}",
    class = c("geojson", "json"))
  points <- geojson_sf(points)
  plot(points)

  erase_poly <- structure('{
  "type": "Feature",
  "properties": {},
  "geometry": {
  "type": "Polygon",
  "coordinates": [
  [
  [51, -40],
  [55, -40],
  [55, -45],
  [51, -45],
  [51, -40]
  ]
  ]
  }
  }', class = c("geojson", "json"))
  erase_poly <- geojson_sf(erase_poly)

  out <- ms_erase(points, erase_poly)
  plot(out, add = TRUE)
}


rmapshaper documentation built on April 11, 2023, 5:55 p.m.