These functions operate on one or more geography vectors and return a geography vector.

s2_boundary(x)

s2_centroid(x)

s2_closest_point(x, y)

s2_minimum_clearance_line_between(x, y)

s2_difference(x, y, options = s2_options())

s2_sym_difference(x, y, options = s2_options())

s2_intersection(x, y, options = s2_options())

s2_union(x, y = NULL, options = s2_options())

s2_snap_to_grid(x, grid_size)

s2_rebuild(x, options = s2_options())

s2_buffer_cells(
x,
distance,
max_cells = 1000,
min_level = -1,
)

s2_centroid_agg(x, na.rm = FALSE)

s2_union_agg(x, options = s2_options(), na.rm = FALSE)

## Arguments

x geography vectors. These inputs are passed to as_s2_geography(), so you can pass other objects (e.g., character vectors of well-known text) directly. geography vectors. These inputs are passed to as_s2_geography(), so you can pass other objects (e.g., character vectors of well-known text) directly. An s2_options() object describing the polygon/polyline model to use and the snap level. The grid size to which coordinates should be snapped; will be rounded to the nearest power of 10. The minimum distance between vertexes to use when simplifying a geography. Radius of the earth. Defaults to the average radius of the earth in meters as defined by s2_earth_radius_meters(). The distance to buffer, in units of radius. The maximum number of cells to approximate a buffer. The minimum cell level used to approximate a buffer (1 - 30). Setting this value too high will result in unnecessarily large geographies, but may help improve buffers along long, narrow regions. For aggregate calculations use na.rm = TRUE to drop missing values.

## Model

The geometry model indicates whether or not a geometry includes its boundaries. Boundaries of line geometries are its end points. OPEN geometries do not contain their boundary (model = "open"); CLOSED geometries (model = "closed") contain their boundary; SEMI-OPEN geometries (model = "semi-open") contain half of their boundaries, such that when two polygons do not overlap or two lines do not cross, no point exist that belong to more than one of the geometries. (This latter form, half-closed, is not present in the OpenGIS "simple feature access" (SFA) standard nor DE9-IM on which that is based). The default values for s2_contains() (open) and covers/covered_by (closed) correspond to the SFA standard specification of these operators.

BigQuery's geography function reference:

## Examples

# returns the boundary:
# empty for point, endpoints of a linestring,
# perimeter of a polygon
s2_boundary("POINT (-64 45)")
#> <s2_geography[1]>
#> [1] <POINT EMPTY>s2_boundary("LINESTRING (0 0, 10 0)")
#> <s2_geography[1]>
#> [1] <MULTIPOINT ((0 0), (10 0))>s2_boundary("POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0))")
#> <s2_geography[1]>
#> [1] <LINESTRING (0 0, 10 0, 10 10, 0 10, 0 0...>
# returns the area-weighted centroid, element-wise
s2_centroid("POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0))")
#> <s2_geography[1]>
#> [1] <POINT (5 5.00595863)>s2_centroid("LINESTRING (0 0, 10 0)")
#> <s2_geography[1]>
#> [1] <POINT (5 0)>
# returns the unweighted centroid of the entire input
s2_centroid_agg(c("POINT (0 0)", "POINT (10 0)"))
#> <s2_geography[1]>
#> [1] <POINT (5 0)>
# returns the closest point on x to y
s2_closest_point(
"POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0))",
"POINT (0 90)" # north pole!
)
#> <s2_geography[1]>
#> [1] <POINT (5 10.037423)>
# returns the shortest possible line between x and y
s2_minimum_clearance_line_between(
"POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0))",
"POINT (0 90)" # north pole!
)
#> <s2_geography[1]>
#> [1] <LINESTRING (5 10.037423, 0 90)>
# binary operations: difference, symmetric difference, intersection and union
s2_difference(
"POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0))",
"POLYGON ((5 5, 15 5, 15 15, 5 15, 5 5))",
# 32 bit platforms may need to set snap rounding
s2_options(snap = s2_snap_level(30))
)
#> <s2_geography[1]>
#> [1] <POLYGON ((5.00000004 10.0374231, 3.55739019e-08 10, 3.55739019e-08 3.55739019e-08, 10 3.50334544e-08, 10 5.01900178...>
s2_sym_difference(
"POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0))",
"POLYGON ((5 5, 15 5, 15 15, 5 15, 5 5))",
# 32 bit platforms may need to set snap rounding
s2_options(snap = s2_snap_level(30))
)
#> <s2_geography[1]>
#> [1] <MULTIPOLYGON (((5.00000004 10.0374231, 3.55739019e-08 10, 3.55739019e-08 3.55739019e-08, 10 3.50334544e-08, 10 5.01900178...>
s2_intersection(
"POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0))",
"POLYGON ((5 5, 15 5, 15 15, 5 15, 5 5))",
# 32 bit platforms may need to set snap rounding
s2_options(snap = s2_snap_level(30))
)
#> <s2_geography[1]>
#> [1] <POLYGON ((5.00000004 4.99999997, 10 5.01900178, 10 9.99999999, 5.00000004 10.0374231, 5.00000004 4.99999997...>
s2_union(
"POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0))",
"POLYGON ((5 5, 15 5, 15 15, 5 15, 5 5))",
# 32 bit platforms may need to set snap rounding
s2_options(snap = s2_snap_level(30))
)
#> <s2_geography[1]>
#> [1] <POLYGON ((5.00000004 10.0374231, 3.55739019e-08 10, 3.55739019e-08 3.55739019e-08, 10 3.50334544e-08, 10 5.01900178...>
# use s2_union_agg() to aggregate geographies in a vector
s2_union_agg(
c(
"POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0))",
"POLYGON ((5 5, 15 5, 15 15, 5 15, 5 5))"
),
# 32 bit platforms may need to set snap rounding
s2_options(snap = s2_snap_level(30))
)
#> <s2_geography[1]>
#> [1] <POLYGON ((5.00000004 10.0374231, 3.55739019e-08 10, 3.55739019e-08 3.55739019e-08, 10 3.50334544e-08, 10 5.01900178...>
# snap to grid rounds coordinates to a specified grid size
s2_snap_to_grid("POINT (0.333333333333 0.666666666666)", 1e-2)
#> <s2_geography[1]>
#> [1] <POINT (0.33 0.67)>