sf (version 1.1-1)

tidyverse: Tidyverse methods for sf objects

Description

Tidyverse methods for sf objects. Geometries are sticky, use as.data.frame to let dplyr's own methods drop them. Use these methods after loading the tidyverse package with the generic (or after loading package tidyverse).

Usage

filter.sf(.data, ..., .dots)
arrange.sf(.data, ..., .dots)
group_by.sf(.data, ..., add = FALSE)
ungroup.sf(x, ...)
rowwise.sf(x, ...)
mutate.sf(.data, ..., .dots)
transmute.sf(.data, ..., .dots)
select.sf(.data, ...)
rename.sf(.data, ...)
rename_with.sf(.data, .fn, .cols, ...)
slice.sf(.data, ..., .dots)
summarise.sf(.data, ..., .dots, do_union = TRUE, is_coverage = FALSE)
count.sf(x, ..., wt = NULL, sort = FALSE, name = "n", .drop_geometry = FALSE)
distinct.sf(.data, ..., .keep_all = FALSE, exact = FALSE, par = 0)
gather.sf(
  data,
  key,
  value,
  ...,
  na.rm = FALSE,
  convert = FALSE,
  factor_key = FALSE
)
pivot_longer.sf(
  data,
  cols,
  names_to = "name",
  names_prefix = NULL,
  names_sep = NULL,
  names_pattern = NULL,
  names_ptypes = NULL,
  names_transform = NULL,
  names_repair = "check_unique",
  values_to = "value",
  values_drop_na = FALSE,
  values_ptypes = NULL,
  values_transform = NULL,
  ...
)
pivot_wider.sf(
  data,
  ...,
  id_cols = NULL,
  id_expand = FALSE,
  names_from = name,
  names_prefix = "",
  names_sep = "_",
  names_glue = NULL,
  names_sort = FALSE,
  names_vary = "fastest",
  names_expand = FALSE,
  names_repair = "check_unique",
  values_from = value,
  values_fill = NULL,
  values_fn = NULL,
  unused_fn = NULL
)
spread.sf(
  data,
  key,
  value,
  fill = NA,
  convert = FALSE,
  drop = TRUE,
  sep = NULL
)
sample_n.sf(tbl, size, replace = FALSE, weight = NULL, .env = parent.frame())
sample_frac.sf(
  tbl,
  size = 1,
  replace = FALSE,
  weight = NULL,
  .env = parent.frame()
)
group_split.sf(.tbl, ..., .keep = TRUE)
nest.sf(.data, ...)
separate.sf(
  data,
  col,
  into,
  sep = "[^[:alnum:]]+",
  remove = TRUE,
  convert = FALSE,
  extra = "warn",
  fill = "warn",
  ...
)
separate_rows.sf(data, ..., sep = "[^[:alnum:]]+", convert = FALSE)
unite.sf(data, col, ..., sep = "_", remove = TRUE)
unnest.sf(data, ..., .preserve = NULL)
drop_na.sf(x, ...)
inner_join.sf(x, y, by = NULL, copy = FALSE, suffix = c(".x", ".y"), ...)
left_join.sf(x, y, by = NULL, copy = FALSE, suffix = c(".x", ".y"), ...)
right_join.sf(x, y, by = NULL, copy = FALSE, suffix = c(".x", ".y"), ...)
full_join.sf(x, y, by = NULL, copy = FALSE, suffix = c(".x", ".y"), ...)
semi_join.sf(x, y, by = NULL, copy = FALSE, suffix = c(".x", ".y"), ...)
anti_join.sf(x, y, by = NULL, copy = FALSE, suffix = c(".x", ".y"), ...)

Value

an object of class sf

Arguments

.data

data object of class sf

...

other arguments

.dots

see corresponding function in package dplyr

add

see corresponding function in dplyr

x, y

A pair of data frames, data frame extensions (e.g. a tibble), or lazy data frames (e.g. from dbplyr or dtplyr). See Methods, below, for more details.

.fn, .cols

see original docs

do_union

logical; in case summary does not create a geometry column, should geometries be created by unioning using st_union, or simply by combining using st_combine? Using st_union resolves internal boundaries, but in case of unioning points, this will likely change the order of the points; see Details.

is_coverage

logical; if do_union is TRUE, use an optimized algorithm for features that form a polygonal coverage (have no overlaps)

wt

see original function docs

sort

see original function docs

name

see original function docs

.drop_geometry

logical; if TRUE, remove geometry column before computing counts

.keep_all

see corresponding function in dplyr

exact

logical; if TRUE use st_equals_exact for geometry comparisons

par

numeric; passed on to st_equals_exact

data

see original function docs

key

see original function docs

value

see original function docs

na.rm

see original function docs

convert

see separate_rows

factor_key

see original function docs

cols

see original function docs

names_to, names_pattern, names_ptypes, names_transform

see tidyr::pivot_longer()

names_prefix, names_sep, names_repair

see original function docs.

values_to, values_drop_na, values_ptypes, values_transform

See tidyr::pivot_longer()

id_cols, id_expand, names_from, names_sort, names_glue, names_vary, names_expand

see tidyr::pivot_wider()

values_from, values_fill, values_fn, unused_fn

see tidyr::pivot_wider()

fill

see original function docs

drop

see original function docs

sep

see separate_rows

tbl

see original function docs

size

see original function docs

replace

see original function docs

weight

see original function docs

.env

see original function docs

.tbl

see original function docs

.keep

see original function docs

col

see separate

into

see separate

remove

see separate

extra

see separate

.preserve

see unnest

by

A join specification created with join_by(), or a character vector of variables to join by.

If NULL, the default, *_join() will perform a natural join, using all variables in common across x and y. A message lists the variables so that you can check they're correct; suppress the message by supplying by explicitly.

To join on different variables between x and y, use a join_by() specification. For example, join_by(a == b) will match x$a to y$b.

To join by multiple variables, use a join_by() specification with multiple expressions. For example, join_by(a == b, c == d) will match x$a to y$b and x$c to y$d. If the column names are the same between x and y, you can shorten this by listing only the variable names, like join_by(a, c).

join_by() can also be used to perform inequality, rolling, and overlap joins. See the documentation at ?join_by for details on these types of joins.

For simple equality joins, you can alternatively specify a character vector of variable names to join by. For example, by = c("a", "b") joins x$a to y$a and x$b to y$b. If variable names differ between x and y, use a named character vector like by = c("x_a" = "y_a", "x_b" = "y_b").

To perform a cross-join, generating all combinations of x and y, see cross_join().

copy

If x and y are not from the same data source, and copy is TRUE, then y will be copied into the same src as x. This allows you to join tables across srcs, but it is a potentially expensive operation so you must opt into it.

suffix

If there are non-joined duplicate variables in x and y, these suffixes will be added to the output to disambiguate them. Should be a character vector of length 2.

Details

select keeps the geometry regardless whether it is selected or not; to deselect it, first pipe through as.data.frame to let dplyr's own select drop it.

In case one or more of the arguments (expressions) in the summarise call creates a geometry list-column, the first of these will be the (active) geometry of the returned object. If this is not the case, a geometry column is created, depending on the value of do_union.

In case do_union is FALSE, summarise will simply combine geometries using c.sfg. When polygons sharing a boundary are combined, this leads to geometries that are invalid; see for instance https://github.com/r-spatial/sf/issues/681.

The functions count and tally drop all geometries. For counting geometries use summarise(.data, n = n(), .by = "geometry").

distinct gives distinct records for which all attributes and geometries are distinct; st_equals is used to find out which geometries are distinct.

nest assumes that a simple feature geometry list-column was among the columns that were nested.

Examples

if (require(dplyr, quietly = TRUE)) {
 nc = read_sf(system.file("shape/nc.shp", package="sf"))
 nc |> filter(AREA > .1) |> plot()
 # plot 10 smallest counties in grey:
 st_geometry(nc) |> plot()
 nc |> select(AREA) |> arrange(AREA) |> slice(1:10) |> plot(add = TRUE, col = 'grey')
 title("the ten counties with smallest area")
 nc2 <- nc |> mutate(area10 = AREA/10)
 nc |> slice(1:2)
}
# plot 10 smallest counties in grey:
if (require(dplyr, quietly = TRUE)) {
 st_geometry(nc) |> plot()
 nc |> select(AREA) |> arrange(AREA) |> slice(1:10) |> plot(add = TRUE, col = 'grey')
 title("the ten counties with smallest area")
}
if (require(dplyr, quietly = TRUE)) {
 nc$area_cl = cut(nc$AREA, c(0, .1, .12, .15, .25))
 nc |> group_by(area_cl) |> class()
}
if (require(dplyr, quietly = TRUE)) {
 nc2 <- nc |> mutate(area10 = AREA/10)
}
if (require(dplyr, quietly = TRUE)) {
 nc |> transmute(AREA = AREA/10) |> class()
}
if (require(dplyr, quietly = TRUE)) {
 nc |> select(SID74, SID79) |> names()
 nc |> select(SID74, SID79) |> class()
}
if (require(dplyr, quietly = TRUE)) {
 nc2 <- nc |> rename(area = AREA)
}
if (require(dplyr, quietly = TRUE)) {
 nc |> slice(1:2)
}
if (require(dplyr, quietly = TRUE)) {
 nc$area_cl = cut(nc$AREA, c(0, .1, .12, .15, .25))
 nc.g <- nc |> group_by(area_cl)
 nc.g |> summarise(mean(AREA))
 nc.g |> summarise(mean(AREA)) |> plot(col = grey(3:6 / 7))
 nc |> as.data.frame() |> summarise(mean(AREA))
 # counting geometries (after duplicating each row):
 nc.dupl <- nc[rep(seq_along(nc), each = 2), ]
 nc.dupl |> summarise(n = n(), .by = "geometry")
}
if (require(dplyr, quietly = TRUE)) {
  nc$area_cl <- cut(nc$AREA, c(0, .1, .12, .15, .25))
  nc |> count(area_cl, .drop_geometry = TRUE)
}
if (require(dplyr, quietly = TRUE)) {
 nc[c(1:100, 1:10), ] |> distinct() |> nrow()
}
if (require(tidyr, quietly = TRUE) && require(dplyr, quietly = TRUE) && "geometry" %in% names(nc)) {
 nc |> select(SID74, SID79) |> gather("VAR", "SID", -geometry) |> summary()
}
if (require(tidyr, quietly = TRUE) && require(dplyr, quietly = TRUE) && "geometry" %in% names(nc)) {
 nc$row = 1:100 # needed for spread to work
 nc |> select(SID74, SID79, geometry, row) |>
	gather("VAR", "SID", -geometry, -row) |>
	spread(VAR, SID) |> head()
}
if (require(tidyr, quietly = TRUE) && require(dplyr, quietly = TRUE)) {
 storms.sf = st_as_sf(storms, coords = c("long", "lat"), crs = 4326)
 x <- storms.sf |> group_by(name, year) |> nest()
 trs = lapply(x$data, function(tr) st_cast(st_combine(tr), "LINESTRING")[[1]]) |>
    st_sfc(crs = 4326)
 trs.sf = st_sf(x[,1:2], trs)
 plot(trs.sf["year"], axes = TRUE)
}