Skip to content

r-spatial/sfdbi

Repository files navigation

sfdbi

Codecov test coverage Lifecycle: experimental

Overview

sfdbi helps you manipulate spatial data stored in a database as if it was in-memory with standard sf and dplyr functions. With sfdbi, you can read and write sf objects to databases and translate sf operations to SQL. We support postgis, but feel free to open an issue for other backends and show you interest.

Installation

You can install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("r-spatial/sfdbi")

sfdbi is not available on CRAN yet.

Example

This is how you write and read spatial data to a database. Note that sfdbi works best with dplyr.

library(sfdbi)
library(sf)
#> Linking to GEOS 3.8.0, GDAL 3.0.2, PROJ 6.2.1
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

con <- DBI::dbConnect(
  RPostgres::Postgres(), 
  host = "localhost", 
  port = 25432
) %>%
  postgis()

# Create a spatial table
pyramids <- tribble(  
  ~what, ~geom,
  "Giza",        "POINT(31.1342 29.9792)",
  "Khafre",      "POINT(31.130833 29.976111)", 
  "Menkaure",    "POINT(31.128333 29.9725)",
  "Khentkaus I", "POINT(31.135608 29.973406)", 
  "Sphynx",      "POINT(31.137778 29.975278)",
  ) %>% 
  mutate(
    geom = st_as_sfc(geom, crs = 4326)
  )

# Copy spatial data to database
x <- copy_to(con, pyramids)
#> Note: method with signature 'DBIObject#sf' chosen for function 'dbDataType',
#>  target signature 'PqConnection#sf'.
#>  "PqConnection#ANY" would also be valid

# Queue operations to the database
y <- x %>% 
  mutate(
    geom = geom %>% 
      st_transform(23035L) %>% 
      st_buffer(10)
  )

# View the query to be executed
y %>% show_query()
#> <SQL>
#> SELECT "what", st_buffer(st_transform("geom", 23035), 10.0) AS "geom"
#> FROM "pyramids"

# Execute operation in the database and load it in R memory
collect(y)
#> # A tibble: 5 x 2
#>   what                                                                      geom
#>   <chr>                                                            <POLYGON [m]>
#> 1 Giza       ((899053.3 3323853, 899053.1 3323851, 899052.6 3323849, 899051.7 3…
#> 2 Khafre     ((898740.5 3323498, 898740.3 3323496, 898739.7 3323494, 898738.8 3…
#> 3 Menkaure   ((898513.4 3323089, 898513.2 3323087, 898512.7 3323085, 898511.7 3…
#> 4 Khentkaus… ((899212.6 3323215, 899212.4 3323213, 899211.8 3323211, 899210.9 3…
#> 5 Sphynx     ((899414.7 3323430, 899414.5 3323428, 899413.9 3323426, 899413 332…

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages