tangler: Re-apply anonymisation using a stored detangler object

In tangles: Anonymisation of Spatial Point Patterns and Grids

Re-apply anonymisation using a stored detangler object

Description

For a given detangler object (from tangles), spatial coordinates or raster data can be re-anonymised using the same transformation sequence. This allows consistent tangling of related datasets using a fixed spatial masking.

Usage

tangler(
  data = NULL,
  tanglerInfo = NULL,
  raster_object = FALSE,
  stub = NULL,
  saveTangles = FALSE,
  exportShapefile = FALSE,
  path = NULL
)

Arguments

data	A 2-column matrix, data.frame of coordinates, an sf POINT object, or a terra::SpatRaster object.
tanglerInfo	A detangler object returned by tangles, which encodes the anonymisation steps and hash key.
raster_object	Logical; set to TRUE if input is a raster object. This ensures the output is reconstructed as a terra::SpatRaster.
stub	Optional character string for file naming. Combined with the hash key for saved outputs.
saveTangles	Logical; if TRUE, the tangler output is saved as an .rds object, and raster layers as .tif.
exportShapefile	Logical; if TRUE and input is point-based, a shapefile of the tangled data is also exported (no CRS assigned).
path	Directory to write output files if saving is enabled. Defaults tempdir().

Value

Returns either:

• A data.frame of spatial coordinates (for point-based data), or

• A terra::SpatRaster (for raster input)

If saveTangles = TRUE, corresponding outputs are written to .rds, and optionally .tif (for raster) or shapefile (for point).

Note

This function re-applies a saved transformation sequence. It does not generate new spatial shifts.

When working with raster data, only 90°, 180°, or 270° rotations are supported. If the stored detangler contains unsupported rotation angles, the function will stop to avoid corrupting the raster structure.

Shapefiles are written without CRS metadata to preserve anonymity. Coordinates retain spatial structure but not geolocation accuracy.

Author(s)

Brendan Malone

References

• CM O’Keefe, S Otorepec, M Elliot, E Mackey, and K O’Hara (2017) The De-Identification Decision Making Framework. CSIRO Reports EP173122 and EP175702. \Sexpr[results=rd]{tools:::Rd_expr_doi("10.4225/08/59c169433efd4")}

Examples

## Example 1: Tangling a point data.frame
library(digest)
set.seed(123)
pts <- data.frame(X = runif(100), Y = runif(100))

# Anonymise original points
t1 <- tangles(data = pts, depth = 4)

# Tangling a second dataset using the same detangler
tangled_again <- tangler(
  data = pts,
  tanglerInfo = t1[[2]],
  stub = "pt_demo",
  saveTangles = TRUE,
  exportShapefile = TRUE
)

## Example 2: Tangling an sf POINT object
library(sf)
sf_pts <- st_as_sf(pts, coords = c("X", "Y"))

tangled_sf <- tangler(
  data = sf_pts,
  tanglerInfo = t1[[2]],
  stub = "sf_demo",
  saveTangles = TRUE,
  exportShapefile = TRUE
)

## Example 3: Tangling a raster using stored detangler
library(terra)
ext_path <- system.file("extdata", package = "tangles")
rast.files <- list.files(path = ext_path, full.names = TRUE)
rasters <- terra::rast(rast.files)

# Must use a detangler with 90°/180°/270° rotations for raster compatibility
t2 <- tangles(data = rasters, depth = 3, rasterdata = TRUE, raster_object = TRUE)

tangled_rast <- tangler(
  data = rasters,
  tanglerInfo = t2[[2]],
  raster_object = TRUE,
  stub = "r_demo",
  saveTangles = TRUE
)

tangles documentation built on June 8, 2025, 11:38 a.m.