Skip to contents

Coordinate Validation and Cleaning

Source: R/validate_process_coordinates.R
validate_process_coordinates.Rd

Validates coordinate data (points) and returns a cleaned sf object with standardized columns. Accepts either an sf object with POINT geometry or a data.frame with longitude and latitude columns. Checks include missing coordinates, DMS detection with optional conversion (via the suggested 'parzer' package), range, precision, flip detection, and country containment.

Usage

validate_process_coordinates(
  data,
  name = "points",
  lon_col = NULL,
  lat_col = NULL,
  adm0_sf = NULL,
  geometry_crs = 4326,
  min_decimals = 4,
  id_col = NULL,
  fix_issues = TRUE,
  quiet = FALSE
)

Arguments

data

An sf object with POINT geometry or a data.frame containing coordinate columns.

name

Character string identifying the dataset (for reporting).

lon_col

Character string, column name for longitude (if data is not sf).

lat_col

Character string, column name for latitude (if data is not sf).

adm0_sf

Optional sf polygon/multipolygon of the country to validate points against (EPSG:4326 preferred). When provided, the function tests that all points fall within the given shape. No admin attributes are joined to the points.

geometry_crs

Target CRS for output geometry and for operations with adm0_sf (default EPSG:4326). Lat/Lon attributes remain in EPSG:4326.

min_decimals

Minimum decimal places required for lon/lat to be considered precise (default 4).

id_col

Optional character name of a unique identifier column (e.g., facility ID). When provided, duplicate reporting includes IDs for coordinate-duplicate rows to aid de-dup decisions.

fix_issues

Logical, whether to attempt automatic fixes (drop invalids, set CRS, standardize columns, remove duplicates, attempt flip fixes when adm0_sf is available).

quiet

Logical, whether to suppress progress messages.

Value

A list with validation results. Key elements:

  • issues: Character vector of issues detected.

  • final_points_df: Cleaned sf object of points.

  • invalid_rows: sf/data.frame of rows failing basic coordinate checks.

  • checks$duplicate_rows: Duplicates by ID + coordinates (standardized sf).

  • checks$all_shared_locations: All records sharing a location.

  • checks$all_shared_locations_ungrouped: Only the duplicate rows at shared locations.

  • checks$coordinates_adm0: Points outside ADM0 with original lon/lat restored from geometry.

  • column_dictionary: Data frame mapping old to standardized column names.

Examples

if (FALSE) { # \dontrun{
# Example data with common issues: missing, DMS, flipped, imprecise
df <- data.frame(
  hf = paste0("HF_", 1:6),
  lon = c(-13.26077, "13\u00B015'38\"W", 8.5, -12.09057, NA, 181),
  lat = c(8.464283, "8\u00B027'51\"N", -13.26, 8.770261, 8.1, 0.1)
)

# Simple rectangle as country polygon (e.g., Sierra Leone-ish bbox)
adm0_sf <- sf::st_as_sfc(
  sf::st_bbox(c(xmin = -14, ymin = 6.9, xmax = -10, ymax = 10.1),
              crs = sf::st_crs(4326))
) |>
  sf::st_sf()

# Validate and auto-fix issues (DMS parsing requires 'parzer')
res <- validate_process_coordinates(
  data = df,
  name = "hf_points",
  lon_col = "lon",
  lat_col = "lat",
  adm0_sf = adm0_sf,
  min_decimals = 4,
  fix_issues = TRUE,
  quiet = TRUE
)

# Cleaned points
res$final_points_df
} # }