Skip to contents

Spatial Vector Validation and Cleaning

Source: R/validate_process_spatial.R
validate_process_spatial.Rd

Validates spatial vector data and creates admin-level aggregations with geometry hashes. Follows the same validation approach as validate_process_coordinates() with modular checks and standardized outputs.

Usage

validate_process_spatial(
  shp,
  name = "shapefile",
  adm0_col = NULL,
  adm1_col = NULL,
  adm2_col = NULL,
  adm3_col = NULL,
  adm4_col = NULL,
  fix_issues = TRUE,
  quiet = FALSE,
  geometry_crs = 4326,
  drop_z = TRUE
)

Arguments

shp

An sf object to validate

name

Character string identifying the dataset (for reporting)

adm0_col

Character string, column name for country level

adm1_col

Character string, column name for region/province level

adm2_col

Character string, column name for district level

adm3_col

Character string, column name for chiefdom/commune level

adm4_col

Character string, column name for sub-chiefdom/village level

fix_issues

Logical, whether to attempt automatic fixes

quiet

Logical, whether to suppress progress messages

geometry_crs

Target CRS for output geometry (default EPSG:4326). After CRS validation/fixes, geometries are transformed to this CRS for subsequent processing and output.

drop_z

Logical, whether to drop Z/M coordinates before validation. Defaults to TRUE to avoid s2 warnings for 2D processing.

Value

A list with validation results. Key elements:

  • issues: Character vector of issues detected.

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

  • checks$duplicate_rows: Duplicates by admin names + geometry.

  • checks$all_shared_locations: All records sharing geometry.

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

  • checks$invalid_geometries: Rows with invalid geometries before fixing.

  • checks$empty_geometries: Rows with empty geometries.

  • checks$self_intersecting: Rows with self-intersecting geometries.

  • column_dictionary: Data frame with column_name and description.

  • final_spat_vec: List of admin-level aggregations (adm0, adm1, adm2, adm3, adm4).

  • geometry_types: Tibble summarizing geometry types and counts.

  • spatial_extent: Named vector with xmin, ymin, xmax, ymax.

Examples

if (FALSE) { # \dontrun{
# Basic validation with Sierra Leone data (adm3)
result <- validate_process_spatial(
  shp = shp_raw,
  name = "Sierra Leone boundaries",
  adm0_col = "country",
  adm1_col = "region",
  adm2_col = "district",
  adm3_col = "chiefdom"
)

# Validation with adm4
result <- validate_process_spatial(
  shp = shp_raw,
  name = "Sierra Leone boundaries",
  adm0_col = "country",
  adm1_col = "region",
  adm2_col = "district",
  adm3_col = "chiefdom",
  adm4_col = "village"
)

# Access cleaned admin levels
shp_adm0 <- result$final_spat_vec$adm0
shp_adm1 <- result$final_spat_vec$adm1
shp_adm2 <- result$final_spat_vec$adm2
shp_adm3 <- result$final_spat_vec$adm3
shp_adm4 <- result$final_spat_vec$adm4

# Check validation status
if (length(result$issues) == 0) {
  cli::cli_alert_success("Ready for SNT analysis!")
}

# Access invalid rows (if any)
invalid_rows <- result$invalid_rows

# Access duplicate rows (if any)
duplicate_rows <- result$checks$duplicate_rows
shared_locations <- result$checks$all_shared_locations

# Check column structure as data frame
col_dict <- result$column_dictionary
# Shows data frame with shapefile, old_name, new_name, description
print(col_dict)
} # }