adding a prune data package method

NAMESPACE

@@ -28,6 +28,7 @@ export(obfuscate_gps)
 export(obfuscate_lat)
 export(obfuscate_lon)
 export(othertext_lookup)
+export(prune_datapackage)
 export(read_excel_all_sheets)
 export(read_googlesheets)
 export(remove_deletions)

R/expand_frictionless_metadata.R → R/modify_frictionless_metadata.R

@@ -8,6 +8,8 @@
 #' @param resource_name Character. Item within the datapackage to be updated
 #' @param resource_path Character. Path to csv file
 #' @param data_package_path Character. Path to datapackage.json file
+#' @param prune_datapackage Logical. Should properties not in the structural metadata
+#' be removed?
 #'
 #' @return Updates the datapackage, returns nothing
 #' @export
@@ -46,7 +48,8 @@
 expand_frictionless_metadata <- function(structural_metadata,
                                          resource_name,
                                          resource_path,
-                                         data_package_path ){
+                                         data_package_path,
+                                         prune_datapackage = TRUE){
 
   data_package <- frictionless::read_package(data_package_path)
 
@@ -82,6 +85,10 @@ expand_frictionless_metadata <- function(structural_metadata,
     my_data_schema$fields[[idx]] <- x
   }
 
+  if(prune_datapackage){
+    my_data_schema <- prune_datapackage(my_data_schema,structural_metadata)
+  }
+
   # update the datapackage.json
   data_package <- data_package|>
     frictionless::remove_resource(resource_name) |>
@@ -95,3 +102,37 @@ expand_frictionless_metadata <- function(structural_metadata,
 
   invisible()
 }
+
+
+#' Prune data pacakge
+#'
+#' method to remove properties from the metadata for a dataset in a datapackage
+#'
+#' @param my_data_schema list. schema object from frictionless
+#' @param structural_metadata dataframe. structural metadata for a dataset
+#'
+#' @return pruned data_schema -
+#' @export
+#'
+prune_datapackage <- function(my_data_schema, structural_metadata){
+
+  # get property names
+  property_names <- names(structural_metadata)
+
+  # add minimal property values
+  property_names_complete <- append(c("name","type"),property_names) |>
+    unique()
+
+  # create storage object
+  my_data_schema_pruned <- my_data_schema
+
+  # map over fields and remove metadata items not in property names complete
+  my_data_schema_pruned$fields <- purrr::map(my_data_schema$fields, function(schema_item){
+
+    properties_to_drop <- names(schema_item) %in% property_names_complete
+    out <- schema_item[properties_to_drop]
+    return(out)
+  })
+
+  return(my_data_schema_pruned)
+}

vignettes/metadata.Rmd

@@ -215,6 +215,17 @@ expand_frictionless_metadata(structural_metadata = structural_metadata,
                              resource_path = "data_examples/my_data.csv",
                              data_package_path = "data_examples/datapackage.json")
 
+## remove an element from the structural metadata and datapackage
+
+# dropping the comments field because comments field
+structural_metadata <- structural_metadata[-5]
+
+expand_frictionless_metadata(structural_metadata = structural_metadata,
+                             resource_name = "my_data", # name of the file with no extension
+                             resource_path = "data_examples/my_data.csv",
+                             data_package_path = "data_examples/datapackage.json",
+                             prune_datapackage = TRUE) # this is the default
+
 # there are methods for embargoing or restricting deposits in {deposits}