pumi 0.12.2 → 0.13.1

data/lib/pumi/data_parser.rb

@@ -0,0 +1,75 @@
+require "pathname"
+require "csv"
+require "yaml"
+
+# https://en.wikipedia.org/wiki/Administrative_divisions_of_Cambodia
+# https://en.wikipedia.org/wiki/Romanization_of_Khmer
+# https://en.wikipedia.org/wiki/United_Nations_Group_of_Experts_on_Geographical_Names
+
+module Pumi
+  class DataParser
+    CSV_HEADERS = %w[type code name_km name_latin reference note1 note2].freeze
+
+    AdministrativeUnit = Struct.new(:en, :km, :latin, :code_length, :group, keyword_init: true)
+    ADMINISTRATIVE_UNITS = {
+      "ស្រុក" => AdministrativeUnit.new(en: "District", km: "ស្រុក", latin: "Srok", code_length: 4, group: "districts"),
+      "ខណ្ឌ" => AdministrativeUnit.new(en: "Section", km: "ខណ្ឌ", latin: "Khan", code_length: 4, group: "districts"),
+      "ក្រុង" => AdministrativeUnit.new(en: "Municipality", km: "ក្រុង", latin: "Krong", code_length: 4, group: "districts"),
+      "ឃុំ" => AdministrativeUnit.new(en: "Commune", km: "ឃុំ", latin: "Khum", code_length: 6, group: "communes"),
+      "សង្កាត់" => AdministrativeUnit.new(en: "Quarter", km: "សង្កាត់", latin: "Sangkat", code_length: 6, group: "communes"),
+      "ភូមិ" => AdministrativeUnit.new(en: "Village", km: "ភូមិ", latin: "Phum", code_length: 8, group: "villages")
+    }.freeze
+
+    def load_data!(source_dir: "tmp")
+      data = {}
+
+      source_files(source_dir).each do |file|
+        CSV.read(file, headers: CSV_HEADERS).each do |row|
+          code = row.fetch("code")
+          next if code.to_s.gsub(/\D/, "").empty?
+
+          code = code.rjust(code.length + 1, "0") if code.length.odd?
+          administrative_unit = ADMINISTRATIVE_UNITS.fetch(row.fetch("type"))
+
+          next if administrative_unit.code_length != code.length
+
+          data[administrative_unit.group] ||= {}
+          data[administrative_unit.group][code] = {
+            "name" => {
+              "km" => row.fetch("name_km"),
+              "latin" => row.fetch("name_latin")
+            },
+            "administrative_unit" => {
+              "km" => administrative_unit.km,
+              "latin" => administrative_unit.latin,
+              "en" => administrative_unit.en
+            }
+          }
+        end
+      end
+
+      data
+    end
+
+    def write_data!(data, destination_dir: "data")
+      return if data.empty?
+
+      data_groups.each do |group|
+        File.write(
+          "#{destination_dir}/#{group}.yml",
+          { group => data.fetch(group).sort.to_h }.to_yaml
+        )
+      end
+    end
+
+    private
+
+    def source_files(source_dir)
+      Pathname.glob("#{source_dir}/*.csv").select(&:file?)
+    end
+
+    def data_groups
+      ADMINISTRATIVE_UNITS.values.map(&:group).uniq
+    end
+  end
+end

data/lib/pumi/version.rb

@@ -1,3 +1,3 @@
 module Pumi
-  VERSION = "0.12.2".freeze
+  VERSION = "0.13.1".freeze
 end

data/lib/pumi.rb

@@ -23,3 +23,4 @@ require_relative "pumi/district"
 require_relative "pumi/commune"
 require_relative "pumi/village"
 require_relative "pumi/parser"
+require_relative "pumi/data_parser"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pumi
 version: !ruby/object:Gem::Version
-  version: 0.12.2
+  version: 0.13.1
 platform: ruby
 authors:
 - David Wilkie
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2023-02-01 00:00:00.000000000 Z
+date: 2023-03-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bump
@@ -274,6 +274,7 @@ files:
 - lib/pumi.rb
 - lib/pumi/administrative_unit.rb
 - lib/pumi/commune.rb
+- lib/pumi/data_parser.rb
 - lib/pumi/data_store.rb
 - lib/pumi/district.rb
 - lib/pumi/location.rb