normalize_country 0.2.0 → 0.3.1

data/lib/normalize_country/scanner.rb

@@ -0,0 +1,121 @@
+module NormalizeCountry
+  class Tokenizer
+    # Need numbers to match ISO codes
+    # Need "-" to match emoji names
+    # Need regex to match emojis
+    PATTERN = /[[:word:]]+/
+
+    def initialize(s)
+      @scanner = StringScanner.new(s)
+    end
+
+    def scan
+      @scanner.scan_until(PATTERN)
+      @scanner.matched
+    end
+
+    def peek
+      match = scan
+      @scanner.unscan if match
+      match
+    end
+
+    def end?
+      peek.nil?
+    end
+  end
+
+  class Scanner
+    def initialize(options = nil)
+      options ||= {}
+
+      @to = options[:to] || NormalizeCountry.to
+      @table = lookup_table(options[:from] || NormalizeCountry.formats)   # need aliases!!!
+    end
+
+    def convert(text)
+      s = Tokenizer.new(text)
+
+      matches = []
+      stack = []
+      match_position = @table
+
+      while !s.end?
+        word = s.peek.downcase
+        if !match_position[word]
+          s.scan
+          next
+        end
+
+        stack << s.scan
+        alternatives = match_position[stack[-1].downcase]
+
+        peek = s.peek
+        if alternatives && peek && alternatives[peek.downcase]
+          match_position = alternatives
+          next
+        end
+
+        if match_position[stack[-1].downcase][:match]
+          text = stack.join(" ")
+          matches << { :matched => text, :converted => NormalizeCountry(text, :to => @to) }
+
+        end
+
+        stack.clear
+        match_position = @table
+      end
+
+      matches
+    end
+
+    private
+
+    def lookup_table(from_formats)
+      table = {}
+
+      NormalizeCountry::Countries.values.uniq.each_with_object(table) do |country, o|
+        # country.name # eq all
+        # no way to get aliases
+        from_formats.each do |format|
+          name = country[format]
+          next unless name
+
+          head = o
+          parts = name.split(/[[:space:]]+/)
+          parts.each_with_index do |word, i|
+            # options[:case_sensitive_codes] = [x,y] # or true
+            # options[:case_sensitive_formats] = [x,y]
+            # options[:case_sensitive_formats] = true   # alpha2, alpha3, fifa, ioc
+            word.downcase! #unless parts.size == 1 && ABBRV_FORMATS.include?(format)
+
+            # if head[word]
+            #   if i == parts.size - 1
+            #     head[word][:match] = true
+            #   else
+            #     head = head[word]
+            #   end
+            # else
+            #   head[word] = {}
+            #   if i == parts.size - 1
+            #     head[word][:match] = true
+            #   else
+            #     head = head[word]
+            #   end
+            # end
+
+            head[word] = {} unless head[word]
+
+            if i == parts.size - 1
+              head[word][:match] = true
+            else
+              head = head[word]
+            end
+          end
+        end
+
+        o
+      end
+    end
+  end
+end

data/spec/normalize_country_spec.rb

@@ -4,7 +4,7 @@ require "minitest/autorun"
 require "normalize_country"
 
 describe NormalizeCountry do
-  COUNTRY_COUNT = 247
+  COUNTRY_COUNT = 249
 
   it "normalizes to a country's ISO name by default" do
     NormalizeCountry.convert("USA").must_equal("United States")
@@ -110,15 +110,9 @@ describe NormalizeCountry do
 
   describe ".formats" do
     it "returns a list of supported formats" do
-      expected = [:alpha2, :alpha3, :fifa, :ioc, :iso_name, :numeric, :official, :short, :simple, :emoji]
+      expected = [:alpha2, :alpha3, :fifa, :ioc, :iso_name, :numeric, :official, :short, :shortcode, :simple, :emoji]
       formats  = NormalizeCountry.formats
-
-      # Ugh, support this in 1.8.7 for a least one version
-      if Symbol < Comparable
-        formats.sort.must_equal(expected.sort)
-      else
-        formats.sort_by { |f| f.to_s }.must_equal(expected.sort_by { |f| f.to_s })
-      end
+      formats.sort.must_equal(expected.sort)
     end
   end
 

metadata

@@ -1,29 +1,29 @@
 --- !ruby/object:Gem::Specification
 name: normalize_country
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.3.1
 platform: ruby
 authors:
 - Skye Shaw
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2016-06-13 00:00:00.000000000 Z
+date: 2021-01-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: '0.9'
+        version: 12.3.3
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - "~>"
+    - - ">="
       - !ruby/object:Gem::Version
-        version: '0.9'
+        version: 12.3.3
 - !ruby/object:Gem::Dependency
   name: minitest
   requirement: !ruby/object:Gem::Requirement
@@ -54,6 +54,7 @@ files:
 - bin/normalize_country
 - lib/normalize_country.rb
 - lib/normalize_country/countries/en.yml
+- lib/normalize_country/scanner.rb
 - spec/normalize_country_spec.rb
 homepage: http://github.com/sshaw/normalize_country
 licenses:
@@ -75,7 +76,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.5.1
+rubygems_version: 2.7.6
 signing_key: 
 specification_version: 4
 summary: Convert country names and codes to a standard