inci_score 3.0.3 → 3.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 3acb19f9ac522f6a50378ec318c6466ae9b67257
-  data.tar.gz: 3d46458825c9d1fb3ff74db0779b329b63c5f97a
+  metadata.gz: 8aab0ebc926de421bb3aaecf7c52a8522c8f47e4
+  data.tar.gz: 1306f8fd4375384c4c2965155759cfe251b515fb
 SHA512:
-  metadata.gz: 9ab029ca732a8605502a7263fa3d7fc96ad20be2ca6aaf9b142e599746eae8216bd326ae318b92aed19a24e157456fe63dd99642d52f536c0a3ea0a4cd59e5bd
-  data.tar.gz: 6d23d330b4267abc0d2ba4241c52d0f55d94922bc1dbb776842e3266afd699f366ade1c648f2e76c3ba2a85db7a680aac59257f683d9d85b5fb2b9b352eae536
+  metadata.gz: 0ce833b953c31f2097356e87c2c33fbbe49e866e26e541b836b8bfba7b2b999ef7cd5ae3d3e81b92ad05e473aabdfec7fb0c45496f7cd3aa096103e3b9fac8a1
+  data.tar.gz: 046d1e9c4a481919546f27287442401ee05d9d2265cd1413eff5bcfb5f75b6de1619439bba3aaeef78b6fb6f8173a93ca70737a165d311c7614990242f46d88b

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    inci_score (3.0.1)
+    inci_score (3.1.0)
       RubyInline (~> 3)
       puma (~> 3)
 

data/README.md

@@ -105,9 +105,8 @@ curl http://127.0.0.1:9292?src=aqua,dimethicone
 ### Getting help
 You can get CLI interface help by:
 ```shell
-Usage: inci_score --src="aqua, parfum, etc" --precise
+Usage: inci_score --src="aqua, parfum, etc"
     -s, --src=SRC                    The INCI list: "aqua, parfum, etc"
-    -p, --precise                    Compute components more precisely (slower)
         --http=PORT                  Start HTTP server on the specified port
     -h, --help                       Prints this help
 ```
@@ -130,12 +129,12 @@ I registered these benchmarks with a MacBook PRO 15 mid 2015 having these specs:
 As always i used [wrk](https://github.com/wg/wrk) as the loading tool.
 I measured the library three times, picking the best lap.  
 ```shell
-wrk -t 4 -c 100 -d 30s --timeout 2000 "http://0.0.0.0:9292/?src=<source>&precise=true"
+wrk -t 4 -c 100 -d 30s --timeout 2000 "http://0.0.0.0:9292/?src=<source>"
 ```
 
 ### Results
 | Source                      | Throughput (req/s) |
 | --------------------------: | -----------------: |
-| aqua,parfum,zeolite         |          18784.21  |
-| agua,porfum,zeolithe        |           1087.88  |
+| aqua,parfum,zeolite         |          20296.75  |
+| agua,porfum,zeolithe        |           1098.45  |
 | agua/water,porfum/fragrance |           1599.47  |

data/config/catalog.yml

@@ -1,5 +1,6 @@
 ---
 aqua: 0
+water: 0
 parfum: 3
 fragrance: 3
 phosphatidylcholine: 1

data/lib/inci_score/api.rb

@@ -12,8 +12,7 @@ module InciScore
     def call(env)
       req = Rack::Request.new(env)
       src = req.params["src"]
-      precise = req.params["precise"]
-      json = src ? Computer.new(src: src, catalog: catalog, precise: !!precise).call.to_json : %q({"error": "no valid source"})
+      json = src ? Computer.new(src: src, catalog: catalog).call.to_json : %q({"error": "no valid source"})
       ["200", {"Content-Type" => "application/json"}, [json]]
     end
   end

data/lib/inci_score/cli.rb

@@ -10,28 +10,23 @@ module InciScore
       @catalog = catalog
       @src = nil
       @port = nil
-      @precise = nil
     end
 
     def call(server_klass: Server, computer_klass: Computer)
       parser.parse!(@args)
       return server_klass.new(port: @port, preload: true).run if @port
       return @io.puts(%q{Specify inci list as: --src="aqua, parfum, etc"}) unless @src
-      @io.puts computer_klass.new(src: @src, catalog: @catalog, precise: @precise).call
+      @io.puts computer_klass.new(src: @src, catalog: @catalog).call
     end
 
     private def parser
       OptionParser.new do |opts|
-        opts.banner = %q{Usage: inci_score --src="aqua, parfum, etc" --precise}
+        opts.banner = %q{Usage: inci_score --src="aqua, parfum, etc"}
 
         opts.on("-sSRC", "--src=SRC", %q{The INCI list: "aqua, parfum, etc"}) do |src|
           @src = src
         end
 
-        opts.on("-p", "--precise", "Compute components more precisely (slower)") do |precise|
-          @precise = precise
-        end
-
         opts.on("--http=PORT", "Start HTTP server on the specified port") do |port|
           @port = port
         end

data/lib/inci_score/computer.rb

@@ -12,14 +12,12 @@ module InciScore
     def initialize(src:, 
                    catalog: Catalog.fetch, 
                    tolerance: TOLERANCE, 
-                   rules: Normalizer::DEFAULT_RULES, 
-                   precise: false)
+                   rules: Normalizer::DEFAULT_RULES)
       @src = src
       @catalog = catalog
       @tolerance = Float(tolerance)
       @rules = rules
       @unrecognized = []
-      @precise = precise
     end
 
     def call
@@ -34,15 +32,12 @@ module InciScore
     end
 
     private def ingredients
-      @ingredients ||= begin
-                         tokens = Normalizer.new(src: @src, rules: @rules).call
-                         Ingredient.bulk(tokens)
-                       end
+      @ingredients ||= Normalizer.new(src: @src, rules: @rules).call
     end
 
     private def components
       @components ||= ingredients.map do |ingredient|
-        Recognizer.new(ingredient, @catalog).call(@precise).tap do |component|
+        Recognizer.new(ingredient, @catalog).call.tap do |component|
           @unrecognized << ingredient unless component
         end
       end.compact

data/lib/inci_score/ingredient.rb

@@ -5,13 +5,9 @@ module InciScore
     PARENTHESIS = %w[( ) [ ]]
     DETAILS_RULE = /(\(.+\)|\[.+\])/
 
-    def self.bulk(tokens)
-      tokens.map { |raw| new(raw) }
-    end
-
     def initialize(raw)
-      @raw = raw
-      @tokens = raw.split(SLASH_RULE).map(&:strip)
+      @raw = raw.to_s
+      @tokens = @raw.split(SLASH_RULE).map(&:strip)
     end
 
     def to_s
@@ -28,7 +24,7 @@ module InciScore
     end
 
     private def synonims
-      @tokens[1, @tokens.size]
+      @tokens[1, @tokens.size].to_a
     end
 
     private def details

data/lib/inci_score/recognizer.rb

@@ -8,30 +8,29 @@ module InciScore
 
     attr_reader :applied
 
-    def initialize(ingredient, catalog, rules = DEFAULT_RULES)
-      @ingredient = ingredient
+    def initialize(ingredient, catalog, rules = DEFAULT_RULES, wrapper = Ingredient)
+      @ingredient = wrapper.new(ingredient)
       @catalog = catalog
       @rules = rules
       @applied = []
     end
 
-    def call(precise = false)
+    def call
       return if @ingredient.to_s.empty?
-      component = find_component(precise)
+      component = find_component
       return unless component
       Component.new(component, @catalog[component])
     end 
 
-    private def find_component(precise)
+    private def find_component
       @rules.reduce(nil) do |component, rule|
         break(component) if component
         applied << rule
-        apply(rule, precise)
+        apply(rule)
       end
     end
 
-    private def apply(rule, precise)
-      return rule.call(@ingredient.to_s, @catalog) unless precise
+    private def apply(rule)
       @ingredient.values.map do |value|
         rule.call(value, @catalog)
       end.find(&:itself)

data/lib/inci_score/recognizer_rules.rb

@@ -58,14 +58,14 @@ module InciScore
         def call(src, catalog)
           tokens(src).each do |token|
             catalog.each do |component, _| 
-              return component if component.matches?(/\b#{Regexp.escape(token)}\b/)
+              return component if component.include?(token)
             end
           end
           nil
         end
 
         def tokens(src)
-          (src.split(" ") - UNMATCHABLE).reject { |t| t.size < TOLERANCE }.sort_by!(&:size).reverse!
+          (src.split(" ") - UNMATCHABLE).reject { |t| t.size < TOLERANCE }.sort! { |a, b| b.size <=> a.size }
         end
       end
     end

data/lib/inci_score/refinements.rb

@@ -10,11 +10,6 @@ module InciScore
       def distance(t)
         InciScore::LevenshteinC.new.call(self.downcase, self.size, t.downcase, t.size)
       end
-
-      def matches?(re)
-        return match?(re) if defined?("".match?)
-        match(re)
-      end
     end
   end
 end

data/lib/inci_score/version.rb

@@ -1,3 +1,3 @@
 module InciScore
-  VERSION = "3.0.3"
+  VERSION = "3.1.0"
 end

data/spec/integration/api_spec.rb

@@ -12,7 +12,7 @@ describe InciScore::Api do
     status, score = Stubs.statuses[i], Stubs.scores[i]
 
     it "cosmetic #{i} must get a proper response" do
-      get "/", src: src, precise: true
+      get "/", src: src
       assert last_response.ok?
       last_response.content_type.must_equal "application/json"
       body = JSON::parse(last_response.body)

data/spec/stubs.rb

@@ -48,7 +48,7 @@ END
   end
   
   def scores
-    [79, 66, 72, 63, 63, 75, 75, 56, 73, 63]
+    [79, 64, 72, 63, 61, 82, 75, 56, 75, 62]
   end
 
   def statuses

data/spec/unit/cli_spec.rb

@@ -12,7 +12,7 @@ describe InciScore::CLI do
     begin
       InciScore::CLI.new(args: %w[--help], io: io, catalog: Stubs.catalog).call
     rescue SystemExit
-      io.string.must_equal "Usage: inci_score --src=\"aqua, parfum, etc\" --precise\n    -s, --src=SRC                    The INCI list: \"aqua, parfum, etc\"\n    -p, --precise                    Compute components more precisely (slower)\n        --http=PORT                  Start HTTP server on the specified port\n    -h, --help                       Prints this help\n"
+      io.string.must_equal "Usage: inci_score --src=\"aqua, parfum, etc\"\n    -s, --src=SRC                    The INCI list: \"aqua, parfum, etc\"\n        --http=PORT                  Start HTTP server on the specified port\n    -h, --help                       Prints this help\n"
     end
   end
 

data/spec/unit/ingredient_spec.rb

@@ -1,12 +1,6 @@
 require "helper"
 
 describe InciScore::Ingredient do
-  it "must create instances in bulk" do
-    InciScore::Ingredient.bulk(Stubs.ingredients.first).each do |ingredient|
-      ingredient.must_be_instance_of InciScore::Ingredient
-    end
-  end
-
   it "must be represented as a string" do
     ingredient = InciScore::Ingredient.new("acrylamidopropyltrimonium chloride/acrylates copolymer")
     ingredient.to_s.must_equal "acrylamidopropyltrimonium chloride/acrylates copolymer"

data/spec/unit/recognizer_spec.rb

@@ -1,5 +1,4 @@
 require "helper"
-require "inci_score/recognizer"
 
 describe InciScore::Recognizer do
   let(:rules) { [] }
@@ -22,6 +21,12 @@ describe InciScore::Recognizer do
     recognizer.applied.must_equal InciScore::Recognizer::DEFAULT_RULES[0,1]
   end
 
+  it "must recognize by key by using synonim" do
+    recognizer = InciScore::Recognizer.new("olio di oliva/olea europea", Stubs.catalog)
+    recognizer.call
+    recognizer.applied.must_equal InciScore::Recognizer::DEFAULT_RULES[0,1]
+  end
+
   it "must recognize by key and levenshtein" do
     recognizer = InciScore::Recognizer.new("agua", Stubs.catalog)
     recognizer.call
@@ -39,11 +44,4 @@ describe InciScore::Recognizer do
     recognizer.call
     recognizer.applied.must_equal InciScore::Recognizer::DEFAULT_RULES
   end
-
-  it "must recognize component precisely" do
-    ingredient = InciScore::Ingredient.new("olio di oliva/olea europea")
-    recognizer = InciScore::Recognizer.new(ingredient, Stubs.catalog)
-    recognizer.call(true)
-    recognizer.applied.must_equal InciScore::Recognizer::DEFAULT_RULES[0,1]
-  end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: inci_score
 version: !ruby/object:Gem::Version
-  version: 3.0.3
+  version: 3.1.0
 platform: ruby
 authors:
 - costajob
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-10-19 00:00:00.000000000 Z
+date: 2017-10-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: puma