inci_score 2.2.1 → 2.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 8bc9d1f6ee80c730c5d5f12837fed8d774950369
-  data.tar.gz: ccb0394552ba0e2966c5260e84c338aa61b58a09
+  metadata.gz: 9d5e69b680cde0fed823fd4672b8b4ec16638fe9
+  data.tar.gz: 488e31a72fd25148a61b32c5fa2a589b6ef438c3
 SHA512:
-  metadata.gz: 629adf023e81fa545af1dbb675abe4467ef9aae42bb9f22e73d691339a8e9e585e7719902cf2f3b09c9a4d64c46032d17e9d72d92a378ec6a8d4a4beb9ff8e25
-  data.tar.gz: 36af9780ea2534d18a92be8d122649596a692f18720d5da5eb63b1a44cc9d7c4afe6c55b9a57fc33810f61d8fba5811912481ec0f70b975c751fc7d26334a203
+  metadata.gz: 9e9eb717b3460fd139330acd40f7eec880b8b42f6c806f2ff71efdd9f8b2f8a4376a82c2f4133fcc619fa102781e7b96f4a17bb8bb8b81ed245ba11e76a85271
+  data.tar.gz: 7b29bf4e4ab728772be548ba3f81f87ecb6e4b14d8f8e6952a194469314d1445996d9761c90372d19fa99de3030a61f33be7761aaddc98580bc72ce0100ac020

data/README.md

@@ -145,11 +145,10 @@ I measured the library three times, picking the best lap.
 The following script command is used:
 
 ```shell
-wrk -t 4 -c 100 -d 30s --timeout 2000 http://127.0.0.1:9292/?src=<list_of_ingredients>
+wrk -t 4 -c 100 -d 30s --timeout 2000 http://127.0.0.1:9292/?src=aqua,parfum,zeolithe
 ```
 
 ### Results
-| Ingredients              | Throughput (req/s) | Latency in ms (avg/stdev/max) |
-| :----------------------- | -----------------: | ----------------------------: |
-| aqua,parfum,zeolite      |           26054.91 |               0.63/1.03/79.86 |
-| agua,porfum,zaolite      |             953.44 |              14.67/5.15/82.31 |
+| Throughput (req/s) | Latency in ms (avg/stdev/max) |
+| -----------------: | ----------------------------: |
+|           50060.92 |                0.31/0.45/3.83 |

data/Rakefile

@@ -1,23 +1,23 @@
-require 'bundler/gem_tasks'
-require 'rake/testtask'
+require "bundler/gem_tasks"
+require "rake/testtask"
 
 namespace :spec do
   Rake::TestTask.new(:unit) do |t|
-    t.libs << 'spec'
-    t.libs << 'lib'
-    t.test_files = FileList['spec/unit/*_spec.rb']
+    t.libs << "spec"
+    t.libs << "lib"
+    t.test_files = FileList["spec/unit/*_spec.rb"]
   end
 
   Rake::TestTask.new(:integration) do |t|
-    t.libs << 'spec'
-    t.libs << 'lib'
-    t.test_files = FileList['spec/integration/*_spec.rb']
+    t.libs << "spec"
+    t.libs << "lib"
+    t.test_files = FileList["spec/integration/*_spec.rb"]
   end
 
   Rake::TestTask.new(:bench) do |t|
-    t.libs << 'spec'
-    t.libs << 'lib'
-    t.test_files = FileList['spec/bench/*_bench.rb']
+    t.libs << "spec"
+    t.libs << "lib"
+    t.test_files = FileList["spec/bench/*_bench.rb"]
   end
 end
 

data/bin/console

@@ -1,7 +1,7 @@
 #!/usr/bin/env ruby
 
-require 'bundler/setup'
-require 'inci_score'
-require 'irb'
-require 'irb/completion'
+require "bundler/setup"
+require "inci_score"
+require "irb"
+require "irb/completion"
 IRB.start

data/lib/inci_score/app.rb

@@ -1,5 +1,5 @@
-require 'rack'
-require 'inci_score'
+require "rack"
+require "inci_score"
 
 module InciScore
   module App
@@ -12,8 +12,9 @@ module InciScore
     def call(env)
       req = Rack::Request.new(env)
       src = req.params["src"]
-      json = src ? Computer.new(src: src, catalog: catalog).call.to_json : %q({"error": "no valid source"})
-      ['200', {'Content-Type' => 'application/json'}, [json]]
+      precise = req.params["precise"]
+      json = src ? Computer.new(src: src, catalog: catalog, precise: !!precise).call.to_json : %q({"error": "no valid source"})
+      ["200", {"Content-Type" => "application/json"}, [json]]
     end
   end
 end

data/lib/inci_score/catalog.rb

@@ -1,10 +1,10 @@
-require 'yaml'
+require "yaml"
 
 module InciScore
   module Catalog
     extend self
 
-    YAML_PATH = File::expand_path('../../../config/catalog.yml', __FILE__) 
+    YAML_PATH = File::expand_path("../../../config/catalog.yml", __FILE__) 
 
     def fetch(src = File.read(YAML_PATH))
       YAML::load(src)

data/lib/inci_score/cli.rb

@@ -11,20 +11,21 @@ module InciScore
       @src = nil
       @fresh = nil
       @port = nil
+      @precise = nil
     end
 
     def call(server_klass: Server, computer_klass: Computer, fetcher: Fetcher.new)
       parser.parse!(@args)
       return server_klass.new(port: @port, preload: true).run if @port
-      return @io.puts("Specify inci list as: --src='aqua, parfum, etc'") unless @src
-      @io.puts computer_klass.new(src: @src, catalog: catalog(fetcher)).call
+      return @io.puts(%q{Specify inci list as: --src="aqua, parfum, etc"}) unless @src
+      @io.puts computer_klass.new(src: @src, catalog: catalog(fetcher), precise: @precise).call
     end
 
     private def parser
       OptionParser.new do |opts|
-        opts.banner = %q{Usage: ./bin/inci_score --src='aqua, parfum, etc' --fresh}
+        opts.banner = %q{Usage: inci_score --src="aqua, parfum, etc" --fresh --precise}
 
-        opts.on("-sSRC", "--src=SRC", "The INCI list: 'aqua, parfum, etc'") do |src|
+        opts.on("-sSRC", "--src=SRC", %q{The INCI list: "aqua, parfum, etc"}) do |src|
           @src = src
         end
 
@@ -32,6 +33,10 @@ module InciScore
           @fresh = fresh
         end
 
+        opts.on("-p", "--precise", "Compute components more precisely (slower)") do |precise|
+          @precise = precise
+        end
+
         opts.on("--http=PORT", "Start Puma server on the specified port") do |port|
           @port = port
         end

data/lib/inci_score/computer.rb

@@ -1,18 +1,19 @@
-require 'inci_score/normalizer'
-require 'inci_score/recognizer'
-require 'inci_score/scorer'
-require 'inci_score/response'
+require "inci_score/normalizer"
+require "inci_score/recognizer"
+require "inci_score/scorer"
+require "inci_score/response"
 
 module InciScore
   class Computer
     TOLERANCE = 30.0
 
-    def initialize(src:, catalog:, tolerance: TOLERANCE, rules: Normalizer::DEFAULT_RULES)
+    def initialize(src:, catalog:, tolerance: TOLERANCE, rules: Normalizer::DEFAULT_RULES, precise: false)
       @src = src
       @catalog = catalog
       @tolerance = Float(tolerance)
       @rules = rules
       @unrecognized = []
+      @precise = precise
     end
 
     def call
@@ -32,7 +33,7 @@ module InciScore
 
     private def components
       @components ||= ingredients.map do |ingredient|
-        Recognizer.new(ingredient, @catalog).call.tap do |component|
+        Recognizer.new(ingredient, @catalog).call(@precise).tap do |component|
           @unrecognized << ingredient unless component
         end
       end.compact

data/lib/inci_score/fetcher.rb

@@ -1,10 +1,10 @@
-require 'nokogiri'
+require "nokogiri"
 
 module InciScore
   class Fetcher
-    BIODIZIO_URI = 'http://www.biodizionario.it/biodizio.php'
+    BIODIZIO_URI = "http://www.biodizionario.it/biodizio.php"
     SEMAPHORES = %w[vv v g r rr]
-    CSS_QUERY = 'table[width="751"] > tr > td img'
+    CSS_QUERY = %q{table[width="751"] > tr > td img}
 
     def initialize(src = nil)
       @src = src || Thread.new { open(BIODIZIO_URI) }
@@ -12,7 +12,7 @@ module InciScore
 
     def call
       @components ||= Nokogiri::HTML(doc).css(CSS_QUERY).inject({}) do |acc, img|
-        hazard = semaphore(img.attr('src'))
+        hazard = semaphore(img.attr("src"))
         name = img.next_sibling.next_sibling
         desc = name.next_sibling.next_sibling
         name, desc = desc, name if swap?(desc.text)
@@ -26,7 +26,7 @@ module InciScore
     end
 
     private def semaphore(src)
-      src.match(/(#{SEMAPHORES.join('|')}).gif$/)[1]
+      src.match(/(#{SEMAPHORES.join("|")}).gif$/)[1]
     end
 
     private def normalize(node)

data/lib/inci_score/levenshtein.rb

@@ -1,8 +1,8 @@
-require 'inline'
+require "inline"
 
 module InciScore
   class LevenshteinC
-    C_PROGRAM = File::expand_path('../../../ext/levenshtein.c', __FILE__)
+    C_PROGRAM = File::expand_path("../../../ext/levenshtein.c", __FILE__)
 
     inline(:C) do |builder|
       builder.c File::read(C_PROGRAM) 

data/lib/inci_score/normalizer.rb

@@ -1,14 +1,14 @@
-require 'inci_score/normalizer_rules'
+require "inci_score/normalizer_rules"
 
 module InciScore
   class Normalizer
-    DEFAULT_RULES = [Rules::Replacer, Rules::Downcaser, Rules::Beheader, Rules::Separator, Rules::Tokenizer, Rules::Sanitizer, Rules::Desynonymizer]
+    DEFAULT_RULES = [Rules::Replacer, Rules::Downcaser, Rules::Beheader, Rules::Separator, Rules::Tokenizer, Rules::Sanitizer, Rules::Desynonymizer, Rules::Uniquifier]
 
     attr_reader :src
 
-    def initialize(options = {})
-      @src = options[:src] || fail(ArgumentError, 'missing src')
-      @rules = options.fetch(:rules) { DEFAULT_RULES }
+    def initialize(src:, rules: DEFAULT_RULES)
+      @src = src
+      @rules = rules
     end
 
     def call

data/lib/inci_score/normalizer_rules.rb

@@ -88,6 +88,14 @@ module InciScore
           end.reject(&:empty?)
         end
       end
+
+      module Uniquifier
+        extend self
+
+        def call(src)
+          Array(src).uniq
+        end
+      end
     end
   end
 end

data/lib/inci_score/recognizer.rb

@@ -1,4 +1,4 @@
-require 'inci_score/recognizer_rules'
+require "inci_score/recognizer_rules"
 
 module InciScore
   class Recognizer
@@ -10,11 +10,11 @@ module InciScore
       @rules = rules
     end
 
-    def call
+    def call(precise = false)
       @component = @rules.reduce(nil) do |component, rule|
         break(component) if component
         yield(rule) if block_given?
-        rule.call(@src, @catalog)
+        rule.call(@src, @catalog, precise)
       end
       [@component, @catalog[@component]] if @component
     end 

data/lib/inci_score/recognizer_rules.rb

@@ -1,4 +1,4 @@
-require 'inci_score/refinements'
+require "inci_score/refinements"
 
 module InciScore
   using Refinements
@@ -9,7 +9,7 @@ module InciScore
       module Key
         extend self
 
-        def call(src, catalog)
+        def call(src, catalog, precise = false)
           src if catalog.has_key?(src)
         end
       end
@@ -17,13 +17,13 @@ module InciScore
       module Levenshtein
         extend self
 
-        ALTERNATE_SEP  = '/'
+        ALTERNATE_SEP  = "/"
 
-        def call(src, catalog)
+        def call(src, catalog, precise = false)
           size = src.size
           initial = src[0]
           component, distance = catalog.reduce([nil, size]) do |min, (_component, _)|
-            next min unless _component.start_with?(initial)
+            next min unless precise || _component.start_with?(initial)
             match = (n = _component.index(ALTERNATE_SEP)) ? _component[0, n] : _component
             next min if match.size > (size + TOLERANCE)
             dist = src.distance(match)
@@ -39,13 +39,18 @@ module InciScore
 
         MIN_MEANINGFUL = 7
 
-        def call(src, catalog)
+        def call(src, catalog, precise = false)
           return if src.size < TOLERANCE
-          digits = src[0, MIN_MEANINGFUL]
+          digits = src[0, min_meaningful(precise)]
           catalog.detect do |component, _| 
             component.matches?(/^#{Regexp::escape(digits)}/)
           end.to_a.first
         end
+
+        def min_meaningful(precise)
+          return MIN_MEANINGFUL unless precise
+          MIN_MEANINGFUL + 2
+        end
       end
 
       module Tokens
@@ -53,7 +58,7 @@ module InciScore
 
         UNMATCHABLE = %w[extract oil sodium acid sulfate]
       
-        def call(src, catalog)
+        def call(src, catalog, precise = false)
           tokens(src).each do |token|
             catalog.each do |component, _| 
               return component if component.matches?(/\b#{Regexp.escape(token)}\b/)
@@ -63,7 +68,7 @@ module InciScore
         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_by!(&:size).reverse!
         end
       end
     end

data/lib/inci_score/server.rb

@@ -1,5 +1,6 @@
-require 'etc'
-require 'puma'
+require "etc"
+require "puma/launcher"
+require "puma/configuration"
 
 module InciScore
   class Server
@@ -29,13 +30,21 @@ module InciScore
         c.rackup RACKUP_FILE
         c.bind "tcp://#{DEFAULT_HOST}:#{@port}"
         c.workers @workers if workers?
-        c.threads *@threads
+        c.threads(*@threads)
         c.preload_app! if @preload
       end
     end
 
     private def workers?
-      @workers > 1 && !Puma.jruby? && !Puma.windows?
+      @workers > 1 && !java? && !windows?
+    end
+
+    private def java?
+      RUBY_VERSION == "java"
+    end
+
+    private def windows?
+      Gem.win_platform?
     end
   end
 end

data/lib/inci_score/version.rb

@@ -1,3 +1,3 @@
 module InciScore
-  VERSION = "2.2.1"
+  VERSION = "2.3.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: inci_score
 version: !ruby/object:Gem::Version
-  version: 2.2.1
+  version: 2.3.0
 platform: ruby
 authors:
 - costajob
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-01-05 00:00:00.000000000 Z
+date: 2017-05-10 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri