inci_score 4.6.0 → 4.6.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 7c1ec290c92d1fb34798dbd04ec8f93a6ad1db12fd124ff2f7d0c4a50c108aa7
-  data.tar.gz: dad169f8a2486aeeac3a22df47097235d965b9435350d4800d662c9022f0a36c
+  metadata.gz: fbbf88644bed7f8afcd802d796c2b49a2428a2eac2bbf68761079b83bc485866
+  data.tar.gz: 014d402eacfdb5512b350fdb7131dee8ebc793df1091153b30c04e000e47d161
 SHA512:
-  metadata.gz: 34394e8b825952f6ff98696ba899592f23a0acd8e3b95331c1d645903ac9795569dc6e115ff1dca5df0ae4aee551b54aae0e32b91faa302b4338b4e37769d797
-  data.tar.gz: 1b932c8c80adfc5b0989383a7c31c7b78432890dd3b86c54c887b1d3146f6629332cb7abd16e50b70343113bf3beb76233f4cd0d77fd684e26e7377ca0a8ee7f
+  metadata.gz: 44d96ce1e40671e04de3f9120465a129abbbc6bbd104c0e639d78be6d03be52645dedad5f5c22f7ff0c2d02c115fbf32d6941e38d9aa86867f9e4e38763719d3
+  data.tar.gz: 4dcfd975146932d346b4ba693621fd00762b83071b6cf91db73757bec4271d530ab253b42a2e647c661d8e37712682e4cef51c9befa0a5b6717232ddc785b77d

data/README.md

@@ -31,11 +31,13 @@ The total score is then calculated on a percent basis.
 Since the ingredients list could come from an unreliable source (e.g. data scanned from a captured image), the gem tries to fuzzy match the ingredients by using different algorithms:
 * exact matching
 * [edit distance](https://en.wikipedia.org/wiki/Levenshtein_distance) behind a specified tolerance
+* known hazards (ie ending in `ethicone`) 
 * first relevant matching digits 
 * matching splitted tokens
 
 ### Sources
-The library accepts the list of ingredients as a single string of text. Since this source could come from an OCR program, the library performs a normalization by stripping invalid characters and removing the unimportant parts.  
+The library accepts the list of ingredients as a single string of text.  
+Since this source could come from an OCR program, the library performs a normalization by stripping invalid characters and removing the unimportant parts.  
 The ingredients are typically separated by comma, although normalizer will detect the most appropriate separator:
 
 ```
@@ -44,6 +46,7 @@ The ingredients are typically separated by comma, although normalizer will detec
 
 ## Installation
 Install the gem from your shell:
+
 ```shell
 gem install inci_score
 ```
@@ -58,18 +61,21 @@ require "inci_score"
 
 inci = InciScore::Computer.new(src: 'aqua, dimethicone').call
 inci.score # 56.25
+inci.precision # 100.0
 ```
 
-As you see the results are wrapped by an *InciScore::Response* object, this is useful when dealing with the CLI and HTTP interfaces (read below).
+As you see the results are wrapped by an *InciScore::Response* object, this is useful when dealing with the CLI (read below).
 
 #### Unrecognized components
-The API treats unrecognized components as a common case by just marking the object as non valid and raise a warning in case more than 30% of the ingredients are not found.  
+The API treats unrecognized components as a common case by just marking the object as non valid.  
 In such case the score is computed anyway by considering only recognized components.  
-Is still possible to query the object for its state:
+You can check the `precision` value, which is zero for unrecognized components, and changes based on the applied recognizer rule (100% when exact matching).
 
 ```ruby
 inci = InciScore::Computer.new(src: 'ingredients:aqua,noent1,noent2')
 inci.valid? # false
+inci.score # 100.0
+inci.precision # 33.33
 inci.unrecognized # ["noent1", "noent2"]
 ```
 
@@ -82,7 +88,7 @@ inci_score --src="ingredients: aqua, dimethicone, pej-10, noent"
 TOTAL SCORE:
       	53.22
 PRECISION:
-      	75.0
+      	71.54
 COMPONENTS:
       	aqua (0), dimethicone (4), peg-10 (3)
 UNRECOGNIZED:
@@ -91,6 +97,7 @@ UNRECOGNIZED:
 
 #### Getting help
 You can get CLI interface help by:
+
 ```shell
 Usage: inci_score --src="aqua, parfum, etc"
     -s, --src=SRC                    The INCI list: "aqua, parfum, etc"

data/config/catalog.yml

@@ -2818,7 +2818,7 @@ pantothenic acid polypeptide: 0
 papaver orientale: 0
 papaver rhoeas: 0
 paraffin: 3
-paraffinum liquidum: 3
+paraffinum liquidum (mineral oil): 3
 parietaria officinalis: 0
 passiflora incarnata: 0
 passiflora quadrangularis: 0

data/lib/inci_score/cli.rb

@@ -15,7 +15,7 @@ module InciScore
 
     def call
       parser.parse!(args)
-      return io.puts(%q{Specify inci list as: --src='aqua, parfum, etc'}) unless src
+      return io.puts(%q{Specify INCI list as: --src='aqua, parfum, etc'}) unless src
       computer = Computer.new(src: src)
       io.puts computer.call
     end

data/lib/inci_score/computer.rb

@@ -2,13 +2,14 @@
 
 module InciScore
   class Computer
-    TOLERANCE = 30.0
+    TOLERANCE = 50.0
     DECIMALS = 2
 
-    attr_reader :src, :rules, :ingredients, :components, :unrecognized
+    attr_reader :src, :rules, :ingredients, :components, :unrecognized, :precisions
 
     def initialize(src:, rules: Normalizer::DEFAULT_RULES)
       @unrecognized = []
+      @precisions = []
       @src = src
       @rules = rules
       @ingredients = Normalizer.new(src: src, rules: rules).call
@@ -28,7 +29,7 @@ module InciScore
     end
 
     def precision
-      (100 - ((unrecognized.size / Float(ingredients.size)) * 100)).round(DECIMALS)
+      (precisions.sum / ingredients.size).round(DECIMALS)
     end
 
     def valid?
@@ -39,7 +40,9 @@ module InciScore
 
     def fetch_components
       ingredients.map do |ingredient|
-        Recognizer.new(ingredient).call.tap do |component|
+        recognizer = Recognizer.new(ingredient)
+        recognizer.call.tap do |component|
+          precisions << recognizer.precision
           unrecognized << ingredient unless component
         end
       end.compact

data/lib/inci_score/recognizer.rb

@@ -3,19 +3,31 @@
 module InciScore
   class Recognizer
     DEFAULT_RULES = [Rules::Key, Rules::Levenshtein, Rules::Hazard, Rules::Prefix, Rules::Tokens].freeze
+    PRECISION_BASE = 4
 
     attr_reader :ingredient, :rules, :applied
+    attr_accessor :found
 
     def initialize(ingredient, rules = DEFAULT_RULES)
       @ingredient = Ingredient.new(ingredient)
       @rules = rules
       @applied = []
-      freeze
+      @found = false
     end
 
     def call
       return if ingredient.to_s.empty?
-      find_component
+      find_component.tap do |c|
+        self.found = true if c
+      end
+    end
+
+    def precision
+      return 0.0 unless found
+      rule = applied.last
+      index = rules.index(rule) + PRECISION_BASE
+      ratio = Math.log(index, PRECISION_BASE)
+      (100 / ratio).round(2)
     end
 
     private

data/lib/inci_score/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module InciScore
-  VERSION = '4.6.0'
+  VERSION = '4.6.3'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: inci_score
 version: !ruby/object:Gem::Version
-  version: 4.6.0
+  version: 4.6.3
 platform: ruby
 authors:
 - costajob
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-01-05 00:00:00.000000000 Z
+date: 2023-01-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: oj