food_ingredient_parser 1.1.1 → 1.1.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 41da81db767bedd583db2794f44d39d710ddef58
-  data.tar.gz: 9bc24f8e83d04fb608e8fba4364aa26117fd898b
+  metadata.gz: 64ad7a10a1480b520602113bbcdfc10ba1daf8b5
+  data.tar.gz: 4068a9edbe1dca908228f38d2795ad63a5cbcf76
 SHA512:
-  metadata.gz: 7634ebb4a8f3530cab515c5b27d2782df8c2521eb79df67ecbe08945360bcc27d68456d389d69a83808792145ecfb7094b3b97b5c33cbb17119642449813df88
-  data.tar.gz: 3b5c5ed78b7c4e41211855523acaf2a4c519b0e021a32eb698a715f7dccf2beacde229593d8422b71ce7a5a55dc1e266347d263c6629a72cdb308556697f7d93
+  metadata.gz: 73ce876757b08e1d2cf0b5126e8d024b3728260134c3c4f3fe49fee14793da77ecc48d286165dc0c86e8363f2eddf6081355ac26d38f524371b367f4aa3cee23
+  data.tar.gz: befa97dc0fd4605cd2019a2cf7a39aa15d5dfa15acf118f10e9b28e104bbc9bd5925f28286e780acb142eb73a935dba26657f1c6fe9cb766e7f4f88310d5ce55

data/README.md

@@ -65,10 +65,10 @@ Usage: bin/food_ingredient_parser [options] --file|-f <filename>
     -s, --string INGREDIENTS         Parse specified ingredient list.
     -q, --[no-]quiet                 Only show summary.
     -p, --parsed                     Only show lines that were successfully parsed.
+    -n, --noresult                   Only show lines that had no result.
     -r, --parser PARSER              Use specific parser (strict, loose).
     -e, --[no-]escape                Escape newlines
     -c, --[no-]color                 Use color
-    -n, --noresult                   Only show lines that had no result.
     -v, --[no-]verbose               Show more data (parsed tree).
         --version                    Show program version.
     -h, --help                       Show this help
@@ -190,5 +190,4 @@ upgrade to `1.1`.
 [`data/ingredient-samples-nl`](data/ingredient-samples-nl) contains about 150k
 real-world ingredient lists found on the Dutch market. Each line contains one ingredient
 list (newlines are encoded as `\n`, empty lines and those starting with `#` are ignored).
-The strict parser currently parses about three quarter, while the loose parser returns
-something for all of them.
+The strict parser currently parses 80%, while the loose parser returns something for all of them.

data/bin/food_ingredient_parser

@@ -88,10 +88,10 @@ OptionParser.new do |opts|
 
   opts.on("-q", "--[no-]quiet", "Only show summary.") {|q| verbosity = q ? 0 : 1 }
   opts.on("-p", "--parsed", "Only show lines that were successfully parsed.") {|p| print = :parsed }
+  opts.on("-n", "--noresult", "Only show lines that had no result.") {|p| print = :noresult }
   opts.on("-r", "--parser PARSER", "Use specific parser (#{PARSERS.keys.join(", ")}).") {|p| parser_name = p&.downcase&.to_sym }
   opts.on("-e", "--[no-]escape", "Escape newlines") {|e| escape = !!e }
   opts.on("-c", "--[no-]color", "Use color") {|e| color = !!e }
-  opts.on("-n", "--noresult", "Only show lines that had no result.") {|p| print = :noresult }
   opts.on("-v", "--[no-]verbose", "Show more data (parsed tree).") {|v| verbosity = v ? 2 : 1 }
   opts.on(      "--version", "Show program version.") do
     puts("food_ingredient_parser v#{FoodIngredientParser::VERSION}")

data/lib/food_ingredient_parser/loose/parser.rb

@@ -2,6 +2,7 @@ require_relative '../cleaner'
 require_relative 'scanner'
 require_relative 'transform/amount'
 require_relative 'transform/handle_missing_name'
+require_relative 'transform/split_e_numbers'
 
 module FoodIngredientParser::Loose
   class Parser
@@ -20,6 +21,7 @@ module FoodIngredientParser::Loose
       s = FoodIngredientParser::Cleaner.clean(s) if clean
       n = Scanner.new(s).scan
       n = Transform::Amount.transform!(n) if n
+      n = Transform::SplitENumbers.transform!(n) if n
       n = Transform::HandleMissingName.transform!(n) if n && normalize
       n
     end

data/lib/food_ingredient_parser/loose/transform/split_e_numbers.rb

@@ -0,0 +1,50 @@
+require_relative '../node'
+
+module FoodIngredientParser::Loose
+  module Transform
+    class SplitENumbers
+      # Transforms node tree to split e-number combinations.
+      #
+      # @note mark and amount is lost, this is not expected on e-numbers
+
+      SPLIT_RE = /\s*-\s*/.freeze
+      MATCH_RE = /\A\s*(e[0-9]{3}[a-z]?)(?:#{SPLIT_RE}(e[0-9]{3}[a-z]?))+\s*\z/i.freeze
+
+      def self.transform!(node)
+        new(node).transform!
+      end
+
+      def initialize(node)
+        @node = node
+      end
+
+      def transform!
+        transform_node!(@node)
+        @node
+      end
+
+      private
+
+      def transform_node!(node)
+        if node.contains.any?
+          node.contains.each {|n| transform_node!(n) }
+        elsif node.name && m = MATCH_RE.match(node.name.text_value)
+          i = 0
+          while m = node.name.text_value.match(SPLIT_RE, i)
+            node.contains << new_node(node, i, m.begin(0)-1)
+            i = m.end(0)
+          end
+          node.contains << new_node(node, i, node.name.interval.last) if i <= node.name.interval.last
+          node.name = nil
+        end
+      end
+
+      def new_node(node, begins, ends)
+        offset = node.name.interval.first
+        new_node = Node.new(node.input, offset + begins .. offset + ends)
+        new_node.name = Node.new(node.input, new_node.interval)
+        new_node
+      end
+    end
+  end
+end

data/lib/food_ingredient_parser/strict/grammar/ingredient_coloned.treetop

@@ -10,7 +10,8 @@ module FoodIngredientParser::Strict::Grammar
     end
 
     rule ingredient_coloned_inner_list
-      contains:( ingredient_coloned_simple_with_amount_and_nest ws* ( '/'+ ws* ingredient_coloned_simple_with_amount_and_nest )* ) <ListNode>
+      contains:( ingredient_simple_e_number ( ws* dash ws* ingredient_simple_e_number )+ ) <ListNode> /
+      contains:( ingredient_coloned_simple_with_amount_and_nest ( ws* '/'+ ws* ingredient_coloned_simple_with_amount_and_nest )* ) <ListNode>
     end
 
     # @see IngredientSimple#ingredient_simple

data/lib/food_ingredient_parser/strict/grammar/ingredient_simple.treetop

@@ -14,5 +14,9 @@ module FoodIngredientParser::Strict::Grammar
       ing:ingredient_simple                                                      <IngredientNode>
     end
 
+    rule ingredient_simple_e_number
+      name:( [Ee] [0-9] [0-9] [0-9] [a-zA-Z]? ) ![a-zA-Z0-9] <IngredientNode>
+    end
+
   end
 end

data/lib/food_ingredient_parser/strict/grammar/list_coloned.treetop

@@ -13,6 +13,7 @@ module FoodIngredientParser::Strict::Grammar
     end
 
     rule list_coloned_inner_list
+      contains:( ingredient_simple_e_number ( ws* dash ws* ingredient_simple_e_number )+ ) <ListNode> /
       contains:( ingredient ( ws* ',' ws* ingredient )* ) <ListNode>
     end
 

data/lib/food_ingredient_parser/version.rb

@@ -1,4 +1,4 @@
 module FoodIngredientParser
-  VERSION      = '1.1.1'
-  VERSION_DATE = '2018-09-25'
+  VERSION      = '1.1.2'
+  VERSION_DATE = '2018-09-28'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: food_ingredient_parser
 version: !ruby/object:Gem::Version
-  version: 1.1.1
+  version: 1.1.2
 platform: ruby
 authors:
 - wvengen
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-09-25 00:00:00.000000000 Z
+date: 2018-09-28 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: treetop
@@ -50,6 +50,7 @@ files:
 - lib/food_ingredient_parser/loose/transform/amount.rb
 - lib/food_ingredient_parser/loose/transform/amount_from_name.treetop
 - lib/food_ingredient_parser/loose/transform/handle_missing_name.rb
+- lib/food_ingredient_parser/loose/transform/split_e_numbers.rb
 - lib/food_ingredient_parser/strict/grammar.rb
 - lib/food_ingredient_parser/strict/grammar/amount.treetop
 - lib/food_ingredient_parser/strict/grammar/common.treetop