food_ingredient_parser 1.0.0.pre.3 → 1.0.0.pre.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: d687958a818bdfe88bfa30bce9ff2a87db5073fe
-  data.tar.gz: 69c9e1547c02635b305003e16a62666fb9fc4c68
+  metadata.gz: b180a987ae477627cd2046f967d54eeb160fc7ab
+  data.tar.gz: '0981287cb5348c58a5fc2e0fda39c470a0b10746'
 SHA512:
-  metadata.gz: afed94a75e31892e5e80f5b84b3d920634a670fe7dd061a611f2fcbc5adf8ebe294a6709f17a7c2c0fcafc3c6fc12f36c5aff93ce6b1936947fd2984fc393a0c
-  data.tar.gz: a3d81d578349e279f8630b661b26693c008321b5ed1b2fc76872e8de8175c2258afe744e7ca70eb192a4a65357c3995157e776584d0ca3bb1d23c44a6f4173d8
+  metadata.gz: e11033d79eff439544d71c789981361c98192d767687148cca73e02308f288aac474b84f8ffe1b2a08901df481e4a7c03c162bbcc8ed6f5c030b48f12e4ebb61
+  data.tar.gz: a44018d0bc1c77d7907e5fa31cf5fe7e919c2e20bf5480b07bbd0de33ff2e6aec4a2a7981b8a22906c574f9140bede380cc87a7917fc84a0f6685190d26925c7

data/README.md

@@ -108,6 +108,28 @@ parsed 35 (100.0%), no result 0 (0.0%)
 
 If you want to use the output in (shell)scripts, the options `-e -c` may be quite useful.
 
+## `to_html`
+
+When ingredient lists are entered manually, it can be very useful to show how the text is
+recognized. This can help understanding why a certain ingredients list cannot be parsed.
+
+For this you can use the `to_html` method on the parsed output, which returns the original
+text, augmented with CSS classes for different parts.
+
+```ruby
+require 'food_ingredient_parser'
+
+parsed = FoodIngredientParser::Parser.new.parse("Saus (10% tomaat*, zout). * = bio")
+puts parsed.to_html
+```
+
+```html
+<span class='name'>Saus</span> (<span class='amount'>10%</span>
+<span class='name'>tomaat</span><span class='mark'>*</span>,
+<span class='name'>zout</span>). <span class='note'>* = bio</span>
+```
+
+
 ## Test data
 
 [`data/ingredient-samples-nl`](data/ingredient-samples-nl) contains about 150k

data/food_ingredient_parser.gemspec

@@ -4,7 +4,7 @@ require 'food_ingredient_parser/version'
 Gem::Specification.new do |s|
   s.name             = 'food_ingredient_parser'
   s.version          = FoodIngredientParser::VERSION
-  s.date             = '2018-08-31'
+  s.date             = '2018-09-05'
   s.summary          = 'Parser for ingredient lists found on food products.'
   s.authors          = ['wvengen']
   s.email            = ['dev-ruby@willem.engen.nl']

data/lib/food_ingredient_parser/grammar/amount.treetop

@@ -11,19 +11,19 @@ module FoodIngredientParser::Grammar
 
     rule simple_amount
       ( (
-        'of which' / 'at least' /
-        'waarvan' / 'ten minste' / 'tenminste' / 'minimaal'
+        'of which'i / 'at least'i / 'minimal'i / 'maximal'i / 'less than'i / 'more than'i /
+        'waarvan'i / 'ten minste'i / 'tenminste'i / 'minimaal'i / 'maximaal'i / 'minder dan'i / 'meer dan'i
       ) ws* )?
-      [<>]? ws*
+      [±∓~∼∽≂≃≈≲≤<>≥≳]? ws*
       simple_amount_quantity
       ( ws+ (
-        'minimum' /
-        'minimaal' / 'minimum'
+        'minimum'i /
+        'minimaal'i / 'minimum'i
       ) )?
     end
 
     rule simple_amount_quantity
-      number ( ws* '-' ws* number )? ws* ( '%' / 'g' / 'mg' / 'gram' / 'ml' )
+      number ( ws* '-' ws* number )? ws* ( '%' / 'g'i / 'mg'i / 'gram'i / 'ml'i )
     end
 
   end

data/lib/food_ingredient_parser/grammar/common.treetop

@@ -20,7 +20,7 @@ module FoodIngredientParser::Grammar
 
     rule mark
       # mark referencing a footnote
-      [¹²³⁴⁵ᵃᵇᶜᵈᵉᶠᵍªº] '⁾'? / '⁽' [¹²³⁴⁵ᵃᵇᶜᵈᵉᶠᵍªº] '⁾' / [†‡•°#^] / '*'+
+      [¹²³⁴⁵ᵃᵇᶜᵈᵉᶠᵍªº] '⁾'? / '⁽' [¹²³⁴⁵ᵃᵇᶜᵈᵉᶠᵍªº] '⁾' / [†‡•°#^] / '*'+ / '(' ws* ( [†‡•°#^] / '*'+ ) ws* ')'
     end
 
     rule digit

data/lib/food_ingredient_parser/nodes.rb

@@ -1,4 +1,5 @@
 require 'treetop/runtime'
+require_relative 'to_html'
 
 # Needs to be in grammar namespace so Treetop can find the nodes.
 module FoodIngredientParser::Grammar
@@ -18,6 +19,8 @@ module FoodIngredientParser::Grammar
 
   # Root object, contains everything else.
   class RootNode < SyntaxNode
+    include FoodIngredientParser::ToHtml
+
     def to_h
       h = { contains: contains.to_a }
       if notes && notes_ary = to_a_deep(notes, NoteNode)&.map(&:text_value)

data/lib/food_ingredient_parser/parser.rb

@@ -5,20 +5,23 @@ module FoodIngredientParser
 
     # @!attribute [r] parser
     #   @return [Treetop::Runtime::CompiledParser] low-level parser object
+    #   @note This attribute is there for convenience, but may change in the future. Take care.
     attr_reader :parser
 
     # Create a new food ingredient parser
-    # @return [FoodIngredientParser]
+    # @return [FoodIngredientParser::Parser]
     def initialize
       @parser = Grammar::RootParser.new
     end
 
     # Parse food ingredient list text into a structured representation.
+    #
     # @option clean [Boolean] pass +false+ to disable correcting frequently occuring issues
-    # @return [Hash] structured representation of food ingredients
-    def parse(s, clean: true)
+    # @return [FoodIngredientParser::Grammar::RootNode] structured representation of food ingredients
+    # @note Unrecognized options are passed to Treetop, but this is not guarenteed to remain so forever.
+    def parse(s, clean: true, **options)
       s = clean(s) if clean
-      @parser.parse(s)
+      @parser.parse(s, **options)
     end
 
     private

data/lib/food_ingredient_parser/to_html.rb

@@ -0,0 +1,43 @@
+require 'cgi'
+
+# Adds HTML output functionality to a Treetop Node.
+#
+# The node needs to provide a {#to_h} method (for {#to_html_h}).
+#
+module FoodIngredientParser::ToHtml
+
+  # Markup original ingredients list text in HTML.
+  #
+  # The input text is returned as HTML, augmented with CSS classes
+  # on +span+s for +name+, +amount+, +mark+ and +note+.
+  #
+  # @return [String] HTML representation of ingredient list.
+  def to_html
+    node_to_html(self)
+  end
+
+  private
+
+  def node_to_html(node, cls=nil)
+    el_cls = {}               # map of node instances to class names for contained elements
+    terminal = node.terminal? # whether to look at children elements or not
+
+    if node.is_a?(FoodIngredientParser::Grammar::AmountNode)
+      cls ||= "amount"
+    elsif node.is_a?(FoodIngredientParser::Grammar::NoteNode)
+      cls ||= "note"
+      terminal = true # NoteNodes may contain other NoteNodes, we want it flat.
+    elsif node.is_a?(FoodIngredientParser::Grammar::IngredientNode)
+      el_cls[node.name] = "name" if node.respond_to?(:name)
+      el_cls[node.mark] = "mark" if node.respond_to?(:mark)
+    end
+
+    val = if terminal
+      CGI.escapeHTML(node.text_value)
+    else
+      node.elements.map {|el| node_to_html(el, el_cls[el]) }.join("")
+    end
+
+    cls ? "<span class='#{cls}'>#{val}</span>" : val
+  end
+end

data/lib/food_ingredient_parser/version.rb

@@ -1,3 +1,3 @@
 module FoodIngredientParser
-  VERSION = '1.0.0.pre.3'
+  VERSION = '1.0.0.pre.4'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: food_ingredient_parser
 version: !ruby/object:Gem::Version
-  version: 1.0.0.pre.3
+  version: 1.0.0.pre.4
 platform: ruby
 authors:
 - wvengen
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-08-31 00:00:00.000000000 Z
+date: 2018-09-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: treetop
@@ -55,6 +55,7 @@ files:
 - lib/food_ingredient_parser/grammar/root.treetop
 - lib/food_ingredient_parser/nodes.rb
 - lib/food_ingredient_parser/parser.rb
+- lib/food_ingredient_parser/to_html.rb
 - lib/food_ingredient_parser/version.rb
 homepage: https://github.com/q-m/food-ingredient-parser-ruby
 licenses: