food_ingredient_parser 1.0.0 → 1.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 2c529b63bd3a9f6139ed10663b2cef70ff3d1dc6
-  data.tar.gz: a8effec91559e15920794c61b08a486e572032cf
+  metadata.gz: 629792af43b06c646b98c2ab8ee136895123e3e5
+  data.tar.gz: 6abbe88e79943cedb8b8776146aab6fdffd427e1
 SHA512:
-  metadata.gz: 35f27c7d83effc16962a65ac4c8c09fb5694373dbd3d2745c434c37ddcf3fc466264c0f10cbe5054876517839a81bf75ab0b1b9876098c1b04b5312138e06ea1
-  data.tar.gz: 28e517777928262b45836d899ff919f725e09ab6e116fd9f58545262b7ae7151821ae683e6d33146d3b5e20f5c02ad7217c4985267460645b0ba80e1ccc19751
+  metadata.gz: c1728c6edc995f6a5a18b82eae0e217fa18814c75f1b34d33d52abc4f593428020ba6a425ea526519036a5d4a412a80b43aae55147b2a07bae0e0979394b26bd
+  data.tar.gz: a9f43ad8e20ab867293e1db59aeb5b7bd6147a6e560adb58159f345d441c6445be453b7eb4ba01bdd4169ae951444e4bf90a6e7873cf85cfa7557b4ac4859d36

data/README.md

@@ -121,7 +121,7 @@ When ingredient lists are entered manually, it can be very useful to show how th
 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. (Available for strict parser only.)
+text, augmented with CSS classes for different parts.
 
 ```ruby
 require 'food_ingredient_parser'
@@ -178,8 +178,8 @@ Even though the strict parser would not give a result, the loose parser returns:
 
 From the 1.0.0 release, the main interface will be stable. This comprises the two parser's `parse`
 methods (incl. documented options), its `nil` result when parsing failed, and the parsed output's
-`to_h` and `to_html` methods (where available). Please note that parsed node trees may be subject to
-change, even within a major release. Within a minor release, node trees are expected to remain stable.
+`to_h` and `to_html` methods. Please note that parsed node trees may be subject to change, even within
+a major release. Within a minor release, node trees are expected to remain stable.
 
 So if you only use the stable interface (`parse`, `to_h` and `to_html`), you can lock your version
 to e.g. `~> 1.0`. If you depend on more, lock your version against e.g. `~> 1.0.0` and test when you

data/lib/food_ingredient_parser/loose/node.rb

@@ -1,6 +1,10 @@
+require_relative 'to_html'
+
 module FoodIngredientParser::Loose
   # Parsing result.
   class Node
+    include ToHtml
+
     attr_accessor :name, :mark, :amount, :contains, :notes
     attr_reader :input, :interval, :auto_close
 

data/lib/food_ingredient_parser/loose/scanner.rb

@@ -149,9 +149,10 @@ module FoodIngredientParser::Loose
 
     def is_notes_start?
       # @todo use more heuristics: don't assume dot is notes when separator is a dot, and only toplevel?
-      if ( is_mark? && @s[@i+mark_len..-1] =~ /\A\s*=/ ) ||     # "* = Biologisch"
-         ( is_mark? && @s[@i-1] =~ /\s/ ) ||                    # " **Biologisch"
-         ( @s[@i..-1] =~ NOTE_RE )                              # "E=", "Kan sporen van", ...
+      ml = mark_len
+      if ( is_mark? && @s[@i+ml .. -1] =~ /\A\s*=/ ) ||                      # "* = Biologisch"
+         ( is_mark? && @s[@i-1] =~ /\s/ && @s[@i+ml .. -1] =~ /\A\s*\w/ ) || # " **Biologisch"
+         ( @s[@i..-1] =~ NOTE_RE )                                           # "E=", "Kan sporen van", ...
         @i -= 1 # we want to include the mark in the note
         true
       # End of sentence
@@ -196,7 +197,7 @@ module FoodIngredientParser::Loose
       cur.name ||= begin
         i, j = cur.interval.first, @i - 1
         i += mark_len(i) # skip any mark in front
-        Node.new(@s, i .. j) if j > i
+        Node.new(@s, i .. j) if j >= i
       end
     end
 

data/lib/food_ingredient_parser/loose/to_html.rb

@@ -0,0 +1,66 @@
+require 'cgi'
+
+# Adds HTML output functionality to a Node.
+#
+module FoodIngredientParser::Loose
+  module 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, depth=0)
+      children = [*node.contains, *node.notes, node.name, node.amount, node.mark].compact
+      children.sort_by! {|n| n.interval.first }
+
+      html = ""
+      last_idx = node.interval.first - 1
+      children.each do |child|
+        # we don't have nodes for all characters, make sure they are in the output
+        if child.interval.first - 1 > last_idx
+          html += CGI.escapeHTML(node.input[last_idx + 1 .. child.interval.first - 1])
+          last_idx = child.interval.first - 1
+        end
+
+        if child == node.name
+          html += node_to_html_single(child, "name")
+          last_idx = child.interval.last
+        elsif child == node.amount
+          html += node_to_html_single(child, "amount")
+          last_idx = child.interval.last
+        elsif child == node.mark
+          html += node_to_html_single(child, "mark")
+          last_idx = child.interval.last
+        elsif node.notes.include?(child)
+          html += node_to_html_single(child, "note")
+          last_idx = child.interval.last
+        elsif node.contains.include?(child)
+          cls = "depth#{depth}"
+          cls = "contains #{cls}" if depth > 0
+          html += "<span class='#{cls}'>#{node_to_html(child, depth + 1)}</span>"
+          last_idx = child.interval.last
+        end
+      end
+
+      # include any trailing characters
+      if children.any? && last_idx < node.interval.last
+        html += CGI.escapeHTML(node.input[last_idx + 1 .. node.interval.last])
+      end
+
+      html
+    end
+
+    def node_to_html_single(node, cls=nil)
+      ws1, txt, ws2 = node.text_value.match(/\A(\s*)(.*?)(\s*)\z/).captures.map {|s| CGI.escapeHTML(s)}
+      cls && txt.size > 0 ? "#{ws1}<span class='#{cls}'>#{txt}</span>#{ws2}" : "#{ws1}#{txt}#{ws2}"
+    end
+  end
+end

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

@@ -8,7 +8,9 @@ module FoodIngredientParser::Loose
     #
     # When a contained node is found which doesn't have a name:
     # * For the amount (if any): ignore it (as it's often ambiguous which ingredient it belongs to)
-    # * For the marks (if any): ignore it (we might instead add it to the containing ingredients)
+    # * For the marks (if any)
+    #   - if the node has no siblings and no containing ingredients, add the mark to the parent (if any)
+    #   - else ignore it (we might instead add it to the containing ingredients)
     # * For the containing ingredients (if any):
     #   - if the previous ingredient is present and doesn't contain ingredients already,
     #     assume the current contained ingredients are actually part of the previous ingredient.
@@ -45,6 +47,9 @@ module FoodIngredientParser::Loose
             if prev
               # there is a previous ingredient: move children to new parent
               prev.contains.push(*child.contains)
+            elsif child.mark && !node.mark && child.contains.empty? && !child.amount
+              # this is just a mark without siblings: it's a mark for its parent
+              node.mark = child.mark
             else
               # there is no previous ingredient: move children one level up
               new_contains.push(*child.contains)

data/lib/food_ingredient_parser/version.rb

@@ -1,4 +1,4 @@
 module FoodIngredientParser
-  VERSION      = '1.0.0'
-  VERSION_DATE = '2018-09-21'
+  VERSION      = '1.1.0'
+  VERSION_DATE = '2018-09-24'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: food_ingredient_parser
 version: !ruby/object:Gem::Version
-  version: 1.0.0
+  version: 1.1.0
 platform: ruby
 authors:
 - wvengen
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-09-21 00:00:00.000000000 Z
+date: 2018-09-24 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: treetop
@@ -46,6 +46,7 @@ files:
 - lib/food_ingredient_parser/loose/node.rb
 - lib/food_ingredient_parser/loose/parser.rb
 - lib/food_ingredient_parser/loose/scanner.rb
+- lib/food_ingredient_parser/loose/to_html.rb
 - 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