escapement 1.0.0 → 2.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 6fa90e5289492fd35e77857b9caac7a9d70f2bee
-  data.tar.gz: e1786a0edbbb18ec4aa741cd762d73abdd6f20bf
+  metadata.gz: a022be5b1869ab5770ced1ed81a0ffa124739244
+  data.tar.gz: e808af33e9c5a56413dfc3973ccca4ffdce5da60
 SHA512:
-  metadata.gz: a11eca3bcb49edf065e4c5caec62dd61588e77febaea83cbbe757ac8e701ae544aaa1336e6294a69a92899c8e0b5aba4d87d592e543eae3f78f5a276b3ce7750
-  data.tar.gz: a82bf4073571a62cbcac6416680271e253d5d55af1624f9b916a27288d846d803e9434c48f04f8919655d7c49c522102bd9067d684d8548b7773ef3e4fe478a2
+  metadata.gz: 909dd0b5780eea310435ada122f4b93529befa2f2af7be25d5e590cc840349704cc0cedd6e9cf8f9cea8f20f3ac8ddbeb78477ffc77533426e9670c199f4e955
+  data.tar.gz: 35e0febb031007db8d9da3bd114ec530595e6d8eeb9700feebd15699ffaa37ebf35a7025cc9b77c604799df5d49edfe4787cd24f77b9a700518099ccd55dc7f8

data/README.md

@@ -20,7 +20,7 @@ Or install it yourself as:
 
 ## Usage
 
-Basic usage is very straightforward. Escapement will consider all root-level tags as separate paragraphs.
+Basic usage is very straightforward. Escapement will consider all root-level tags as separate elements.
 
 The position values are 0-based and are relative to the plain text result. The first value is the start of the attributed text, and the second is the end of the attributed text.
 
@@ -30,7 +30,19 @@ body = "<p>Isn't <i>Tourbillon</i> a <a href=\"http://google.com\">great</a> wor
 html = Escapement::HTML.new(body)
 html.extract!
 html.results
-# => [{:text=>"Isn't Tourbillon a great word?", :entities=>[{:type=>"italic", :html_tag=>"i", :position=>[6, 16], :attributes=>{}}, {:type=>"link", :html_tag=>"a", :position=>[19, 24], :attributes=>{"href"=>"http://google.com"}}]}] 
+# => [{:type=>"paragraph", :html_tag=>"p", :text=>"Isn't Tourbillon a great word?", :entities=>[{:type=>"italic", :html_tag=>"i", :position=>[6, 16], :attributes=>{}}, {:type=>"link", :html_tag=>"a", :position=>[19, 24], :attributes=>{"href"=>"http://google.com"}}]}]
+```
+
+Escapement also supports lists (with nesting), which treats each list item as a separate paragraph-like element.
+
+``` ruby
+body = "<ul><li>List item 1</li><ul><li><b>Nested</b> list item</li></ul><li>List item 2</li></ul>"
+
+html = Escapement::HTML.new(body)
+html.extract!
+html.results
+
+# => [{:type=>"unordered_list", :html_tag=>"ul", :children=>[{:type=>"list_item", :html_tag=>"li", :text=>"List item 1", :entities=>[]}, {:type=>"unordered_list", :html_tag=>"ul", :children=>[{:type=>"list_item", :html_tag=>"li", :text=>"Nested list item", :entities=>[{:type=>"bold", :html_tag=>"b", :position=>[0, 6], :attributes=>{}}]}]}, {:type=>"list_item", :html_tag=>"li", :text=>"List item 2", :entities=>[]}]}]
 ```
 
 ## How It Works

data/lib/escapement.rb

@@ -1,7 +1,8 @@
 require "nokogiri"
 require "escapement/traversal"
+require "escapement/pretty_names"
 require "escapement/attributes"
-require "escapement/block"
+require "escapement/element"
 require "escapement/html"
 require "escapement/tag"
 require "escapement/version"

data/lib/escapement/element.rb

@@ -0,0 +1,30 @@
+require 'escapement/elements/base'
+require 'escapement/elements/paragraph'
+require 'escapement/elements/list'
+require 'escapement/elements/ordered_list'
+require 'escapement/elements/unordered_list'
+
+module Escapement
+  # An element represents a root-level element in the given
+  # HTML string. Each paragraph has it's own text value and
+  # array of entities.
+  module Element
+    extend self
+
+    NODE_TYPES = [
+      Element::Paragraph,
+      Element::OrderedList,
+      Element::UnorderedList
+    ].freeze
+
+    def factory(node)
+      NODE_TYPES.each do |type|
+        if type.should_handle?(node)
+          return type.new(node)
+        end
+      end
+
+      nil
+    end
+  end
+end

data/lib/escapement/elements/base.rb

@@ -0,0 +1,30 @@
+module Escapement
+  module Element
+    class Base
+      include Traversal
+      include PrettyNames
+
+      attr_reader :node, :result
+
+      def initialize(node)
+        @node = node
+        @entities = []
+        @result = nil
+        @current_position = 0
+      end
+
+      def process!
+        raise "Override"
+      end
+
+      protected
+
+      def node_is_blank?
+        # This will match empty strings, strings with spaces, and
+        # even strings with unicode non-breakable spaces (which can be
+        # produced by  )
+        node.text =~ /\A[[:space:]]*\z/
+      end
+    end
+  end
+end

data/lib/escapement/elements/list.rb

@@ -0,0 +1,20 @@
+module Escapement
+  module Element
+    module List
+      def process!
+        return if node_is_blank?
+
+        @entities = node.children.map { |child|
+          next if child.text?
+          Element.factory(child).tap(&:process!)
+        }.compact
+
+        @result = {
+          type: node_to_type,
+          html_tag: node.name,
+          children: @entities.map(&:result)
+        }
+      end
+    end
+  end
+end

data/lib/escapement/elements/ordered_list.rb

@@ -0,0 +1,11 @@
+module Escapement
+  module Element
+    class OrderedList < Base
+      include List
+
+      def self.should_handle?(node)
+        node.name == 'ol'
+      end
+    end
+  end
+end

data/lib/escapement/elements/paragraph.rb

@@ -0,0 +1,22 @@
+module Escapement
+  module Element
+    class Paragraph < Base
+      def self.should_handle?(node)
+        ['p', 'li'].include? node.name
+      end
+
+      def process!
+        return if node_is_blank?
+
+        process_children
+
+        @result = {
+          type: node_to_type,
+          html_tag: node.name,
+          text: node.text,
+          entities: @entities
+        }
+      end
+    end
+  end
+end

data/lib/escapement/elements/unordered_list.rb

@@ -0,0 +1,11 @@
+module Escapement
+  module Element
+    class UnorderedList < Base
+      include List
+
+      def self.should_handle?(node)
+        node.name == 'ul'
+      end
+    end
+  end
+end

data/lib/escapement/html.rb

@@ -1,22 +1,22 @@
 module Escapement
   # Wrapper around the entire document, which contains an array of
   # results. Each result is the text value and entities for a single
-  # paragraph/block.
+  # element.
   class HTML
-    attr_reader :doc, :blocks, :results
+    attr_reader :doc, :elements, :results
 
     def initialize(html)
       @doc = Nokogiri::HTML(html)
-      @blocks = []
+      @elements = []
       @results = nil
     end
 
-    # Extracts all of the entities for each paragraph/block.
+    # Extracts all of the entities for each element.
     def extract!
       preprocess!
 
-      @blocks = doc.css('body').children.map { |child| Block.new(child).tap(&:process!) }
-      @results = @blocks.reject { |b| b.result.nil? }.map(&:result)
+      @elements = doc.css('body').children.map { |child| Element.factory(child) }.compact
+      @results = @elements.each(&:process!).reject { |e| e.result.nil? }.map(&:result)
     end
 
     private

data/lib/escapement/pretty_names.rb

@@ -0,0 +1,24 @@
+module Escapement
+  module PrettyNames
+    def node_to_type
+      case node.name
+      when 'p' then 'paragraph'
+      when 'ul' then 'unordered_list'
+      when 'ol' then 'ordered_list'
+      when 'a' then 'link'
+      when 'i', 'em' then 'italic'
+      when 'u' then 'underline'
+      when 'strong', 'b' then 'bold'
+      when 'abbr' then 'abbreviation'
+      when 'q' then 'quote'
+      when 'pre' then 'preformatted'
+      when 'img' then 'image'
+      when 'li' then 'list_item'
+      when 'sup' then 'superscript'
+      when 'sub' then 'subscript'
+      when /h\d/ then 'header'
+      else node.name
+      end
+    end
+  end
+end

data/lib/escapement/tag.rb

@@ -4,6 +4,7 @@ module Escapement
   # the traversal until we reach the leaf text node.
   class Tag
     include Traversal
+    include PrettyNames
 
     attr_reader :node, :entities
 
@@ -26,25 +27,6 @@ module Escapement
 
     private
 
-    def node_to_type
-      case node.name
-      when 'p' then 'paragraph'
-      when 'a' then 'link'
-      when 'i', 'em' then 'italic'
-      when 'u' then 'underline'
-      when 'strong', 'b' then 'bold'
-      when 'abbr' then 'abbreviation'
-      when 'q' then 'quote'
-      when 'pre' then 'preformatted'
-      when 'img' then 'image'
-      when 'li' then 'list_item'
-      when 'sup' then 'superscript'
-      when 'sub' then 'subscript'
-      when /h\d/ then 'header'
-      else node.name
-      end
-    end
-
     def filtered_attributes
       method_name = Attributes.respond_to?(node.name) ? node.name : :default
       node.attributes.select(&Attributes.method(method_name))

data/lib/escapement/traversal.rb

@@ -2,7 +2,7 @@ module Escapement
   module Traversal
     # Processes all child nodes of the current node. As the recursion unwinds, we
     # update the entities array such that we're left with a full result set at
-    # the root, which is the Block object.
+    # the root, which is the Element object.
     def process_children
       node.children.each do |child|
         if child.text?

data/lib/escapement/version.rb

@@ -1,3 +1,3 @@
 module Escapement
-  VERSION = "1.0.0"
+  VERSION = "2.0.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: escapement
 version: !ruby/object:Gem::Version
-  version: 1.0.0
+  version: 2.0.0
 platform: ruby
 authors:
 - Ryan LeFevre
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2016-06-30 00:00:00.000000000 Z
+date: 2017-03-21 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
@@ -87,8 +87,14 @@ files:
 - escapement.gemspec
 - lib/escapement.rb
 - lib/escapement/attributes.rb
-- lib/escapement/block.rb
+- lib/escapement/element.rb
+- lib/escapement/elements/base.rb
+- lib/escapement/elements/list.rb
+- lib/escapement/elements/ordered_list.rb
+- lib/escapement/elements/paragraph.rb
+- lib/escapement/elements/unordered_list.rb
 - lib/escapement/html.rb
+- lib/escapement/pretty_names.rb
 - lib/escapement/tag.rb
 - lib/escapement/traversal.rb
 - lib/escapement/version.rb

data/lib/escapement/block.rb

@@ -1,30 +0,0 @@
-module Escapement
-  # A block represents a paragraph, which is a root-level element in the
-  # given HTML string. Each paragraph has it's own text value and array of entities.
-  class Block
-    include Traversal
-
-    attr_reader :node, :result
-
-    def initialize(node)
-      @node = node
-      @entities = []
-      @result = nil
-      @current_position = 0
-    end
-
-    def process!
-      # This will match empty strings, strings with spaces, and
-      # even strings with unicode non-breakable spaces (which can be
-      # produced by  )
-      return if node.text =~ /\A[[:space:]]*\z/
-
-      process_children
-
-      @result = {
-        text: node.text,
-        entities: @entities
-      }
-    end
-  end
-end