Merge pull request #92 from gopeter/master

Added support for hyperlinks (implemented #70 again)

Author: satoryu

lib/docx/containers/paragraph.rb

@@ -55,7 +55,7 @@ def to_html
 
         # Array of text runs contained within paragraph
         def text_runs
-          @node.xpath('w:r|w:hyperlink/w:r').map { |r_node| Containers::TextRun.new(r_node, @document_properties) }
+          @node.xpath('w:r|w:hyperlink').map { |r_node| Containers::TextRun.new(r_node, @document_properties) }
         end
 
         # Iterate over each text run within a paragraph

lib/docx/containers/text_run.rb

@@ -23,6 +23,8 @@ def self.tag
         def initialize(node, document_properties = {})
           @node = node
           @text_nodes = @node.xpath('w:t').map {|t_node| Elements::Text.new(t_node) }
+          @text_nodes = @node.xpath('w:t|w:r/w:t').map {|t_node| Elements::Text.new(t_node) }
+
           @properties_tag = 'rPr'
           @text       = parse_text || ''
           @formatting = parse_formatting || DEFAULT_FORMATTING
@@ -74,6 +76,7 @@ def to_html
           # No need to be granular with font size down to the span level if it doesn't vary.
           styles['font-size'] = "#{font_size}pt" if font_size != @font_size 
           html = html_tag(:span, content: html, styles: styles) unless styles.empty?
+          html = html_tag(:a, content: html, attributes: {href: href, target: "_blank"}) if hyperlink?
           return html
         end
 
@@ -89,6 +92,18 @@ def underlined?
           @formatting[:underline]
         end
 
+        def hyperlink?
+          @node.name == 'hyperlink'
+        end
+
+        def href
+          @document_properties[:hyperlinks][hyperlink_id]
+        end
+
+        def hyperlink_id
+          @node.attributes['id'].value
+        end        
+
         def font_size
           size_tag = @node.xpath('w:rPr//w:sz').first
           size_tag ? size_tag.attributes['val'].value.to_i / 2 : @font_size

lib/docx/document.rb

@@ -45,7 +45,8 @@ def initialize(path_or_io, options = {})
     # This stores the current global document properties, for now
     def document_properties
       {
-        font_size: font_size
+        font_size: font_size,
+        hyperlinks: hyperlinks
       }
     end
 
@@ -83,6 +84,17 @@ def font_size
       size_tag ? size_tag.attributes['val'].value.to_i / 2 : nil
     end
 
+    # Hyperlink targets are extracted from the document.xml.rels file
+    def hyperlinks
+      hyperlink_relationships.each_with_object({}) do |rel, hash|
+        hash[rel.attributes['Id'].value] = rel.attributes['Target'].value
+      end 
+    end
+
+    def hyperlink_relationships
+      @rels.xpath("//xmlns:Relationship[contains(@Type,'hyperlink')]")
+    end    
+
     ##
     # *Deprecated*
     #
@@ -157,6 +169,8 @@ def replace_entry(entry_path, file_contents)
     def load_styles
       @styles_xml = @zip.read('word/styles.xml')
       @styles = Nokogiri::XML(@styles_xml)
+      @rels_xml = @zip.read('word/_rels/document.xml.rels')
+      @rels = Nokogiri::XML(@rels_xml)
     rescue Errno::ENOENT => e
       warn e.message
       nil

lib/docx/elements/element.rb

@@ -65,15 +65,24 @@ def copy
       def html_tag(name, options = {})
         content = options[:content]
         styles = options[:styles]
+        attributes = options[:attributes]
 
         html = "<#{name.to_s}"
+        
         unless styles.nil? || styles.empty?
           styles_array = []
           styles.each do |property, value|
             styles_array << "#{property.to_s}:#{value};"
           end
           html << " style=\"#{styles_array.join('')}\""
         end
+        
+        unless attributes.nil? || attributes.empty?
+          attributes.each do |attr_name, attr_value|
+            html << " #{attr_name}=\"#{attr_value}\""
+          end
+        end
+        
         html << ">"
         html << content if content
         html << "</#{name.to_s}>"

spec/docx/document_spec.rb

@@ -7,7 +7,7 @@
 describe Docx::Document do
   before(:all) do
     @fixtures_path = 'spec/fixtures'
-    @formatting_line_count = 12 # number of lines the formatting.docx file has
+    @formatting_line_count = 13 # number of lines the formatting.docx file has
   end
 
   describe '#open' do
@@ -205,6 +205,8 @@
       expect(@doc.paragraphs[8].text).to eq('This paragraph is aligned right.')
       expect(@doc.paragraphs[9].text).to eq('This paragraph is 14 points.')
       expect(@doc.paragraphs[10].text).to eq('This paragraph has a word at 16 points.')
+      expect(@doc.paragraphs[11].text).to eq('This sentence has different formatting in different places.')
+      expect(@doc.paragraphs[12].text).to eq('This sentence has a hyperlink.')
     end
 
     it 'should contain a paragraph with multiple text runs' do
@@ -354,6 +356,7 @@
       @span_regex = /(\<span).+((?<=\>)\w+)(<\/span>)/
       @em_regex = /(\<em).+((?<=\>)\w+)(\<\/em\>)/
       @strong_regex = /(\<strong).+((?<=\>)\w+)(\<\/strong\>)/
+      @anchor_tag_regex = /\<a href="(.+)" target="_blank"\>(.+)\<\/a>/
     end
 
     it 'should wrap pragraphs in a p tag' do
@@ -434,6 +437,12 @@
     it 'should join paragraphs with newlines' do
       expect(@doc.to_html.scan(%(<p style="font-size:11pt;">Normal</p>\n<p style="font-size:11pt;"><em>Italic</em></p>\n<p style="font-size:11pt;"><strong>Bold</strong></p>)).size).to eq 1
     end
+
+    it 'should convert hyperlinks to anchor tags' do
+      scan = @doc.to_html.scan(@anchor_tag_regex).flatten
+      expect(scan[0]).to eq "http://www.google.com/"
+      expect(scan[1]).to eq "hyperlink"
+    end
   end
 
   describe 'replacing contents' do