basic to_html implemenation

Author: higginsdragon

lib/docx/containers/paragraph.rb

@@ -41,6 +41,18 @@ def to_s
           text_runs.map(&:text).join('')
         end
 
+        # Return paragraph as a <p></p> HTML fragment with formatting based on properties.
+        def to_html
+          html = ''
+          text_runs.each do |text_run|
+            html << text_run.to_html
+          end
+          styles = { 'font-size' => "#{font_size}pt" }
+          styles['text-align'] = alignment if alignment
+          html_tag(:p, content: html, styles: styles)
+        end
+
+
         # 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) }

lib/docx/containers/text_run.rb

@@ -56,7 +56,20 @@ def parse_formatting
         def to_s
           @text
         end
-        
+
+        # Return text as a HTML fragment with formatting based on properties.
+        def to_html
+          html = @text
+          html = html_tag(:em, content: html) if italicized?
+          html = html_tag(:strong, content: html) if bolded?
+          styles = {}
+          styles['text-decoration'] = 'underline' if underlined?
+          # 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?
+          return html
+        end
+
         def italicized?
           @formatting[:italic]
         end

lib/docx/document.rb

@@ -88,6 +88,11 @@ def to_s
       paragraphs.map(&:to_s).join("\n")
     end
 
+    # Output entire document as a String HTML fragment
+    def to_html
+      paragraphs.map(&:to_html).join('\n')
+    end
+
     # Save document to provided path
     # call-seq:
     #   save(filepath) => void

lib/docx/elements/element.rb

@@ -55,6 +55,30 @@ def copy
         self.class.new(@node.dup)
       end
 
+      # A method to wrap content in an HTML tag.
+      # Currently used in paragraph and text_run for the to_html methods
+      #
+      # content:: The base text content for the tag.
+      # styles:: Hash of the inline CSS styles to be applied. e.g.
+      #          { 'font-size' => '12pt', 'text-decoration' => 'underline' }
+      #
+      def html_tag(name, options = {})
+        content = options[:content]
+        styles = options[:styles]
+
+        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
+        html << ">"
+        html << content if content
+        html << "</#{name.to_s}>"
+      end
+
       module ClassMethods
         def create_with(element)
           # Need to somehow get the xml document accessible here by default, but this is alright in the interim

spec/docx/document_spec.rb

@@ -5,6 +5,7 @@
 describe Docx::Document do
   before(:all) do
     @fixtures_path = "spec/fixtures"
+    @formatting_line_count = 11 # number of lines the formatting.docx file has
   end
 
   describe 'reading' do
@@ -166,7 +167,7 @@
     end
 
     it 'should have the correct text' do
-      @doc.paragraphs.size.should eq 11
+      @doc.paragraphs.size.should eq @formatting_line_count
       @doc.paragraphs[0].text.should eq 'Normal'
       @doc.paragraphs[1].text.should eq 'Italic'
       @doc.paragraphs[2].text.should eq 'Bold'
@@ -304,4 +305,70 @@
       end
     end
   end
+
+
+  describe 'outputting html' do
+    before do
+      @doc = Docx::Document.open(@fixtures_path + '/formatting.docx')
+      @formatted_line = @doc.paragraphs[5]
+    end
+
+    it 'should wrap pragraphs in a p tag' do
+      scan = @doc.paragraphs[0].to_html.scan(/(^\<p).+((?<=\>)\w+)(\<\/p>$)/).flatten
+      scan.first.should eq '<p'
+      scan.last.should eq '</p>'
+      scan[1].should eq 'Normal'
+    end
+   
+    it 'should emphasize italicized text' do
+      scan = @doc.paragraphs[1].to_html.scan(/(\<em\>)(\w+)(\<\/em\>)/).flatten
+      scan.first.should eq '<em>'
+      scan.last.should eq '</em>'
+      scan[1].should eq 'Italic'
+    end
+
+    it 'should strong bolded text' do
+      scan = @doc.paragraphs[2].to_html.scan(/(\<strong\>)(\w+)(\<\/strong\>)/).flatten
+      scan.first.should eq '<strong>'
+      scan.last.should eq '</strong>'
+      scan[1].should eq 'Bold'
+    end
+
+    it 'should underline underlined text' do
+      scan = @doc.paragraphs[3].to_html.scan(/\<span\s+([^\>]+)/).flatten
+      scan.first.should eq 'style="text-decoration:underline;"'
+    end
+
+    it 'should justify paragraphs' do
+      regex = /^<p[^\"]+.(?<=\")([^\"]+)/
+      @doc.paragraphs[6].to_html.scan(regex).flatten.first.split(';').include?('text-align:center').should be_true
+      @doc.paragraphs[7].to_html.scan(regex).flatten.first.split(';').include?('text-align:left').should be_false
+      @doc.paragraphs[8].to_html.scan(regex).flatten.first.split(';').include?('text-align:right').should be_true
+    end
+
+    it "should set font size on styled paragraphs" do
+      regex = /(\<p{1})[^\>]+style\=\"([^\"]+).+(<\/p>)/      
+      scan = @doc.paragraphs[9].to_html.scan(regex).flatten
+      scan.first.should eq '<p'
+      scan.last.should eq '</p>'
+      scan[1].split(';').include?('font-size:14pt').should be_true
+    end
+
+    it 'should set font size on styled text runs' do
+      regex = /(\<span)[^\>]+style\=\"([^\"]+)[^\<]+(<\/span>)/
+      scan = @doc.paragraphs[10].to_html.scan(regex).flatten
+      scan.first.should eq '<span'
+      scan.last.should eq '</span>'
+      scan[1].split(';').include?('font-size:16pt').should be_true
+    end
+
+    it 'should output an entire document as html' do
+      @doc.to_html.scan(/(\<p)/).flatten.size.should eq @formatting_line_count
+    end
+
+    it 'should output styled html' do
+      @formatted_line.to_html.scan('<span style="text-decoration:underline;"><strong><em>all</em></strong></span>').size.should eq 1
+    end
+
+  end
 end