Merge pull request #173 from ruby-docx/add-header-extraction

Add ability to read document headers

Author: satoryu

lib/docx/document.rb

@@ -22,7 +22,7 @@ module Docx
   class Document
     include Docx::SimpleInspect
 
-    attr_reader :xml, :doc, :zip, :styles
+    attr_reader :xml, :doc, :zip, :styles, :headers
 
     def initialize(path_or_io, options = {})
       @replace = {}
@@ -40,6 +40,7 @@ def initialize(path_or_io, options = {})
       @document_xml = document.get_input_stream.read
       @doc = Nokogiri::XML(@document_xml)
       load_styles
+      load_headers
       yield(self) if block_given?
     ensure
       @zip.close unless @zip.nil?
@@ -200,6 +201,15 @@ def with_zip64_disabled
       Zip.write_zip64_support = previous
     end
 
+    def load_headers
+      header_files = @zip.glob("word/header*.xml").map{|h| h.name}
+      filename_and_contents_pairs = header_files.map do |file|
+        simple_file_name = file.sub(/^word\//, "").sub(/\.xml$/, "")
+        [simple_file_name, Nokogiri::XML(@zip.read(file))]
+      end
+      @headers = Hash[filename_and_contents_pairs]
+    end
+
     def load_styles
       @styles_xml = @zip.read('word/styles.xml')
       @styles = Nokogiri::XML(@styles_xml)

spec/docx/document_spec.rb

@@ -60,6 +60,18 @@
     end
   end
 
+  describe 'read headers' do
+    before do
+      @doc = Docx::Document.open(@fixtures_path + '/multi_doc.docx')
+    end
+
+    it 'can extract headers' do
+      expect(@doc.headers).to_not be_nil
+      expect(@doc.headers.keys).to eq ["header1"]
+      expect(@doc.headers["header1"].text).to eq "Hello from the header."
+    end
+  end
+
   describe 'read tables' do
     before do
       @doc = Docx::Document.open(@fixtures_path + '/tables.docx')