docx.rb

gistfile1.rb

# Abstract access to internal files in docx format. Right now this uses system
# zip and tempfiles because the available ruby zip libraries don't seem to be
# able to produce files that Word will read.
#
# Author: Ben Nagy
# Copyright: Copyright (c) Ben Nagy, 2006-2013.
# License: The MIT License
# (See http://www.opensource.org/licenses/mit-license.php for details.)

require 'tempfile'

class Docx

  def initialize( source_fname, tmpdir=Dir.tmpdir )
    if `which zip`.empty? or `which unzip`.empty?
      raise "#{__FILE__}: Unable to find system zip / unzip binaries."
    end
    @file_copy=Tempfile.new ['foo','.docx'], tmpdir, 'wb+'
    File.open( File.expand_path( source_fname ), 'rb') {|io|
      @file_copy.write io.read
    }
    @file_copy.rewind
    @unzipped_dirname="#{@file_copy.path}.unzipped"
    `unzip #{@file_copy.path} -d #{@unzipped_dirname}`
  end

  def files
    Dir["#{@unzipped_dirname}/**/*"].reject {|fname|
      File.directory? fname
    }
  end

  # Add or replace
  def write( fname, replace_str )
    @unzipped_dirname_hackycrap||=@unzipped_dirname + '/'
    relname=fname.rpartition( @unzipped_dirname_hackycrap  ).last
    old=Dir.pwd
    Dir.chdir @unzipped_dirname
    File.open( relname, 'wb+' ) {|io| io.write replace_str}
    `zip -r #{@file_copy.path} #{relname}`
    Dir.chdir old
  end

  def read
    File.open( @file_copy.path, 'rb' ) {|io| io.read}
  end

  def close
    @file_copy.close
    @file_copy.unlink
    `rm -rf #{@unzipped_dirname}`
  end

end