roo 2.0.1 → 2.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: db86b4586034783940303cfd17a7ee5c60277a51
-  data.tar.gz: 2cf9222f2f2759f164164f745c72abe6b901df98
+  metadata.gz: 97a621e940844c1142964501c7a78fa0cd054c86
+  data.tar.gz: 9db0bb973c120511a2bba167b152b3ff6dadb488
 SHA512:
-  metadata.gz: 442e05a8bffdc839c45a8684e8882f4327098387617b76027b6b5c0543c2a7b4be284c511c35f8333c9bc1147e31350d1f6f67174fc62e5b7fc45520ab0ad307
-  data.tar.gz: 4ba67602e5772cae52f41ea0357ddd6835d25c35be57797307a6194a46ad977d7922b884b4fa826c91efd8533c6f3e56995e8cb0a13146a32e4a3ee7d5e42ae9
+  metadata.gz: 0831fd41e73b58b153b04ebda1ccbed591219b15d69070e4a5b50c4298c2b12a396feb0ffcd3f0d6a0b94ea6848897ae2324cd12cee721c38beea82c8e512b26
+  data.tar.gz: e409801f0e341dcde7333830ec08b4150ec7e8310924583fac260ace3816c2bb950bb419c3d2060578fbd1b7e129c34bb2f825ffcde60fdde1c4a9fb384f2937

data/.travis.yml

@@ -10,4 +10,5 @@ matrix:
   allow_failures:
     - rvm: ruby-head
     - rvm: jruby-19mode
+    - rvm: rbx-2
 bundler_args: --without local_development

data/CHANGELOG.md

@@ -1,3 +1,14 @@
+## [2.1.0] - 2015-07-18
+### Added
+- Added support for Excel 2007 `xlsm` files. [#232](https://github.com/roo-rb/roo/pull/232)
+- Roo::Excelx returns an enumerator when calling each_row_streaming without a block. [#224](https://github.com/roo-rb/roo/pull/224)
+- Returns an enumerator when calling `each` without a block. [#219](https://github.com/roo-rb/roo/pull/219)
+
+### Fixed
+- Removed tabs and windows CRLF. [#235](https://github.com/roo-rb/roo/pull/235), [#234](https://github.com/roo-rb/roo/pull/234)
+- Fixed Regexp to only check for valid URI's when opening a spreadsheet. [#229](https://github.com/roo-rb/roo/pull/228)
+- Open streams in Roo:Excelx correctly. [#222](https://github.com/roo-rb/roo/pull/222)
+
 ## [2.0.1] - 2015-06-01
 ### Added
 - Return an enumerator when calling '#each' without a block [#219](https://github.com/roo-rb/roo/pull/219)
@@ -464,7 +475,7 @@
 
 ## [0.2.4] - 2007-06-16
 ### Fixed
-- ID 11605	Two cols with same value: crash roo (openoffice version only)
+- ID 11605  Two cols with same value: crash roo (openoffice version only)
 
 ## [0.2.3] - 2007-06-02
 ### Changed / Added

data/README.md

@@ -12,7 +12,7 @@ Roo implements read access for all common spreadsheet types. It can handle:
 
 In addition, the [roo-xls](https://github.com/roo-rb/roo-xls) and [roo-google](https://github.com/roo-rb/roo-google) gems exist to extend Roo to support reading classic Excel formats (i.e. `.xls` and ``Excel2003XML``) and read/write access for Google spreadsheets.
 
-# #Installation
+## Installation
 
 Install as a gem
 
@@ -21,7 +21,7 @@ Install as a gem
 Or add it to your Gemfile
 
 ```ruby
-gem 'roo', '~> 2.0.0'
+gem 'roo', '~> 2.1.0'
 ```
 ## Usage
 
@@ -102,7 +102,7 @@ s.cell(1,'A',s.sheets[1])
 ```
 
 #### Querying a spreadsheet
-Use ``each`` with a ``block`` to iterate over each row.
+Use ``each`` to iterate over each row.
 
 If each is given a hash with the names of some columns, then each will generate a hash with the columns supplied for each row.
 
@@ -143,7 +143,7 @@ sheet.to_xml
 sheet.to_yaml
 ```
 
-### Excel (xlsx) Support
+### Excel (xlsx and xlsm) Support
 
 Stream rows from an Excelx spreadsheet.
 

data/lib/roo.rb

@@ -1,3 +1,4 @@
+require 'roo/constants'
 require 'roo/spreadsheet'
 require 'roo/base'
 
@@ -10,17 +11,18 @@ module Roo
   CLASS_FOR_EXTENSION = {
     ods: Roo::OpenOffice,
     xlsx: Roo::Excelx,
+    xlsm: Roo::Excelx,
     csv: Roo::CSV
   }
 
   def self.const_missing(const_name)
     case const_name
     when :Excel
-      raise "Excel support has been extracted to roo-xls due to its dependency on the GPL'd spreadsheet gem. Install roo-xls to use Roo::Excel."
+      raise ROO_EXCEL_NOTICE
     when :Excel2003XML
-      raise "Excel SpreadsheetML support has been extracted to roo-xls. Install roo-xls to use Roo::Excel2003XML."
+      raise ROO_EXCELML_NOTICE
     when :Google
-      raise "Google support has been extracted to roo-google. Install roo-google to use Roo::Google."
+      raise ROO_GOOGLE_NOTICE
     else
       super
     end

data/lib/roo/base.rb

@@ -401,25 +401,26 @@ class Roo::Base
 
   protected
 
-  def file_type_check(filename, ext, name, warning_level, packed = nil)
+  def file_type_check(filename, exts, name, warning_level, packed = nil)
     if packed == :zip
-      # lalala.ods.zip => lalala.ods
-      # hier wird KEIN unzip gemacht, sondern nur der Name der Datei
-      # getestet, falls es eine gepackte Datei ist.
+      # spreadsheet.ods.zip => spreadsheet.ods
+      # Decompression is not performed here, only the 'zip' extension
+      # is removed from the file.
       filename = File.basename(filename, File.extname(filename))
     end
 
     if uri?(filename) && qs_begin = filename.rindex('?')
       filename = filename[0..qs_begin - 1]
     end
-    if File.extname(filename).downcase != ext
+    exts = Array(exts)
+    if !exts.include?(File.extname(filename).downcase)
       case warning_level
       when :error
-        warn file_type_warning_message(filename, ext)
+        warn file_type_warning_message(filename, exts)
         fail TypeError, "#{filename} is not #{name} file"
       when :warning
         warn "are you sure, this is #{name} spreadsheet file?"
-        warn file_type_warning_message(filename, ext)
+        warn file_type_warning_message(filename, exts)
       when :ignore
         # ignore
       else
@@ -481,10 +482,12 @@ class Roo::Base
     filename
   end
 
-  def file_type_warning_message(filename, ext)
-    "use #{Roo::CLASS_FOR_EXTENSION.fetch(ext.sub('.', '').to_sym)}.new to handle #{ext} spreadsheet files. This has #{File.extname(filename).downcase}"
+  def file_type_warning_message(filename, exts)
+    *rest, last_ext = exts
+    ext_list = rest.any? ? "#{rest.join(', ')} or #{last_ext}" : last_ext
+    "use #{Roo::CLASS_FOR_EXTENSION.fetch(last_ext.sub('.', '').to_sym)}.new to handle #{ext_list} spreadsheet files. This has #{File.extname(filename).downcase}"
   rescue KeyError
-    raise "unknown file type: #{ext}"
+    raise "unknown file types: #{ext_list}"
   end
 
   def find_by_row(row_index)

data/lib/roo/constants.rb

@@ -0,0 +1,5 @@
+module Roo
+  ROO_EXCEL_NOTICE   = "Excel support has been extracted to roo-xls due to its dependency on the GPL'd spreadsheet gem. Install roo-xls to use Roo::Excel.".freeze
+  ROO_EXCELML_NOTICE = "Excel SpreadsheetML support has been extracted to roo-xls. Install roo-xls to use Roo::Excel2003XML.".freeze
+  ROO_GOOGLE_NOTICE = "Google support has been extracted to roo-google. Install roo-google to use Roo::Google.".freeze
+end

data/lib/roo/excelx.rb

@@ -89,7 +89,7 @@ module Roo
       sheet_options[:expand_merged_ranges] = (options[:expand_merged_ranges] || false)
 
       unless is_stream?(filename_or_stream)
-        file_type_check(filename_or_stream, '.xlsx', 'an Excel-xlsx', file_warning, packed)
+        file_type_check(filename_or_stream, %w[.xlsx .xlsm], 'an Excel 2007', file_warning, packed)
         basename = File.basename(filename_or_stream)
       end
 
@@ -321,7 +321,12 @@ module Roo
     # Yield an array of Excelx::Cell
     # Takes options for sheet, pad_cells, and max_rows
     def each_row_streaming(options = {})
-      sheet_for(options.delete(:sheet)).each_row(options) { |row| yield row }
+      sheet = sheet_for(options.delete(:sheet))
+      if block_given?
+        sheet.each_row(options) { |row| yield row }
+      else
+        sheet.to_enum(:each_row, options)
+      end
     end
 
     private
@@ -409,19 +414,13 @@ module Roo
       @sheet_files = []
 
       unless is_stream?(zipfilename_or_stream)
-        process_zipfile_entries Zip::File.open(zipfilename_or_stream).to_a.sort_by(&:name)
+        zip_file = Zip::File.open(zipfilename_or_stream)
       else
-        stream = Zip::InputStream.open zipfilename_or_stream
-        begin
-          entries = []
-          while (entry = stream.get_next_entry)
-            entries << entry
-          end
-          process_zipfile_entries entries
-        ensure
-          stream.close
-        end
+        zip_file = Zip::CentralDirectory.new
+        zip_file.read_from_stream zipfilename_or_stream
       end
+
+      process_zipfile_entries zip_file.to_a.sort_by(&:name)
     end
 
     def process_zipfile_entries(entries)

data/lib/roo/libre_office.rb

@@ -1,5 +1,4 @@
 require 'roo/open_office'
 
 # LibreOffice is just an alias for Roo::OpenOffice class
-class Roo::LibreOffice < Roo::OpenOffice
-end
+Roo::LibreOffice = Roo::OpenOffice

data/lib/roo/open_office.rb

@@ -4,318 +4,306 @@ require 'cgi'
 require 'zip/filesystem'
 require 'roo/font'
 
-class Roo::OpenOffice < Roo::Base
-  # initialization and opening of a spreadsheet file
-  # values for packed: :zip
-  def initialize(filename, options={})
-    packed = options[:packed]
-    file_warning = options[:file_warning] || :error
-
-    @only_visible_sheets = options[:only_visible_sheets]
-    file_type_check(filename,'.ods','an Roo::OpenOffice', file_warning, packed)
-    @tmpdir = make_tmpdir(File.basename(filename), options[:tmpdir_root])
-    @filename = local_filename(filename, @tmpdir, packed)
-    #TODO: @cells_read[:default] = false
-    Zip::File.open(@filename) do |zip_file|
-      if content_entry = zip_file.glob("content.xml").first
-        roo_content_xml_path = File.join(@tmpdir, 'roo_content.xml')
+module Roo
+  class OpenOffice < Roo::Base
+    ERROR_MISSING_CONTENT_XML = 'file missing required content.xml'.freeze
+    XPATH_FIND_TABLE_STYLES   = "//*[local-name()='automatic-styles']".freeze
+    XPATH_LOCAL_NAME_TABLE    = "//*[local-name()='table']".freeze
+
+    # initialization and opening of a spreadsheet file
+    # values for packed: :zip
+    def initialize(filename, options = {})
+      packed       = options[:packed]
+      file_warning = options[:file_warning] || :error
+
+      @only_visible_sheets = options[:only_visible_sheets]
+      file_type_check(filename, '.ods', 'an Roo::OpenOffice', file_warning, packed)
+      @tmpdir   = make_tmpdir(File.basename(filename), options[:tmpdir_root])
+      @filename = local_filename(filename, @tmpdir, packed)
+      # TODO: @cells_read[:default] = false
+      open_oo_file(options)
+      super(filename, options)
+      initialize_default_variables
+    rescue => e # clean up any temp files, but only if an error was raised
+      close
+      raise e
+    end
+
+    def open_oo_file(options)
+      Zip::File.open(@filename) do |zip_file|
+        content_entry = zip_file.glob('content.xml').first
+        fail ArgumentError, ERROR_MISSING_CONTENT_XML unless content_entry
+
+        roo_content_xml_path = ::File.join(@tmpdir, 'roo_content.xml')
         content_entry.extract(roo_content_xml_path)
-        decrypt_if_necessary(
-          zip_file,
-          content_entry,
-          roo_content_xml_path,
-          options
-        )
-      else
-        raise ArgumentError, 'file missing required content.xml'
+        decrypt_if_necessary(zip_file, content_entry, roo_content_xml_path, options)
       end
     end
-    super(filename, options)
-    @formula = Hash.new
-    @style = Hash.new
-    @style_defaults = Hash.new { |h,k| h[k] = [] }
-    @table_display = Hash.new { |h,k| h[k] = true }
-    @font_style_definitions = Hash.new
-    @comment = Hash.new
-    @comments_read = Hash.new
-  rescue => e # clean up any temp files, but only if an error was raised
-    close
-    raise e
-  end
 
-  def method_missing(m,*args)
-    read_labels
-    # is method name a label name
-    if @label.has_key?(m.to_s)
-      row,col = label(m.to_s)
-      cell(row,col)
-    else
-      # call super for methods like #a1
-      super
+    def initialize_default_variables
+      @formula                = {}
+      @style                  = {}
+      @style_defaults         = Hash.new { |h, k| h[k] = [] }
+      @table_display          = Hash.new { |h, k| h[k] = true }
+      @font_style_definitions = {}
+      @comment                = {}
+      @comments_read          = {}
     end
-  end
 
-  # Returns the content of a spreadsheet-cell.
-  # (1,1) is the upper left corner.
-  # (1,1), (1,'A'), ('A',1), ('a',1) all refers to the
-  # cell at the first line and first row.
-  def cell(row, col, sheet=nil)
-    sheet ||= default_sheet
-    read_cells(sheet)
-    row,col = normalize(row,col)
-    if celltype(row,col,sheet) == :date
-      yyyy,mm,dd = @cell[sheet][[row,col]].to_s.split('-')
-      return Date.new(yyyy.to_i,mm.to_i,dd.to_i)
-    end
-    @cell[sheet][[row,col]]
-  end
+    def method_missing(m, *args)
+      read_labels
+      # is method name a label name
+      if @label.key?(m.to_s)
+        row, col = label(m.to_s)
+        cell(row, col)
+      else
+        # call super for methods like #a1
+        super
+      end
+    end
 
-  # Returns the formula at (row,col).
-  # Returns nil if there is no formula.
-  # The method #formula? checks if there is a formula.
-  def formula(row,col,sheet=nil)
-    sheet ||= default_sheet
-    read_cells(sheet)
-    row,col = normalize(row,col)
-    @formula[sheet][[row,col]]
-  end
+    # Returns the content of a spreadsheet-cell.
+    # (1,1) is the upper left corner.
+    # (1,1), (1,'A'), ('A',1), ('a',1) all refers to the
+    # cell at the first line and first row.
+    def cell(row, col, sheet = nil)
+      sheet ||= default_sheet
+      read_cells(sheet)
+      row, col = normalize(row, col)
+      if celltype(row, col, sheet) == :date
+        yyyy, mm, dd = @cell[sheet][[row, col]].to_s.split('-')
+        return Date.new(yyyy.to_i, mm.to_i, dd.to_i)
+      end
 
-  # Predicate methods really should return a boolean
-  # value. Hopefully no one was relying on the fact that this
-  # previously returned either nil/formula
-  def formula?(*args)
-    !!formula(*args)
-  end
+      @cell[sheet][[row, col]]
+    end
+
+    # Returns the formula at (row,col).
+    # Returns nil if there is no formula.
+    # The method #formula? checks if there is a formula.
+    def formula(row, col, sheet = nil)
+      sheet ||= default_sheet
+      read_cells(sheet)
+      row, col = normalize(row, col)
+      @formula[sheet][[row, col]]
+    end
 
-  # returns each formula in the selected sheet as an array of elements
-  # [row, col, formula]
-  def formulas(sheet=nil)
-    sheet ||= default_sheet
-    read_cells(sheet)
-    if @formula[sheet]
+    # Predicate methods really should return a boolean
+    # value. Hopefully no one was relying on the fact that this
+    # previously returned either nil/formula
+    def formula?(*args)
+      !!formula(*args)
+    end
+
+    # returns each formula in the selected sheet as an array of elements
+    # [row, col, formula]
+    def formulas(sheet = nil)
+      sheet ||= default_sheet
+      read_cells(sheet)
+      return [] unless @formula[sheet]
       @formula[sheet].each.collect do |elem|
         [elem[0][0], elem[0][1], elem[1]]
       end
-    else
-      []
     end
-  end
 
-  # Given a cell, return the cell's style
-  def font(row, col, sheet=nil)
-    sheet ||= default_sheet
-    read_cells(sheet)
-    row,col = normalize(row,col)
-    style_name = @style[sheet][[row,col]] || @style_defaults[sheet][col - 1] || 'Default'
-    @font_style_definitions[style_name]
-  end
+    # Given a cell, return the cell's style
+    def font(row, col, sheet = nil)
+      sheet ||= default_sheet
+      read_cells(sheet)
+      row, col   = normalize(row, col)
+      style_name = @style[sheet][[row, col]] || @style_defaults[sheet][col - 1] || 'Default'
+      @font_style_definitions[style_name]
+    end
 
-  # returns the type of a cell:
-  # * :float
-  # * :string
-  # * :date
-  # * :percentage
-  # * :formula
-  # * :time
-  # * :datetime
-  def celltype(row,col,sheet=nil)
-    sheet ||= default_sheet
-    read_cells(sheet)
-    row,col = normalize(row,col)
-    if @formula[sheet][[row,col]]
-      return :formula
-    else
-      @cell_type[sheet][[row,col]]
+    # returns the type of a cell:
+    # * :float
+    # * :string
+    # * :date
+    # * :percentage
+    # * :formula
+    # * :time
+    # * :datetime
+    def celltype(row, col, sheet = nil)
+      sheet ||= default_sheet
+      read_cells(sheet)
+      row, col = normalize(row, col)
+      @formula[sheet][[row, col]] ? :formula : @cell_type[sheet][[row, col]]
     end
-  end
 
-  def sheets
-    unless @table_display.any?
-      doc.xpath("//*[local-name()='automatic-styles']").each do |style|
-        read_table_styles(style)
+    def sheets
+      unless @table_display.any?
+        doc.xpath(XPATH_FIND_TABLE_STYLES).each do |style|
+          read_table_styles(style)
+        end
       end
+      doc.xpath(XPATH_LOCAL_NAME_TABLE).map do |sheet|
+        if !@only_visible_sheets || @table_display[attribute(sheet, 'style-name')]
+          sheet.attributes['name'].value
+        end
+      end.compact
     end
-    doc.xpath("//*[local-name()='table']").map do |sheet|
-      if !@only_visible_sheets || @table_display[attr(sheet,'style-name')]
-        sheet.attributes["name"].value
-      end
-    end.compact
-  end
 
-  # version of the Roo::OpenOffice document
-  # at 2007 this is always "1.0"
-  def officeversion
-    oo_version
-    @officeversion
-  end
-
-  # shows the internal representation of all cells
-  # mainly for debugging purposes
-  def to_s(sheet=nil)
-    sheet ||= default_sheet
-    read_cells(sheet)
-    @cell[sheet].inspect
-  end
+    # version of the Roo::OpenOffice document
+    # at 2007 this is always "1.0"
+    def officeversion
+      oo_version
+      @officeversion
+    end
 
-  # returns the row,col values of the labelled cell
-  # (nil,nil) if label is not defined
-  def label(labelname)
-    read_labels
-    unless @label.size > 0
-      return nil,nil,nil
+    # shows the internal representation of all cells
+    # mainly for debugging purposes
+    def to_s(sheet = nil)
+      sheet ||= default_sheet
+      read_cells(sheet)
+      @cell[sheet].inspect
     end
-    if @label.has_key? labelname
-      return @label[labelname][1].to_i,
+
+    # returns the row,col values of the labelled cell
+    # (nil,nil) if label is not defined
+    def label(labelname)
+      read_labels
+      return [nil, nil, nil] if @label.size < 1 || !@label.key?(labelname)
+      [
+        @label[labelname][1].to_i,
         ::Roo::Utils.letter_to_number(@label[labelname][2]),
         @label[labelname][0]
-    else
-      return nil,nil,nil
+      ]
     end
-  end
 
-  # Returns an array which all labels. Each element is an array with
-  # [labelname, [row,col,sheetname]]
-  def labels(sheet=nil)
-    read_labels
-    @label.map do |label|
-      [ label[0], # name
-        [ label[1][1].to_i, # row
+    # Returns an array which all labels. Each element is an array with
+    # [labelname, [row,col,sheetname]]
+    def labels(_sheet = nil)
+      read_labels
+      @label.map do |label|
+        [label[0], # name
+         [label[1][1].to_i, # row
           ::Roo::Utils.letter_to_number(label[1][2]), # column
           label[1][0], # sheet
-        ] ]
+         ]]
+      end
     end
-  end
 
-  # returns the comment at (row/col)
-  # nil if there is no comment
-  def comment(row,col,sheet=nil)
-    sheet ||= default_sheet
-    read_cells(sheet)
-    row,col = normalize(row,col)
-    return nil unless @comment[sheet]
-    @comment[sheet][[row,col]]
-  end
+    # returns the comment at (row/col)
+    # nil if there is no comment
+    def comment(row, col, sheet = nil)
+      sheet ||= default_sheet
+      read_cells(sheet)
+      row, col = normalize(row, col)
+      return nil unless @comment[sheet]
+      @comment[sheet][[row, col]]
+    end
 
-  # returns each comment in the selected sheet as an array of elements
-  # [row, col, comment]
-  def comments(sheet=nil)
-    sheet ||= default_sheet
-    read_comments(sheet) unless @comments_read[sheet]
-    if @comment[sheet]
+    # returns each comment in the selected sheet as an array of elements
+    # [row, col, comment]
+    def comments(sheet = nil)
+      sheet ||= default_sheet
+      read_comments(sheet) unless @comments_read[sheet]
+      return [] unless @comment[sheet]
       @comment[sheet].each.collect do |elem|
-        [elem[0][0],elem[0][1],elem[1]]
+        [elem[0][0], elem[0][1], elem[1]]
       end
-    else
-      []
     end
-  end
 
-  private
-
-  # If the ODS file has an encryption-data element, then try to decrypt.
-  # If successful, the temporary content.xml will be overwritten with
-  # decrypted contents.
-  def decrypt_if_necessary(
-    zip_file,
-    content_entry,
-    roo_content_xml_path, options
-  )
-    # Check if content.xml is encrypted by extracting manifest.xml
-    # and searching for a manifest:encryption-data element
-
-    if manifest_entry = zip_file.glob("META-INF/manifest.xml").first
-      roo_manifest_xml_path = File.join(@tmpdir, "roo_manifest.xml")
-      manifest_entry.extract(roo_manifest_xml_path)
-      manifest = ::Roo::Utils.load_xml(roo_manifest_xml_path)
-
-      # XPath search for manifest:encryption-data only for the content.xml
-      # file
-
-      encryption_data = manifest.xpath(
-        "//manifest:file-entry[@manifest:full-path='content.xml']"\
+    private
+
+    # If the ODS file has an encryption-data element, then try to decrypt.
+    # If successful, the temporary content.xml will be overwritten with
+    # decrypted contents.
+    def decrypt_if_necessary(
+      zip_file,
+      content_entry,
+      roo_content_xml_path, options
+    )
+      # Check if content.xml is encrypted by extracting manifest.xml
+      # and searching for a manifest:encryption-data element
+
+      if (manifest_entry = zip_file.glob('META-INF/manifest.xml').first)
+        roo_manifest_xml_path = File.join(@tmpdir, 'roo_manifest.xml')
+        manifest_entry.extract(roo_manifest_xml_path)
+        manifest        = ::Roo::Utils.load_xml(roo_manifest_xml_path)
+
+        # XPath search for manifest:encryption-data only for the content.xml
+        # file
+
+        encryption_data = manifest.xpath(
+          "//manifest:file-entry[@manifest:full-path='content.xml']"\
         "/manifest:encryption-data"
-      ).first
+        ).first
 
-      # If XPath returns a node, then we know content.xml is encrypted
+        # If XPath returns a node, then we know content.xml is encrypted
 
-      if !encryption_data.nil?
+        unless encryption_data.nil?
 
-        # Since we know it's encrypted, we check for the password option
-        # and if it doesn't exist, raise an argument error
+          # Since we know it's encrypted, we check for the password option
+          # and if it doesn't exist, raise an argument error
 
-        password = options[:password]
-        if !password.nil?
-          perform_decryption(
-            encryption_data,
-            password,
-            content_entry,
-            roo_content_xml_path
-          )
-        else
-          raise ArgumentError,
-            'file is encrypted but password was not supplied'
+          password = options[:password]
+          if !password.nil?
+            perform_decryption(
+              encryption_data,
+              password,
+              content_entry,
+              roo_content_xml_path
+            )
+          else
+            fail ArgumentError, 'file is encrypted but password was not supplied'
+          end
         end
+      else
+        fail ArgumentError, 'file missing required META-INF/manifest.xml'
       end
-    else
-      raise ArgumentError, 'file missing required META-INF/manifest.xml'
     end
-  end
 
-  # Process the ODS encryption manifest and perform the decryption
-  def perform_decryption(
-    encryption_data,
-    password,
-    content_entry,
-    roo_content_xml_path
-  )
-    # Extract various expected attributes from the manifest that
-    # describe the encryption
-
-    algorithm_node = encryption_data.xpath("manifest:algorithm").first
-    key_derivation_node =
-      encryption_data.xpath("manifest:key-derivation").first
-    start_key_generation_node =
-      encryption_data.xpath("manifest:start-key-generation").first
-
-    # If we have all the expected elements, then we can perform
-    # the decryption.
-
-    if !algorithm_node.nil? && !key_derivation_node.nil? &&
-        !start_key_generation_node.nil?
-
-      # The algorithm is a URI describing the algorithm used
-      algorithm = algorithm_node['manifest:algorithm-name']
-
-      # The initialization vector is base-64 encoded
-      iv = Base64.decode64(
-        algorithm_node['manifest:initialisation-vector']
-      )
-      key_derivation_name =
-        key_derivation_node['manifest:key-derivation-name']
-      key_size = key_derivation_node['manifest:key-size'].to_i
-      iteration_count =
-        key_derivation_node['manifest:iteration-count'].to_i
-      salt = Base64.decode64(key_derivation_node['manifest:salt'])
-
-      # The key is hashed with an algorithm represented by this URI
-      key_generation_name =
-        start_key_generation_node[
-          'manifest:start-key-generation-name'
-      ]
-        key_generation_size =
-          start_key_generation_node['manifest:key-size'].to_i
+    # Process the ODS encryption manifest and perform the decryption
+    def perform_decryption(
+      encryption_data,
+      password,
+      content_entry,
+      roo_content_xml_path
+    )
+      # Extract various expected attributes from the manifest that
+      # describe the encryption
+
+      algorithm_node            = encryption_data.xpath('manifest:algorithm').first
+      key_derivation_node       =
+        encryption_data.xpath('manifest:key-derivation').first
+      start_key_generation_node =
+        encryption_data.xpath('manifest:start-key-generation').first
+
+      # If we have all the expected elements, then we can perform
+      # the decryption.
+
+      if !algorithm_node.nil? && !key_derivation_node.nil? &&
+         !start_key_generation_node.nil?
+
+        # The algorithm is a URI describing the algorithm used
+        algorithm           = algorithm_node['manifest:algorithm-name']
+
+        # The initialization vector is base-64 encoded
+        iv                  = Base64.decode64(
+          algorithm_node['manifest:initialisation-vector']
+        )
+        key_derivation_name = key_derivation_node['manifest:key-derivation-name']
+        key_size            = key_derivation_node['manifest:key-size'].to_i
+        iteration_count     = key_derivation_node['manifest:iteration-count'].to_i
+        salt                = Base64.decode64(key_derivation_node['manifest:salt'])
+
+        # The key is hashed with an algorithm represented by this URI
+        key_generation_name =
+          start_key_generation_node[
+            'manifest:start-key-generation-name'
+          ]
+        key_generation_size = start_key_generation_node['manifest:key-size'].to_i
 
         hashed_password = password
-        key = nil
+        key             = nil
 
-        if key_generation_name.eql?(
-        "http://www.w3.org/2000/09/xmldsig#sha256"
-      )
-        hashed_password = Digest::SHA256.digest(password)
-      else
-        raise ArgumentError, 'Unknown key generation algorithm ' +
-          key_generation_name
-      end
+        if key_generation_name == 'http://www.w3.org/2000/09/xmldsig#sha256'
+
+          hashed_password = Digest::SHA256.digest(password)
+        else
+          fail ArgumentError, "Unknown key generation algorithm #{key_generation_name}"
+        end
 
         cipher = find_cipher(
           algorithm,
@@ -336,210 +324,177 @@ class Roo::OpenOffice < Roo::Base
             Zlib::Inflate.new(-Zlib::MAX_WBITS).inflate(decrypted)
           )
         rescue StandardError => error
-          raise ArgumentError,
-            'Invalid password or other data error: ' + error.to_s
+          raise ArgumentError, "Invalid password or other data error: #{error}"
         end
-    else
-      raise ArgumentError,
-        'manifest.xml missing encryption-data elements'
+      else
+        fail ArgumentError, 'manifest.xml missing encryption-data elements'
+      end
     end
-  end
 
-  # Create a cipher based on an ODS algorithm URI from manifest.xml
-  def find_cipher(
-    algorithm,
-    key_derivation_name,
-    hashed_password,
-    salt,
-    iteration_count,
-    iv
-  )
-    cipher = nil
-    if algorithm.eql? "http://www.w3.org/2001/04/xmlenc#aes256-cbc"
+    # Create a cipher based on an ODS algorithm URI from manifest.xml
+    # params: algorithm, key_derivation_name, hashed_password, salt, iteration_count, iv
+    def find_cipher(*args)
+      fail ArgumentError, 'Unknown algorithm ' + algorithm unless args[0] == 'http://www.w3.org/2001/04/xmlenc#aes256-cbc'
+
       cipher = OpenSSL::Cipher.new('AES-256-CBC')
       cipher.decrypt
       cipher.padding = 0
-      cipher.key = find_cipher_key(
-        cipher,
-        key_derivation_name,
-        hashed_password,
-        salt,
-        iteration_count
-      )
-      cipher.iv = iv
-    else
-      raise ArgumentError, 'Unknown algorithm ' + algorithm
-    end
-    cipher
-  end
+      cipher.key     = find_cipher_key(cipher, *args[1..4])
+      cipher.iv      = args[5]
 
-  # Create a cipher key based on an ODS algorithm string from manifest.xml
-  def find_cipher_key(
-    cipher,
-    key_derivation_name,
-    hashed_password,
-    salt,
-    iteration_count
-  )
-    if key_derivation_name.eql? "PBKDF2"
-      key = OpenSSL::PKCS5.pbkdf2_hmac_sha1(
-        hashed_password,
-        salt,
-        iteration_count,
-        cipher.key_len
-      )
-    else
-      raise ArgumentError, 'Unknown key derivation name ' +
-        key_derivation_name
-    end
-    key
-  end
+      cipher
+    end
 
-  # Block decrypt raw bytes from the zip file based on the cipher
-  def decrypt(content_entry, cipher)
-    # Zip::Entry.extract writes a 0-length file when trying
-    # to extract an encrypted stream, so we read the
-    # raw bytes based on the offset and lengths
-    decrypted = ""
-    File.open(@filename, "rb") do |zipfile|
-      zipfile.seek(
-        content_entry.local_header_offset +
-        content_entry.calculate_local_header_size
-      )
-      total_to_read = content_entry.compressed_size
-
-      block_size = 4096
-      block_size = total_to_read if block_size > total_to_read
-
-      while buffer = zipfile.read(block_size)
-        decrypted += cipher.update(buffer)
-        total_to_read -= buffer.length
-
-        break if total_to_read == 0
-
-        block_size = total_to_read  if block_size > total_to_read
-      end
+    # Create a cipher key based on an ODS algorithm string from manifest.xml
+    def find_cipher_key(*args)
+      fail ArgumentError, 'Unknown key derivation name ', args[1] unless args[1] == 'PBKDF2'
+
+      OpenSSL::PKCS5.pbkdf2_hmac_sha1(args[2], args[3], args[4], args[0].key_len)
     end
 
-    decrypted + cipher.final
-  end
+    # Block decrypt raw bytes from the zip file based on the cipher
+    def decrypt(content_entry, cipher)
+      # Zip::Entry.extract writes a 0-length file when trying
+      # to extract an encrypted stream, so we read the
+      # raw bytes based on the offset and lengths
+      decrypted = ''
+      File.open(@filename, 'rb') do |zipfile|
+        zipfile.seek(
+          content_entry.local_header_offset +
+            content_entry.calculate_local_header_size
+        )
+        total_to_read = content_entry.compressed_size
 
-  def doc
-    @doc ||= ::Roo::Utils.load_xml(File.join(@tmpdir, "roo_content.xml"))
-  end
+        block_size = 4096
+        block_size = total_to_read if block_size > total_to_read
+
+        while (buffer = zipfile.read(block_size))
+          decrypted     += cipher.update(buffer)
+          total_to_read -= buffer.length
 
-  # read the version of the OO-Version
-  def oo_version
-    doc.xpath("//*[local-name()='document-content']").each do |office|
-      @officeversion = attr(office,'version')
+          break if total_to_read == 0
+
+          block_size = total_to_read if block_size > total_to_read
+        end
+      end
+
+      decrypted + cipher.final
+    end
+
+    def doc
+      @doc ||= ::Roo::Utils.load_xml(File.join(@tmpdir, 'roo_content.xml'))
+    end
+
+    # read the version of the OO-Version
+    def oo_version
+      doc.xpath("//*[local-name()='document-content']").each do |office|
+        @officeversion = attribute(office, 'version')
+      end
     end
-  end
 
-  # helper function to set the internal representation of cells
-  def set_cell_values(sheet,x,y,i,v,value_type,formula,table_cell,str_v,style_name)
-    key = [y,x+i]
-    @cell_type[sheet] = {} unless @cell_type[sheet]
-    @cell_type[sheet][key] = Roo::OpenOffice.oo_type_2_roo_type(value_type)
-    @formula[sheet] = {} unless @formula[sheet]
-    if formula
-      ['of:', 'oooc:'].each do |prefix|
-        if formula[0,prefix.length] == prefix
-          formula = formula[prefix.length..-1]
+    # helper function to set the internal representation of cells
+    def set_cell_values(sheet, x, y, i, v, value_type, formula, table_cell, str_v, style_name)
+      key = [y, x + i]
+      @cell_type[sheet] ||= {}
+      @cell_type[sheet][key] = value_type.to_sym if value_type
+      @formula[sheet] ||= {}
+      if formula
+        ['of:', 'oooc:'].each do |prefix|
+          if formula[0, prefix.length] == prefix
+            formula = formula[prefix.length..-1]
+          end
         end
+        @formula[sheet][key] = formula
       end
-      @formula[sheet][key] = formula
-    end
-    @cell[sheet] = {} unless @cell[sheet]
-    @style[sheet] = {} unless @style[sheet]
-    @style[sheet][key] = style_name
-    case @cell_type[sheet][key]
-    when :float
-      @cell[sheet][key] = v.to_f
-    when :string
-      @cell[sheet][key] = str_v
-    when :date
-      #TODO: if table_cell.attributes['date-value'].size != "XXXX-XX-XX".size
-      if attr(table_cell,'date-value').size != "XXXX-XX-XX".size
-        #-- dann ist noch eine Uhrzeit vorhanden
-        #-- "1961-11-21T12:17:18"
-        @cell[sheet][key] = DateTime.parse(attr(table_cell,'date-value').to_s)
-        @cell_type[sheet][key] = :datetime
+      @cell[sheet] ||= {}
+      @style[sheet] ||= {}
+      @style[sheet][key] = style_name
+      case @cell_type[sheet][key]
+      when :float, :percentage
+        @cell[sheet][key] = v.to_f
+      when :string
+        @cell[sheet][key] = str_v
+      when :date
+        # TODO: if table_cell.attributes['date-value'].size != "XXXX-XX-XX".size
+        if attribute(table_cell, 'date-value').size != 'XXXX-XX-XX'.size
+          #-- dann ist noch eine Uhrzeit vorhanden
+          #-- "1961-11-21T12:17:18"
+          @cell[sheet][key]      = DateTime.parse(attribute(table_cell, 'date-value').to_s)
+          @cell_type[sheet][key] = :datetime
+        else
+          @cell[sheet][key] = table_cell.attributes['date-value']
+        end
+      when :time
+        hms = v.split(':')
+        @cell[sheet][key] = hms[0].to_i * 3600 + hms[1].to_i * 60 + hms[2].to_i
       else
-        @cell[sheet][key] = table_cell.attributes['date-value']
+        @cell[sheet][key] = v
       end
-    when :percentage
-      @cell[sheet][key] = v.to_f
-    when :time
-      hms = v.split(':')
-      @cell[sheet][key] = hms[0].to_i*3600 + hms[1].to_i*60 + hms[2].to_i
-    else
-      @cell[sheet][key] = v
     end
-  end
 
-  # read all cells in the selected sheet
-  #--
-  # the following construct means '4 blanks'
-  # some content <text:s text:c="3"/>
-  #++
-  def read_cells(sheet = default_sheet)
-    validate_sheet!(sheet)
-    return if @cells_read[sheet]
-
-    sheet_found = false
-    doc.xpath("//*[local-name()='table']").each do |ws|
-      if sheet == attr(ws,'name')
+    # read all cells in the selected sheet
+    #--
+    # the following construct means '4 blanks'
+    # some content <text:s text:c="3"/>
+    #++
+    def read_cells(sheet = default_sheet)
+      validate_sheet!(sheet)
+      return if @cells_read[sheet]
+
+      sheet_found = false
+      doc.xpath("//*[local-name()='table']").each do |ws|
+        next unless sheet == attribute(ws, 'name')
+
         sheet_found = true
-        col = 1
-        row = 1
+        col         = 1
+        row         = 1
         ws.children.each do |table_element|
           case table_element.name
           when 'table-column'
             @style_defaults[sheet] << table_element.attributes['default-cell-style-name']
           when 'table-row'
             if table_element.attributes['number-rows-repeated']
-              skip_row = attr(table_element,'number-rows-repeated').to_s.to_i
-              row = row + skip_row - 1
+              skip_row = attribute(table_element, 'number-rows-repeated').to_s.to_i
+              row      = row + skip_row - 1
             end
             table_element.children.each do |cell|
-              skip_col = attr(cell, 'number-columns-repeated')
-              formula = attr(cell,'formula')
-              value_type = attr(cell,'value-type')
-              v =  attr(cell,'value')
-              style_name = attr(cell,'style-name')
+              skip_col   = attribute(cell, 'number-columns-repeated')
+              formula    = attribute(cell, 'formula')
+              value_type = attribute(cell, 'value-type')
+              v          = attribute(cell, 'value')
+              style_name = attribute(cell, 'style-name')
               case value_type
               when 'string'
-                str_v  = ''
+                str_v      = ''
                 # insert \n if there is more than one paragraph
                 para_count = 0
                 cell.children.each do |str|
                   # begin comments
-=begin
-- <table:table-cell office:value-type="string">
-  - <office:annotation office:display="true" draw:style-name="gr1" draw:text-style-name="P1" svg:width="1.1413in" svg:height="0.3902in" svg:x="2.0142in" svg:y="0in" draw:caption-point-x="-0.2402in" draw:caption-point-y="0.5661in">
-      <dc:date>2011-09-20T00:00:00</dc:date>
-      <text:p text:style-name="P1">Kommentar fuer B4</text:p>
-    </office:annotation>
-    <text:p>B4 (mit Kommentar)</text:p>
-  </table:table-cell>
-=end
+                  #=begin
+                  #- <table:table-cell office:value-type="string">
+                  #  - <office:annotation office:display="true" draw:style-name="gr1" draw:text-style-name="P1" svg:width="1.1413in" svg:height="0.3902in" svg:x="2.0142in" svg:y="0in" draw:caption-point-x="-0.2402in" draw:caption-point-y="0.5661in">
+                  #      <dc:date>2011-09-20T00:00:00</dc:date>
+                  #      <text:p text:style-name="P1">Kommentar fuer B4</text:p>
+                  #    </office:annotation>
+                  #    <text:p>B4 (mit Kommentar)</text:p>
+                  #  </table:table-cell>
+                  #=end
                   if str.name == 'annotation'
                     str.children.each do |annotation|
-                      if annotation.name == 'p'
-                        # @comment ist ein Hash mit Sheet als Key (wie bei @cell)
-                        # innerhalb eines Elements besteht ein Eintrag aus einem
-                        # weiteren Hash mit Key [row,col] und dem eigentlichen
-                        # Kommentartext als Inhalt
-                        @comment[sheet] = Hash.new unless @comment[sheet]
-                        key = [row,col]
-                        @comment[sheet][key] = annotation.text
-                      end
+                      next unless annotation.name == 'p'
+                      # @comment ist ein Hash mit Sheet als Key (wie bei @cell)
+                      # innerhalb eines Elements besteht ein Eintrag aus einem
+                      # weiteren Hash mit Key [row,col] und dem eigentlichen
+                      # Kommentartext als Inhalt
+                      @comment[sheet]      = Hash.new unless @comment[sheet]
+                      key                  = [row, col]
+                      @comment[sheet][key] = annotation.text
                     end
                   end
                   # end comments
                   if str.name == 'p'
-                    v = str.content
-                    str_v += "\n" if para_count > 0
+                    v          = str.content
+                    str_v      += "\n" if para_count > 0
                     para_count += 1
                     if str.children.size > 1
                       str_v += children_to_string(str.children)
@@ -548,32 +503,28 @@ class Roo::OpenOffice < Roo::Base
                         str_v += child.content #.text
                       end
                     end
-                    str_v.gsub!(/&apos;/,"'")  # special case not supported by unescapeHTML
+                    str_v.gsub!(/&apos;/, "'") # special case not supported by unescapeHTML
                     str_v = CGI.unescapeHTML(str_v)
                   end # == 'p'
                 end
               when 'time'
                 cell.children.each do |str|
-                  if str.name == 'p'
-                    v = str.content
-                  end
+                  v = str.content if str.name == 'p'
                 end
               when '', nil, 'date', 'percentage', 'float'
                 #
               when 'boolean'
-                v = attr(cell,'boolean-value').to_s
-              else
-                # raise "unknown type #{value_type}"
+                v = attribute(cell, 'boolean-value').to_s
               end
               if skip_col
-                if v != nil or cell.attributes['date-value']
-                  0.upto(skip_col.to_i-1) do |i|
-                    set_cell_values(sheet,col,row,i,v,value_type,formula,cell,str_v,style_name)
+                if !v.nil? || cell.attributes['date-value']
+                  0.upto(skip_col.to_i - 1) do |i|
+                    set_cell_values(sheet, col, row, i, v, value_type, formula, cell, str_v, style_name)
                   end
                 end
                 col += (skip_col.to_i - 1)
               end # if skip
-              set_cell_values(sheet,col,row,0,v,value_type,formula,cell,str_v,style_name)
+              set_cell_values(sheet, col, row, 0, v, value_type, formula, cell, str_v, style_name)
               col += 1
             end
             row += 1
@@ -581,101 +532,83 @@ class Roo::OpenOffice < Roo::Base
           end
         end
       end
+      doc.xpath("//*[local-name()='automatic-styles']").each do |style|
+        read_styles(style)
+      end
+
+      fail RangeError unless sheet_found
+
+      @cells_read[sheet]    = true
+      @comments_read[sheet] = true
     end
-    doc.xpath("//*[local-name()='automatic-styles']").each do |style|
-      read_styles(style)
-    end
-    if !sheet_found
-      raise RangeError
-    end
-    @cells_read[sheet] = true
-    @comments_read[sheet] = true
-  end
 
-  # Only calls read_cells because Roo::Base calls read_comments
-  # whereas the reading of comments is done in read_cells for Roo::OpenOffice-objects
-  def read_comments(sheet=nil)
-    read_cells(sheet)
-  end
+    # Only calls read_cells because Roo::Base calls read_comments
+    # whereas the reading of comments is done in read_cells for Roo::OpenOffice-objects
+    def read_comments(sheet = nil)
+      read_cells(sheet)
+    end
 
-  def read_labels
-    @label ||= Hash[doc.xpath("//table:named-range").map do |ne|
-      #-
-      # $Sheet1.$C$5
-      #+
-      name = attr(ne,'name').to_s
-      sheetname,coords = attr(ne,'cell-range-address').to_s.split('.$')
-      col, row = coords.split('$')
-      sheetname = sheetname[1..-1] if sheetname[0,1] == '$'
-      [name, [sheetname,row,col]]
-    end]
-  end
+    def read_labels
+      @label ||= Hash[doc.xpath('//table:named-range').map do |ne|
+        #-
+        # $Sheet1.$C$5
+        #+
+        name              = attribute(ne, 'name').to_s
+        sheetname, coords = attribute(ne, 'cell-range-address').to_s.split('.$')
+        col, row          = coords.split('$')
+        sheetname         = sheetname[1..-1] if sheetname[0, 1] == '$'
+        [name, [sheetname, row, col]]
+      end]
+    end
 
-  def read_styles(style_elements)
-    @font_style_definitions['Default'] = Roo::Font.new
-    style_elements.each do |style|
-      next unless style.name == 'style'
-      style_name = attr(style,'name')
-      style.each do |properties|
-        font = Roo::OpenOffice::Font.new
-        font.bold = attr(properties,'font-weight')
-        font.italic = attr(properties,'font-style')
-        font.underline = attr(properties,'text-underline-style')
-        @font_style_definitions[style_name] = font
+    def read_styles(style_elements)
+      @font_style_definitions['Default'] = Roo::Font.new
+      style_elements.each do |style|
+        next unless style.name == 'style'
+        style_name = attribute(style, 'name')
+        style.each do |properties|
+          font                                = Roo::OpenOffice::Font.new
+          font.bold                           = attribute(properties, 'font-weight')
+          font.italic                         = attribute(properties, 'font-style')
+          font.underline                      = attribute(properties, 'text-underline-style')
+          @font_style_definitions[style_name] = font
+        end
       end
     end
-  end
 
-  def read_table_styles(styles)
-    styles.children.each do |style|
-      next unless style.name == 'style'
-      style_name = attr(style,'name')
-      style.children.each do |properties|
-        display = attr(properties,'display')
-        next unless display
-        @table_display[style_name] = (display == 'true')
+    def read_table_styles(styles)
+      styles.children.each do |style|
+        next unless style.name == 'style'
+        style_name = attribute(style, 'name')
+        style.children.each do |properties|
+          display = attribute(properties, 'display')
+          next unless display
+          @table_display[style_name] = (display == 'true')
+        end
       end
     end
-  end
-
-  A_ROO_TYPE = {
-    "float"      => :float,
-    "string"     => :string,
-    "date"       => :date,
-    "percentage" => :percentage,
-    "time"       => :time,
-  }
-
-  def self.oo_type_2_roo_type(ootype)
-    return A_ROO_TYPE[ootype]
-  end
 
-  # helper method to convert compressed spaces and other elements within
-  # an text into a string
-  def children_to_string(children)
-    result = ''
-    children.each {|child|
-      if child.text?
-        result = result + child.content
-      else
-        if child.name == 's'
-          compressed_spaces = child.attributes['c'].to_s.to_i
-          # no explicit number means a count of 1:
-          if compressed_spaces == 0
-            compressed_spaces = 1
-          end
-          result = result + " "*compressed_spaces
+    # helper method to convert compressed spaces and other elements within
+    # an text into a string
+    # FIXME: add a test for compressed_spaces == 0. It's not currently tested.
+    def children_to_string(children)
+      children.map do |child|
+        if child.text?
+          child.content
         else
-          result = result + child.content
+          if child.name == 's'
+            compressed_spaces = child.attributes['c'].to_s.to_i
+            # no explicit number means a count of 1:
+            compressed_spaces == 0 ? ' ' : ' ' * compressed_spaces
+          else
+            child.content
+          end
         end
-      end
-    }
-    result
-  end
+      end.join
+    end
 
-  def attr(node, attr_name)
-    if node.attributes[attr_name]
-      node.attributes[attr_name].value
+    def attribute(node, attr_name)
+      node.attributes[attr_name].value if node.attributes[attr_name]
     end
   end
 end