hmcgowan-roo 1.2.4 → 1.3.0

data/{README.txt → README.markdown}

@@ -1,11 +1,19 @@
-README for roo
-==============
+# README for Roo
 
-Installation:
+This is the semi-official roo repository. I've been unable to contact the maintainer so this
+repository should allow us to continue in the interim. If you'd like to contribute I'm happy 
+to pull your changes in and will be making periodic releases from here until we can get
+the gems on RubyForge. 
 
+
+## Installation
+
+    # Run the following if you haven't done so before:
+    gem sources -a http://gems.github.com
+    # Install the gem:
     sudo gem install roo
 
-Usage:
+## Usage:
 
     require 'rubygems'
     require 'roo'

data/lib/roo/excel.rb

@@ -1,8 +1,12 @@
 require 'rubygems'
 gem 'spreadsheet', '>= 0.6.3.1'
 require 'spreadsheet'
-CHARGUESS = false
-require 'charguess' if CHARGUESS
+CHARGUESS = begin
+  require 'charguess'
+  true
+rescue LoadError => e
+  false
+end
 
 # ruby-spreadsheet has a font object so we're extending it 
 # with our own functionality but still providing full access
@@ -79,18 +83,7 @@ class Excel < GenericSpreadsheet
   def sheets
     result = []
     @workbook.worksheets.each do |worksheet| 
-      # TODO: is there a better way to do conversion?
-      if CHARGUESS
-        encoding = CharGuess::guess(worksheet.name)
-        encoding = 'unicode' unless encoding
-
-
-        result << Iconv.new('utf-8',encoding).iconv(
-          worksheet.name
-        )
-      else
-        result << platform_specific_iconv(worksheet.name)
-      end
+      result << normalize_string(worksheet.name)
     end
     return result
   end
@@ -243,11 +236,7 @@ class Excel < GenericSpreadsheet
     return name-1 if name.kind_of?(Fixnum)
     i = 0
     @workbook.worksheets.each do |worksheet|
-      # TODO: is there a better way to do conversion?
-      return i if name == platform_specific_iconv(worksheet.name)
-      #Iconv.new('utf-8','unicode').iconv(
-      #        @workbook.worksheet(i).name
-      #      )
+      return i if name == normalize_string(worksheet.name)
       i += 1
     end
     raise StandardError, "sheet '#{name}' not found"
@@ -272,7 +261,16 @@ class Excel < GenericSpreadsheet
     }
     ! content
   end
-
+  
+  def normalize_string(value)
+    value = every_second_null?(value) ? remove_every_second_null(value) : value
+    if CHARGUESS && encoding = CharGuess::guess(value)
+      Iconv.new('utf-8', encoding)
+    else
+      platform_specific_iconv(value)
+    end
+  end
+  
   def platform_specific_iconv(value)
     case RUBY_PLATFORM.downcase
     when /darwin/
@@ -381,7 +379,7 @@ class Excel < GenericSpreadsheet
     format = row.format(idx)
     if format.date_or_time?
       cell = row.at(idx)
-      cell.to_s.to_f > 0 ? true : false # cell value must be numeric
+      true if Float(cell) > 0 rescue false
     else
       false
     end  

data/lib/roo/excelx.rb

@@ -1,6 +1,7 @@
 
 require 'rubygems'
-require 'rexml/document'
+gem 'libxml-ruby', '>= 0.8.3'
+require 'xml'
 require 'fileutils'
 require 'zip/zipfilesystem'
 require 'date'
@@ -102,12 +103,12 @@ class Excelx < GenericSpreadsheet
       @file_nr = @@nr
       extract_content(@filename)
       file = File.new(File.join(@tmpdir, @file_nr.to_s+"_roo_workbook.xml"))
-      @workbook_doc = REXML::Document.new file
+      @workbook_doc = XML::Parser.io(file).parse
       file.close
       @shared_table = []
       if File.exist?(File.join(@tmpdir, @file_nr.to_s+'_roo_sharedStrings.xml'))
         file = File.new(File.join(@tmpdir, @file_nr.to_s+'_roo_sharedStrings.xml'))
-        @sharedstring_doc = REXML::Document.new file
+        @sharedstring_doc = XML::Parser.io(file).parse
         file.close
         read_shared_strings(@sharedstring_doc)
       end
@@ -115,14 +116,14 @@ class Excelx < GenericSpreadsheet
       @style_definitions = Array.new { |h,k| h[k] = {} }
       if File.exist?(File.join(@tmpdir, @file_nr.to_s+'_roo_styles.xml'))
         file = File.new(File.join(@tmpdir, @file_nr.to_s+'_roo_styles.xml'))
-        @styles_doc = REXML::Document.new file
+        @styles_doc = XML::Parser.io(file).parse
         file.close
         read_styles(@styles_doc)
       end
       @sheet_doc = []
       @sheet_files.each_with_index do |item, i|
         file = File.new(item)
-        @sheet_doc[i] = REXML::Document.new file
+        @sheet_doc[i] = XML::Parser.io(file).parse
         file.close
       end
     ensure
@@ -288,18 +289,11 @@ class Excelx < GenericSpreadsheet
   # returns an array of sheet names in the spreadsheet
   def sheets
     return_sheets = []
-    @workbook_doc.each_element do |workbook|
-      workbook.each_element do |el|
-        if el.name == "sheets"
-          el.each_element do |sheet|
-            return_sheets << sheet.attributes['name']
-          end
-        end
-      end
+    @workbook_doc.find("//*[local-name()='sheet']").each do |sheet|
+      return_sheets << sheet.attributes.to_h['name']
     end
     return_sheets
   end
-
   # shows the internal representation of all cells
   # for debugging purposes
   def to_s(sheet=nil)
@@ -404,87 +398,62 @@ class Excelx < GenericSpreadsheet
     raise ArgumentError, "Error: sheet '#{sheet||'nil'}' not valid" if @default_sheet == nil and sheet==nil
     raise RangeError unless self.sheets.include? sheet
     n = self.sheets.index(sheet)
-    @sheet_doc[n].each_element do |worksheet|
-      worksheet.each_element do |elem|
-        if elem.name == 'sheetData'
-          elem.each_element do |sheetdata|
-            if sheetdata.name == 'row'
-              sheetdata.each_element do |row|
-                if row.name == 'c'
-                  s_attribute = row.attributes['s']
-                  if row.attributes['t'] == 's'
-                    tmp_type = :shared
-                  elsif row.attributes['t'] == 'b'
-                    tmp_type = :boolean
-                  else
-                    format = attribute2format(s_attribute)
-                    tmp_type = format2type(format)
-                  end
-                  formula = nil
-                  row.each_element do |cell|
-#                    puts "cell.name: #{cell.name}" if cell.text.include? "22606.5120"
-#                    puts "cell.text: #{cell.text}" if cell.text.include? "22606.5120"
-                    if cell.name == 'f'
-                      formula = cell.text
-                    end
-                    if cell.name == 'v'
-                      #puts "tmp_type: #{tmp_type}" if cell.text.include? "22606.5120"
-                      #puts cell.name
-                      if tmp_type == :time or tmp_type == :datetime #2008-07-26
-                        #p cell.text
-                       # p cell.text.to_f if cell.text.include? "22606.5120"
-                        if cell.text.to_f >= 1.0 # 2008-07-26
-                        #  puts ">= 1.0" if cell.text.include? "22606.5120"
-                         # puts "cell.text.to_f: #{cell.text.to_f}" if cell.text.include? "22606.5120"
-                          #puts "cell.text.to_f.floor: #{cell.text.to_f.floor}" if cell.text.include? "22606.5120"
-                          if (cell.text.to_f - cell.text.to_f.floor).abs > 0.000001 #TODO: 
-                           # puts "abs ist groesser"  if cell.text.include? "22606.5120"
-                            # @cell[sheet][key] = DateTime.parse(tr.attributes['date-value'])
-                            tmp_type = :datetime
-                            
-                          else
-                            #puts ":date"
-                            tmp_type = :date # 2008-07-26
-                          end
-                        else
-                          #puts "<1.0"
-                        end # 2008-07-26
-                      end # 2008-07-26
-                      excelx_type = [:numeric_or_formula,format]
-                      excelx_value = cell.text
-                      if tmp_type == :shared
-                        vt = :string
-                        str_v = @shared_table[cell.text.to_i]
-                        excelx_type = :string
-                      elsif tmp_type == :boolean
-                        vt = :boolean
-                        cell.text.to_i == 1 ? v = 'TRUE' : v = 'FALSE'
-                      elsif tmp_type == :date
-                        vt = :date
-                        v = cell.text
-                      elsif tmp_type == :time
-                        vt = :time
-                        v = cell.text
-                      elsif tmp_type == :datetime
-                        vt = :datetime
-                        v = cell.text
-                      elsif tmp_type == :formula
-                        vt = :formula
-                        v = cell.text.to_f #TODO: !!!!
-                      else
-                        vt = :float
-                        v = cell.text
-                      end
-                      #puts "vt: #{vt}" if cell.text.include? "22606.5120"
-                      x,y = split_coordinate(row.attributes['r'])
-                      tr=nil #TODO: ???s
-                      set_cell_values(sheet,x,y,0,v,vt,formula,tr,str_v,excelx_type,excelx_value,s_attribute)
-                    end
-                  end
-                end
+    @sheet_doc[n].find("//*[local-name()='c']").each do |c|
+       s_attribute = c.attributes.to_h['s'].to_i   # should be here
+       if (c.attributes.to_h['t'] == 's')
+         tmp_type = :shared
+       elsif (c.attributes.to_h['t'] == 'b')
+         tmp_type = :boolean
+       else
+       #  s_attribute = c.attributes.to_h['s'].to_i     # was here
+         format = attribute2format(s_attribute)
+         tmp_type = format2type(format)
+       end
+      formula = nil
+      c.each_element do |cell|
+        if cell.name == 'f'
+          formula = cell.content
+        end
+        if cell.name == 'v'
+          if tmp_type == :time or tmp_type == :datetime
+            if cell.content.to_f >= 1.0 
+              if (cell.content.to_f - cell.content.to_f.floor).abs > 0.000001 
+                tmp_type = :datetime 
+              else
+                tmp_type = :date
               end
-            end
+            else
+            end 
+          end
+          excelx_type = [:numeric_or_formula,format]
+          excelx_value = cell.content
+          if tmp_type == :shared
+            vt = :string
+            str_v = @shared_table[cell.content.to_i]
+            excelx_type = :string
+          elsif tmp_type == :boolean
+            vt = :boolean
+            cell.content.to_i == 1 ? v = 'TRUE' : v = 'FALSE'
+          elsif tmp_type == :date
+            vt = :date
+            v = cell.content
+          elsif tmp_type == :time
+            vt = :time
+            v = cell.content
+          elsif tmp_type == :datetime
+            vt = :datetime
+            v = cell.content
+          elsif tmp_type == :formula
+            vt = :formula
+            v = cell.content.to_f #TODO: !!!!
+          else
+            vt = :float
+            v = cell.content
           end
+          #puts "vt: #{vt}" if cell.text.include? "22606.5120"
+          x,y = split_coordinate(c.attributes.to_h['r'])
+          tr=nil #TODO: ???s
+          set_cell_values(sheet,x,y,0,v,vt,formula,tr,str_v,excelx_type,excelx_value,s_attribute)
         end
       end
     end
@@ -500,17 +469,9 @@ class Excelx < GenericSpreadsheet
   def check_default_sheet
     sheet_found = false
     raise ArgumentError, "Error: default_sheet not set" if @default_sheet == nil
-    @workbook_doc.each_element do |workbook|
-      workbook.each_element do |el|
-        if el.name == "sheets"
-          el.each_element do |sheet|
-            if @default_sheet == sheet.attributes['name']
-              sheet_found = true
-            end
-          end
-        end
-      end
-    end
+    
+    sheet_found = true if sheets.include?(@default_sheet)
+    
     if ! sheet_found
       raise RangeError, "sheet '#{@default_sheet}' not found"
     end
@@ -570,18 +531,21 @@ class Excelx < GenericSpreadsheet
 
   # read the shared strings xml document
   def read_shared_strings(doc)
-    doc.each_element do |sst|
-      if sst.name == 'sst'
-        sst.each_element do |si|
-          if si.name == 'si'
-            si.each_element do |elem|
-              if elem.name == 't'
-                @shared_table << elem.text
-              end
+    doc.find("//*[local-name()='si']").each do |si|
+      shared_table_entry = ''
+      si.each_element do |elem|
+        if (elem.name == 'r')
+          elem.each_element do |r_elem|
+            if (r_elem.name == 't')
+              shared_table_entry << r_elem.content
             end
           end
         end
+        if (elem.name == 't')
+          shared_table_entry = elem.content
+        end
       end
+      @shared_table << shared_table_entry
     end
   end
 
@@ -590,47 +554,39 @@ class Excelx < GenericSpreadsheet
     @numFmts = []
     @cellXfs = []
     fonts = []
-    doc.each_element do |e1|
-      if e1.name == "styleSheet"
-        e1.each_element do |e2|
-          if e2.name == "numFmts"
-            e2.each_element do |e3|
-              if e3.name == 'numFmt'
-                numFmtId = e3.attributes['numFmtId']
-                formatCode = e3.attributes['formatCode']
-                @numFmts << [numFmtId, formatCode]
-              end
-            end
-          elsif e2.name == "fonts"  
-            e2.each_element do |e3|
-              if e3.name == 'font'
-                font = Excelx::Font.new
-                e3.each do |e4|
-                  case e4.name
-                  when 'b'
-                    font.bold = true
-                  when 'i'
-                    font.italic = true
-                  when 'u'
-                    font.underline = true
-                  end  
-                end             
-                fonts << font
-              end
-            end
-          elsif e2.name == "cellXfs"
-            e2.each_element do |e3|
-              if e3.name == 'xf'
-                numFmtId = e3.attributes['numFmtId'] 
-                @cellXfs << [numFmtId]
-                fontId = e3.attributes['fontId'].to_i
-                @style_definitions << fonts[fontId]
+    
+    doc.find("//*[local-name()='numFmt']").each do |numFmt|
+      numFmtId = numFmt.attributes.to_h['numFmtId']
+      formatCode = numFmt.attributes.to_h['formatCode']
+      @numFmts << [numFmtId, formatCode]
+    end
+    doc.find("//*[local-name()='fonts']").each do |fonts_el|
+      fonts_el.each_element do |font_el|
+        if font_el.name == 'font'
+          font = Excelx::Font.new
+          font_el.each_element do |font_sub_el|
+            case font_sub_el.name
+              when 'b'
+                font.bold = true
+              when 'i'
+                font.italic = true
+              when 'u'
+                font.underline = true
               end
-            end
           end
+          fonts << font
         end
       end
     end
+    
+    doc.find("//*[local-name()='cellXfs']").each do |xfs|
+        xfs.each do |xf|
+          numFmtId = xf.attributes.to_h['numFmtId']
+          @cellXfs << [numFmtId]
+          fontId = xf.attributes.to_h['fontId'].to_i
+          @style_definitions << fonts[fontId]
+        end
+    end
   end
 
   # convert internal excelx attribute to a format

data/lib/roo/openoffice.rb

@@ -1,6 +1,6 @@
-
 require 'rubygems'
-require 'rexml/document'
+gem 'libxml-ruby', '>= 0.8.3'
+require 'xml'
 require 'fileutils'
 require 'zip/zipfilesystem'
 require 'date'
@@ -38,7 +38,7 @@ class Openoffice < GenericSpreadsheet
       @file_nr = @@nr
       extract_content
       file = File.new(File.join(@tmpdir, @file_nr.to_s+"_roo_content.xml"))
-      @doc = REXML::Document.new file
+      @doc = XML::Parser.io(file).parse
       file.close
     ensure
       #if ENV["roo_local"] != "thomas-p"
@@ -174,31 +174,14 @@ class Openoffice < GenericSpreadsheet
     end
   end
 
-  # returns an array of sheet names in the spreadsheet
   def sheets
     return_sheets = []
-    oo_document_count = 0
-    @doc.each_element do |oo_document|
-      oo_document_count += 1
-      oo_element_count = 0
-      oo_document.each_element do |oo_element|
-        oo_element_count += 1
-        if oo_element.name == "body"
-          oo_element.each_element do |be|
-            if be.name == "spreadsheet"
-              be.each_element do |se|
-                if se.name == "table"
-                  return_sheets << se.attributes['name']
-                end
-              end
-            end
-          end
-        end
-      end
+    @doc.find("//*[local-name()='table']").each do |sheet|
+      return_sheets << sheet.attributes['name']
     end
     return_sheets
   end
-
+    
   # version of the openoffice document
   # at 2007 this is always "1.0"
   def officeversion
@@ -240,14 +223,14 @@ class Openoffice < GenericSpreadsheet
 
   # read the version of the OO-Version
   def oo_version
-    @doc.each_element do |oo_document|
-      @officeversion = oo_document.attributes['version']
+    @doc.find("//*[local-name()='document-content']").each do |office|
+      @officeversion = office.attributes['version']
     end
   end
 
   # helper function to set the internal representation of cells
-  def set_cell_values(sheet,x,y,i,v,vt,formula,tr,str_v,style_name)
-    key = [y,x+i]
+  def set_cell_values(sheet,x,y,i,v,vt,formula,table_cell,str_v,style_name)
+    key = [y,x+i]    
     @cell_type[sheet] = {} unless @cell_type[sheet]
     @cell_type[sheet][key] = Openoffice.oo_type_2_roo_type(vt)
     @formula[sheet] = {} unless @formula[sheet]
@@ -261,13 +244,13 @@ class Openoffice < GenericSpreadsheet
     when :string
       @cell[sheet][key] = str_v
     when :date
-      if tr.attributes['date-value'].size != "XXXX-XX-XX".size
+      if table_cell.attributes['date-value'].size != "XXXX-XX-XX".size
         #-- dann ist noch eine Uhrzeit vorhanden
         #-- "1961-11-21T12:17:18"
-        @cell[sheet][key] = DateTime.parse(tr.attributes['date-value'])
+        @cell[sheet][key] = DateTime.parse(table_cell.attributes['date-value'])
         @cell_type[sheet][key] = :datetime
       else
-        @cell[sheet][key] = tr.attributes['date-value']
+        @cell[sheet][key] = table_cell.attributes['date-value']
       end
     when :percentage
       @cell[sheet][key] = v.to_f
@@ -289,104 +272,88 @@ class Openoffice < GenericSpreadsheet
     sheet_found = false
     raise ArgumentError, "Error: sheet '#{sheet||'nil'}' not valid" if @default_sheet == nil and sheet==nil
     raise RangeError unless self.sheets.include? sheet
-    oo_document_count = 0
-    @doc.each_element do |oo_document|
-      # @officeversion = oo_document.attributes['version']
-      oo_document_count += 1
-      oo_element_count = 0
-      oo_document.each_element do |oo_element|
-        oo_element_count += 1
-        if oo_element.name == "body"
-          oo_element.each_element do |be|
-            if be.name == "spreadsheet"
-              be.each_element do |se|
-                if se.name == "table"
-                  if se.attributes['name']==sheet
-                    sheet_found = true
-                    x=1
-                    y=1
-                    se.each_element do |te|
-                      if te.name == "table-column"
-                        rep = te.attributes["number-columns-repeated"]
-                        @style_defaults[sheet] << te.attributes["default-cell-style-name"] 
-                      elsif te.name == "table-row"
-                        if te.attributes['number-rows-repeated']
-                          skip_y = te.attributes['number-rows-repeated'].to_i
-                          y = y + skip_y - 1 # minus 1 because this line will be counted as a line element
-                        end
-                        te.each_element do |tr|
-                          if tr.name == 'table-cell'
-                            skip = tr.attributes['number-columns-repeated']
-                            formula = tr.attributes['formula']
-                            vt = tr.attributes['value-type']
-                            v  = tr.attributes['value']
-                            style_name = tr.attributes['style-name']
-                            if vt == 'string'
-                              str_v  = ''
-                              # insert \n if there is more than one paragraph
-                              para_count = 0
-                              tr.each_element do |str|
-                                if str.name == 'p'
-                                  v = str.text
-                                  str_v += "\n" if para_count > 0
-                                  para_count += 1
-                                  if str.children.size > 1
-                                    str_v = children_to_string(str.children)
-                                  else
-                                    str.children.each {|child|
-                                      str_v = str_v + child.to_s #.text
-                                    }
-                                  end
-                                  str_v.gsub!(/&apos;/,"'")  # special case not supported by unescapeHTML
-                                  str_v = CGI.unescapeHTML(str_v)
-                                end # == 'p'
-                              end
-                            elsif vt == 'time'
-                              tr.each_element do |str|
-                                if str.name == 'p'
-                                  v = str.text
-                                end
-                              end
-                            elsif vt == '' or vt == nil
-                              #
-                            elsif vt == 'date'
-                              #
-                            elsif vt == 'percentage'
-                              #
-                            elsif vt == 'float'
-                              #
-                            elsif vt == 'boolean'
-                              v = tr.attributes['boolean-value']
-                              #
-                            else
-                              # raise "unknown type #{vt}"
-                            end
-                            if skip
-                              if v != nil or tr.attributes['date-value']
-                                0.upto(skip.to_i-1) do |i|
-                                  set_cell_values(sheet,x,y,i,v,vt,formula,tr,str_v,style_name)
-                                end
-                              end
-                              x += (skip.to_i - 1)
-                            end # if skip
-                            set_cell_values(sheet,x,y,0,v,vt,formula,tr,str_v,style_name)
-                            x += 1
-                          end
-                        end
-                        y += 1
-                        x = 1
+
+    @doc.find("//*[local-name()='table']").each do |ws|
+      if sheet == ws.attributes['name']
+        sheet_found = true
+        col = 1
+        row = 1
+        ws.each_element 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 = table_element.attributes['number-rows-repeated'].to_i
+              row = row + skip_row - 1
+            end
+            table_element.each_element do |cell|
+              skip_col = cell.attributes['number-columns-repeated']
+              formula = cell.attributes['formula']
+              vt = cell.attributes['value-type']
+              v =  cell.attributes['value']
+              style_name = cell.attributes['style-name']
+              if vt == 'string'
+                str_v  = ''
+                # insert \n if there is more than one paragraph
+                para_count = 0
+                cell.each_element do |str|
+                  if str.name == 'p'
+                    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)
+                    else
+                      str.children.each do |child|
+                        str_v += child.content #.text
                       end
                     end
-                  end # sheet
+                    str_v.gsub!(/&apos;/,"'")  # special case not supported by unescapeHTML
+                    str_v = CGI.unescapeHTML(str_v)
+                  end # == 'p'
+                 end
+              elsif vt == 'time'
+                cell.each_element do |str|
+                  if str.name == 'p'
+                    v = str.content
+                  end
                 end
+              elsif vt == '' or vt == nil
+                #
+              elsif vt == 'date'
+                #
+              elsif vt == 'percentage'
+                #
+              elsif vt == 'float'
+                #
+              elsif vt == 'boolean'
+                v = cell.attributes['boolean-value']
+                #
+              else
+                # raise "unknown type #{vt}"
               end
-            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,vt,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,vt,formula,cell,str_v,style_name)
+              col += 1
+            end    
+            row += 1
+            col = 1
           end
-        elsif oo_element.name == "automatic-styles" 
-          read_styles(oo_element)
         end
       end
     end
+    
+    @doc.find("//*[local-name()='automatic-styles']").each do |style|
+      read_styles(style)
+    end
     if !sheet_found
       raise RangeError
     end
@@ -413,25 +380,7 @@ class Openoffice < GenericSpreadsheet
   def check_default_sheet
     sheet_found = false
     raise ArgumentError, "Error: default_sheet not set" if @default_sheet == nil
-    @doc.each_element do |oo_document|
-      oo_element_count = 0
-      oo_document.each_element do |oo_element|
-        oo_element_count += 1
-        if oo_element.name == "body"
-          oo_element.each_element do |be|
-            if be.name == "spreadsheet"
-              be.each_element do |se|
-                if se.name == "table"
-                  if se.attributes['name'] == @default_sheet
-                    sheet_found = true
-                  end # sheet
-                end
-              end
-            end
-          end
-        end
-      end
-    end
+    sheet_found = true if sheets.include?(@default_sheet)
     if ! sheet_found
       raise RangeError, "sheet '#{@default_sheet}' not found"
     end
@@ -487,8 +436,8 @@ class Openoffice < GenericSpreadsheet
   def children_to_string(children)
     result = ''
     children.each {|child|
-      if child.class == REXML::Text
-        result = result + child.to_s
+      if child.text?
+        result = result + child.content
       else
         if child.name == 's'
           compressed_spaces = child.attributes['c'].to_i
@@ -498,7 +447,7 @@ class Openoffice < GenericSpreadsheet
           end
           result = result + " "*compressed_spaces
         else
-          result = result + child.to_s
+          result = result + child.content
         end
       end
     }

data/test/test_roo.rb

@@ -3817,7 +3817,7 @@ Sheet 3:
   def do_test_xml(oo)
     assert_nothing_raised {oo.to_xml}
     sheetname = oo.sheets.first
-    doc = REXML::Document.new(oo.to_xml)
+    doc = XML::Parser.string(oo.to_xml).parse
     doc.root.each_element {|xml_sheet|
       all_cells = init_all_cells(oo, sheetname)
       x = 0
@@ -3831,7 +3831,7 @@ Sheet 3:
         result = [
           cell.attributes['row'],
           cell.attributes['column'],
-          cell.text,
+          cell.content,
           cell.attributes['type'],
         ]
         assert_equal expected, result
@@ -4847,7 +4847,7 @@ This attached file is the newer format of Microsoft Excel (.xlsx).
      end  
   end    
   
-  def test_cell_excel_boolean
+  def test_cell_boolean
     if EXCEL
       oo = Excel.new(File.join(TESTDIR,"boolean.xls"))
       oo.default_sheet = oo.sheets.first
@@ -4868,6 +4868,31 @@ This attached file is the newer format of Microsoft Excel (.xlsx).
     end
   end
   
+   def test_cell_multiline
+     if EXCEL
+        oo = Excel.new(File.join(TESTDIR,"paragraph.xls"))
+        oo.default_sheet = oo.sheets.first
+        assert_equal "This is a test\nof a multiline\nCell", oo.cell(1,1)
+        assert_equal "This is a test\n¶\nof a multiline\n\nCell", oo.cell(1,2)
+        assert_equal "first p\n\nsecond p\n\nlast p", oo.cell(2,1)
+     end  
+     if OPENOFFICE
+       oo = Openoffice.new(File.join(TESTDIR,"paragraph.ods"))
+       oo.default_sheet = oo.sheets.first
+       assert_equal "This is a test\nof a multiline\nCell", oo.cell(1,1)
+       assert_equal "This is a test\n¶\nof a multiline\n\nCell", oo.cell(1,2)
+       assert_equal "first p\n\nsecond p\n\nlast p", oo.cell(2,1)
+     end  
+     if EXCELX
+        oo = Excelx.new(File.join(TESTDIR,"paragraph.xlsx"))
+        oo.default_sheet = oo.sheets.first
+        assert_equal "This is a test\nof a multiline\nCell", oo.cell(1,1)
+        assert_equal "This is a test\n¶\nof a multiline\n\nCell", oo.cell(1,2)
+        assert_equal "first p\n\nsecond p\n\nlast p", oo.cell(2,1)
+     end
+   end
+  
+  
   # for test_cell_styles
   def verify_cell_fonts(oo)
     oo.default_sheet = oo.sheets.first
@@ -4940,7 +4965,21 @@ This attached file is the newer format of Microsoft Excel (.xlsx).
     if EXCEL
       oo = Excel.new(File.join(TESTDIR,"style.xls"))
         verify_cell_fonts(oo)
-      end
     end
+  end
+
+  # If a cell has a date-like string but is preceeded by a ' 
+  # to force that date to be treated like a string, we were getting an exception.
+  # This test just checks for that exception to make sure it's not raised in this cas
+  def test_date_to_float_conversion
+     if EXCEL
+       assert_nothing_raised(NoMethodError) do
+         oo = Excel.new(File.join(TESTDIR,"datetime_floatconv.xls"))
+         oo.default_sheet = oo.sheets.first
+         oo.cell('a',1)
+         oo.cell('a',2)
+       end
+     end
+  end  
   
 end # class

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: hmcgowan-roo
 version: !ruby/object:Gem::Version 
-  version: 1.2.4
+  version: 1.3.0
 platform: ruby
 authors: 
 - Thomas Preymesser
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-03-12 00:00:00 -07:00
+date: 2009-04-20 00:00:00 -07:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -59,7 +59,7 @@ executables: []
 extensions: []
 
 extra_rdoc_files: 
-- README.txt
+- README.markdown
 - History.txt
 files: 
 - lib/roo
@@ -71,74 +71,6 @@ files:
 - lib/roo/roo_rails_helper.rb
 - lib/roo/version.rb
 - lib/roo.rb
-- test/_ods
-- test/_ods/Configurations2
-- test/_ods/Configurations2/accelerator
-- test/_ods/Configurations2/accelerator/current.xml
-- test/_ods/Configurations2/floater
-- test/_ods/Configurations2/images
-- test/_ods/Configurations2/images/Bitmaps
-- test/_ods/Configurations2/menubar
-- test/_ods/Configurations2/popupmenu
-- test/_ods/Configurations2/progressbar
-- test/_ods/Configurations2/statusbar
-- test/_ods/Configurations2/toolbar
-- test/_ods/content.xml
-- test/_ods/META-INF
-- test/_ods/META-INF/manifest.xml
-- test/_ods/meta.xml
-- test/_ods/mimetype
-- test/_ods/settings.xml
-- test/_ods/style.ods
-- test/_ods/styles.xml
-- test/_ods/Thumbnails
-- test/_ods/Thumbnails/thumbnail.png
-- test/_ods_old
-- test/_ods_old/Configurations2
-- test/_ods_old/Configurations2/accelerator
-- test/_ods_old/Configurations2/accelerator/current.xml
-- test/_ods_old/Configurations2/floater
-- test/_ods_old/Configurations2/images
-- test/_ods_old/Configurations2/images/Bitmaps
-- test/_ods_old/Configurations2/menubar
-- test/_ods_old/Configurations2/popupmenu
-- test/_ods_old/Configurations2/progressbar
-- test/_ods_old/Configurations2/statusbar
-- test/_ods_old/Configurations2/toolbar
-- test/_ods_old/content.xml
-- test/_ods_old/META-INF
-- test/_ods_old/META-INF/manifest.xml
-- test/_ods_old/meta.xml
-- test/_ods_old/mimetype
-- test/_ods_old/settings.xml
-- test/_ods_old/style.ods
-- test/_ods_old/styles.xml
-- test/_ods_old/Thumbnails
-- test/_ods_old/Thumbnails/thumbnail.png
-- test/_xlsx
-- test/_xlsx/[Content_Types].xml
-- test/_xlsx/_rels
-- test/_xlsx/docProps
-- test/_xlsx/docProps/app.xml
-- test/_xlsx/docProps/core.xml
-- test/_xlsx/style.xlsx
-- test/_xlsx/style.xlsx.cpgz
-- test/_xlsx/xl
-- test/_xlsx/xl/_rels
-- test/_xlsx/xl/_rels/workbook.xml.rels
-- test/_xlsx/xl/printerSettings
-- test/_xlsx/xl/printerSettings/printerSettings1.bin
-- test/_xlsx/xl/sharedStrings.xml
-- test/_xlsx/xl/styles.xml
-- test/_xlsx/xl/theme
-- test/_xlsx/xl/theme/theme1.xml
-- test/_xlsx/xl/workbook.xml
-- test/_xlsx/xl/worksheets
-- test/_xlsx/xl/worksheets/_rels
-- test/_xlsx/xl/worksheets/_rels/sheet1.xml.rels
-- test/_xlsx/xl/worksheets/sheet1.xml
-- test/_xlsx/xl/worksheets/sheet2.xml
-- test/_xlsx/xl/worksheets/sheet3.xml
 - test/bbu.ods
 - test/bbu.xls
 - test/bbu.xlsx
@@ -159,6 +91,7 @@ files:
 - test/datetime.ods
 - test/datetime.xls
 - test/datetime.xlsx
+- test/datetime_floatconv.xls
 - test/emptysheets.ods
 - test/emptysheets.xls
 - test/false_encoding.xls
@@ -175,6 +108,9 @@ files:
 - test/only_one_sheet.ods
 - test/only_one_sheet.xls
 - test/only_one_sheet.xlsx
+- test/paragraph.ods
+- test/paragraph.xls
+- test/paragraph.xlsx
 - test/ric.ods
 - test/simple_spreadsheet.ods
 - test/simple_spreadsheet.xls
@@ -190,14 +126,14 @@ files:
 - test/time-test.ods
 - test/time-test.xls
 - test/time-test.xlsx
-- README.txt
+- README.markdown
 - History.txt
 has_rdoc: true
 homepage: http://roo.rubyforge.org
 post_install_message: 
 rdoc_options: 
 - --main
-- README.txt
+- README.markdown
 - --inline-source
 - --charset=UTF-8
 require_paths: