roo 1.1.0 → 1.2.0

data/History.txt

@@ -1,8 +1,17 @@
+== 1.2.0 2008-08-24
+* 3 major enhancements
+  * Excelx: improved the detection of cell type and conversion into roo types
+  * All: to_csv: changed boundaries from first_row,1..last_row,last_column to 1,1..last_row,last_column
+  * All: Environment variable "ROO_TMP" indicate where temporary directories will be created (if not set the default is the current working directory)
+* 2 bugfixes
+  * Excel: improved the detection of last_row/last_column (parseexcel-gem bug?)
+  * Excel/Excelx/Openoffice: temporary directories were not removed at opening a file of the wrong type
 == 1.1.0 2008-07-26
 * 2 major enhancements
   * Excel: speed improvements
-  * Changed the behaviour of reading files with the false type
-* 2 bugfixex
+  * Changed the behavior of reading files with the wrong type
+* 3 bugfixes
+  * Google: added normalize in set_value method
   * Excel: last_row in Excel class did not work properly under some circumstances
   * all: fixed a bug in #to_xml if there is an empty sheet
 == 1.0.2 2008-07-04

data/README.txt

@@ -7,22 +7,36 @@ Installation:
 
 Usage:
 
+    require 'rubygems'
     require 'roo'
 
-    sob = Openoffice("myspreadsheet.ods")              # creates an Openoffice Spreadsheet instance
-    sob = Excel("myspreadsheet.xls")                   # creates an Excel Spreadsheet instance
-    sob = Google("myspreadsheetkey_at_google")         # creates an Google Spreadsheet instance
+    s = Openoffice.new("myspreadsheet.ods")      # creates an Openoffice Spreadsheet instance
+    s = Excel.new("myspreadsheet.xls")           # creates an Excel Spreadsheet instance
+    s = Google.new("myspreadsheetkey_at_google") # creates an Google Spreadsheet instance
+    s = Excelx.new("myspreadsheet.xlsx")         # creates an Excel Spreadsheet instance for Excel .xlsx files
 
-    sob.default_sheet = sob.sheets.first               # first sheet in the spreadsheet file will be used
-    sob.cell(1,1)                                      # returns the content of the first row/first cell in the sheet
-    sob.cell('A',1)                                    # same cell
+    s.default_sheet = s.sheets.first  # first sheet in the spreadsheet file will be used
+
+    # s.sheet is an array which holds the names of the sheets within
+    # a spreadsheet.
+    # you can also write
+    # s.default_sheet = s.sheets[3] or
+    # s.default_sheet = 'Sheet 3'
+
+    s.cell(1,1)                                 # returns the content of the first row/first cell in the sheet
+    s.cell('A',1)                               # same cell
+    s.cell(1,'A')                               # same cell
+    s.cell(1,'A',s.sheets[0])                   # same cell
+
+    # almost all methods have an optional argument 'sheet'.
+    # If this parameter is ommitted, the default_sheet will be used.
     
-    sob.info                                           # prints infos about the spreadsheet file
+    s.info                                      # prints infos about the spreadsheet file
 
-    sob.first_row                                      # the number of the first row
-    sob.last_row                                       # the number of the last row
-    sob.first_column                                   # the number of the first column
-    sob.last_column                                    # the number of the last column
+    s.first_row                                 # the number of the first row
+    s.last_row                                  # the number of the last row
+    s.first_column                              # the number of the first column
+    s.last_column                               # the number of the last column
 
 
 see http://roo.rubyforge.org for a more complete tutorial

data/lib/roo/excel.rb

@@ -24,18 +24,15 @@ class Excel < GenericSpreadsheet
     super()
     @file_warning = file_warning
     @tmpdir = "oo_"+$$.to_s
+    @tmpdir = File.join(ENV['ROO_TMP'], @tmpdir) if ENV['ROO_TMP'] 
     unless File.exists?(@tmpdir)
       FileUtils::mkdir(@tmpdir)
     end
     filename = open_from_uri(filename) if filename[0,7] == "http://"
     filename = unzip(filename) if packed and packed == :zip
-    #if File.extname(filename).downcase != ".xls"
-    #  warn "are you sure, this is an excel file?"
-    #end
-    #@file_warning = :error
-    file_type_check(filename,'.xls','an Excel')
-    @filename = filename
     begin
+      file_type_check(filename,'.xls','an Excel')
+      @filename = filename
       unless File.file?(@filename)
         raise IOError, "file #{@filename} does not exist"
       end
@@ -45,9 +42,8 @@ class Excel < GenericSpreadsheet
       if self.sheets.size == 1
         @default_sheet = self.sheets.first
       end
-      # @first_row = @last_row = @first_column = @last_column = nil
-      #if ENV["roo_local"] != "thomas-p"
     ensure
+      #if ENV["roo_local"] != "thomas-p"
       FileUtils::rm_r(@tmpdir)
       #end
     end
@@ -82,19 +78,6 @@ class Excel < GenericSpreadsheet
     return result
   end
 
-  #TODO: testing only
-  def inject_null_characters(str)
-    if str.class != String
-      return str
-    end
-    new_str=''
-    0.upto(str.size-1) do |i|
-      new_str += str[i,1]
-      new_str += "\000"
-    end
-    new_str
-  end
-
   # returns the content of a cell. The upper left corner is (1,1) or ('A',1)
   def cell(row,col,sheet=nil)
     sheet = @default_sheet unless sheet
@@ -175,7 +158,7 @@ class Excel < GenericSpreadsheet
   private
 
   # determine the first and last boundaries
-  def get_firsts_lasts(sheet=nil)
+  def get_firsts_lasts_parseexcel(sheet=nil)
     sheet = @default_sheet unless sheet
     fr = fc = 999_999
     lr = lc = -999_999
@@ -212,6 +195,46 @@ class Excel < GenericSpreadsheet
       end
     end
     # 2007-11-05 END
+    
+    if lc
+      letzte_spalte_leer = true
+      until ! letzte_spalte_leer
+        worksheet.each(0) {|reihe|
+          if reihe
+            cell = reihe.at(lc-1)
+            if cell
+              case cell.type
+              when :numeric, :date
+                letzte_spalte_leer = false
+              when :text
+                letzte_spalte_leer = false if cell.to_s != "" 
+              end
+            end
+          end
+        }  
+        lc -= 1 if letzte_spalte_leer
+        #puts "letzte Spalte auf #{lc} verringert" if letzte_spalte_leer
+      end    
+    end
+
+    return fr, lr, fc, lc
+  end
+
+  # determine the first and last boundaries
+  def get_firsts_lasts(sheet=nil)
+    fr,lr,fc,lc = get_firsts_lasts_parseexcel(sheet)
+    if lr 
+      # 2007-11-05 BEGIN
+      # parsexcel liefert (mir unverstaendlich) eine Zeile als letzte Zeile
+      # zurueck, die aber leer ist. Deshalb Korrekturfunktion, die wirklich
+      # die letzte nicht leere Zeile liefert.
+      # 2008-07-23 meine Loesung funtionierte auch noch nicht unter allen
+      # Umstaenden row() == nil ergaenzt
+      while row(lr,sheet) == nil || empty_row?(row(lr,sheet))
+        lr -= 1
+      end
+    end
+    # 2007-11-05 END
     @first_row[sheet]    = fr
     @last_row[sheet]     = lr
     @first_column[sheet] = fc
@@ -243,6 +266,16 @@ class Excel < GenericSpreadsheet
     ! content
   end
 
+  def empty_column?(col)
+    content = false
+    col.compact.each {|elem|
+      if elem != ''
+        content = true
+      end
+    }
+    ! content
+  end
+
   def platform_specific_iconv(value)
     case RUBY_PLATFORM.downcase
     when /darwin/
@@ -360,4 +393,18 @@ class Excel < GenericSpreadsheet
     @cells_read[sheet] = true
   end
 
+    #TODO: testing only
+#  def inject_null_characters(str)
+#    if str.class != String
+#      return str
+#    end
+#    new_str=''
+#    0.upto(str.size-1) do |i|
+#      new_str += str[i,1]
+#      new_str += "\000"
+#    end
+#    new_str
+#  end
+#
+
 end

data/lib/roo/excelx.rb

@@ -13,6 +13,70 @@ class  String
 end
 
 class Excelx < GenericSpreadsheet
+  FORMATS = {
+    'General' => :float,
+    '0' => :float,
+    '0.00' => :float,
+    '#,##0' => :float,
+    '#,##0.00' => :float,
+    '0%' => :percentage,
+    '0.00%' => :percentage,
+    '0.00E+00' => :float,
+    '# ?/?' => :float, #??? TODO:
+    '# ??/??' => :float, #??? TODO:
+    'mm-dd-yy' => :date,
+    'd-mmm-yy' => :date,
+    'd-mmm' => :date,
+    'mmm-yy' => :date,
+    'h:mm AM/PM' => :date,
+    'h:mm:ss AM/PM' => :date,
+    'h:mm' => :time,
+    'h:mm:ss' => :time,
+    'm/d/yy h:mm' => :date,
+    '#,##0 ;(#,##0)' => :float,
+    '#,##0 ;[Red](#,##0)' => :float,
+    '#,##0.00;(#,##0.00)' => :float,
+    '#,##0.00;[Red](#,##0.00)' => :float,
+    'mm:ss' => :time,
+    '[h]:mm:ss' => :time,
+    'mmss.0' => :time,
+    '##0.0E+0' => :float,
+    '@' => :float,
+    #-- zusaetzliche Formate, die nicht standardmaessig definiert sind:
+    "yyyy\\-mm\\-dd" => :date,
+    'dd/mm/yy' => :date,
+    'hh:mm:ss' => :time,
+  }
+  STANDARD_FORMATS = { 
+    0 => 'General',
+    1 => '0',
+    2 => '0.00',
+    3 => '#,##0',
+    4 => '#,##0.00',
+    9 => '0%',
+    10 => '0.00%',
+    11 => '0.00E+00',
+    12 => '# ?/?',
+    13 => '# ??/??',
+    14 => 'mm-dd-yy',
+    15 => 'd-mmm-yy',
+    16 => 'd-mmm',
+    17 => 'mmm-yy',
+    18 => 'h:mm AM/PM',
+    19 => 'h:mm:ss AM/PM',
+    20 => 'h:mm',
+    21 => 'h:mm:ss',
+    22 => 'm/d/yy h:mm',
+    37 => '#,##0 ;(#,##0)',
+    38 => '#,##0 ;[Red](#,##0)',
+    39 => '#,##0.00;(#,##0.00)',
+    40 => '#,##0.00;[Red](#,##0.00)',
+    45 => 'mm:ss',
+    46 => '[h]:mm:ss',
+    47 => 'mmss.0',
+    48 => '##0.0E+0',
+    49 => '@',
+  }
   @@nr = 0
 
   # initialization and opening of a spreadsheet file
@@ -21,19 +85,16 @@ class Excelx < GenericSpreadsheet
     super()
     @file_warning = file_warning
     @tmpdir = "oo_"+$$.to_s
+    @tmpdir = File.join(ENV['ROO_TMP'], @tmpdir) if ENV['ROO_TMP'] 
     unless File.exists?(@tmpdir)
       FileUtils::mkdir(@tmpdir)
     end
     filename = open_from_uri(filename) if filename[0,7] == "http://"
     filename = unzip(filename) if packed and packed == :zip
-    #if File.extname(filename).downcase != ".xlsx"
-    #  warn "are you sure, this is an Excel-xlsx file?"
-    #end
-    #@file_warning = :error
-    file_type_check(filename,'.xlsx','an Excel-xlsx')
-    @cells_read = Hash.new
-    @filename = filename
     begin
+      file_type_check(filename,'.xlsx','an Excel-xlsx')
+      @cells_read = Hash.new
+      @filename = filename
       unless File.file?(@filename)
         raise IOError, "file #{@filename} does not exist"
       end
@@ -50,6 +111,13 @@ class Excelx < GenericSpreadsheet
         file.close
         read_shared_strings(@sharedstring_doc)
       end
+      @styles_table = []
+      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
+        file.close
+        read_styles(@styles_doc)
+      end
       @sheet_doc = []
       @sheet_files.each_with_index do |item, i|
         file = File.new(item)
@@ -75,6 +143,8 @@ class Excelx < GenericSpreadsheet
     @last_column = Hash.new
     @header_line = 1
     @excelx_type = Hash.new
+    @excelx_value = Hash.new
+    @s_attribute = Hash.new # TODO: ggf. wieder entfernen nur lokal benoetigt
   end
 
   # Returns the content of a spreadsheet-cell.
@@ -151,14 +221,34 @@ class Excelx < GenericSpreadsheet
     end
   end
 
-  # returns the internal type of a excel cell
+  # returns the internal type of an excel cell
   # * :numeric_or_formula
   # * :string  
+  # Note: this is only available within the Excelx class 
   def excelx_type(row,col,sheet=nil)
-   sheet = @default_sheet unless sheet
+    sheet = @default_sheet unless sheet
+    read_cells(sheet) unless @cells_read[sheet]
+    row,col = normalize(row,col)
+    return @excelx_type[sheet]["#{row},#{col}"]
+  end
+  
+  # returns the internal value of an excelx cell
+  # Note: this is only available within the Excelx class 
+  def excelx_value(row,col,sheet=nil)
+    sheet = @default_sheet unless sheet
+    read_cells(sheet) unless @cells_read[sheet]
+    row,col = normalize(row,col)
+    return @excelx_value[sheet]["#{row},#{col}"]
+  end
+  
+  # returns the internal format of an excel cell
+  def excelx_format(row,col,sheet=nil)
+    sheet = @default_sheet unless sheet
     read_cells(sheet) unless @cells_read[sheet]
     row,col = normalize(row,col)
-    :numeric_or_formula
+    s = @s_attribute[sheet]["#{row},#{col}"]
+    result = attribute2format(s)
+    result
   end
   
   # returns an array of sheet names in the spreadsheet
@@ -204,7 +294,10 @@ class Excelx < GenericSpreadsheet
   private
 
   # helper function to set the internal representation of cells
-  def set_cell_values(sheet,x,y,i,v,vt,formula,tr,str_v,excelx_type=nil)
+  def set_cell_values(sheet,x,y,i,v,vt,formula,tr,str_v,
+      excelx_type=nil,
+      excelx_value=nil,
+      s_attribute=nil)
     key = "#{y},#{x+i}"
     @cell_type[sheet] = {} unless @cell_type[sheet]
     @cell_type[sheet][key] = vt
@@ -229,8 +322,13 @@ class Excelx < GenericSpreadsheet
     end
     @excelx_type[sheet] = {} unless @excelx_type[sheet]
     @excelx_type[sheet][key] = excelx_type
+    @excelx_value[sheet] = {} unless @excelx_value[sheet]
+    @excelx_value[sheet][key] = excelx_value
+    @s_attribute[sheet] = {} unless @s_attribute[sheet]
+    @s_attribute[sheet][key] = s_attribute
   end
 
+  # splits a coordinate like "AA12" into the parts "AA" (String) and 12 (Fixnum)
   def split_coord(s)
     letter = ""
     number = 0
@@ -257,89 +355,20 @@ class Excelx < GenericSpreadsheet
   end
 
   # read all cells in the selected sheet
-  def ALTER_ANSATZ_read_cells(sheet=nil)
-    sheet = @default_sheet unless sheet
-    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
-    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'
-                  if row.attributes['t'] == 's'
-                    tmp_type = :shared
-                  end
-                  if row.attributes['s'] == '2' #or
-                     #row.attributes['s'] == '1' # 2008-07-26
-                    tmp_type = :time
-                  elsif row.attributes['s'] == '1' and row.attributes['t'] == nil # and ergaenzt 2008-07-03
-                    tmp_type = :formula
-                  end
-                  formula = nil
-                  f_element_found = false
-                  row.each_element do |cell|
-                    if cell.name == 'f'
-                      f_element_found = true
-                      formula = cell.text
-                    end
-                    if cell.name == 'v'
-                      if tmp_type == :formula and f_element_found == false
-                        #if cell.text.to_f < 1.0 # 2008-07-26
-                        tmp_type = :time
-                        #else # 2008-07-26
-                        #tmp_type = :date #2008-07-26
-                        #end #2008-07-26
-                      end
-                      if tmp_type == :time #2008-07-26
-                        if cell.text.to_f >= 1.0 # 2008-07-26
-                          tmp_type = :date # 2008-07-26
-                        end # 2008-07-26
-                      end # 2008-07-26
-                      if tmp_type == :shared
-                        vt = :string
-                        str_v = @shared_table[cell.text.to_i]
-                      elsif tmp_type == :date
-                        vt = :date
-                        v = cell.text
-                      elsif tmp_type == :time
-                        vt = :time
-                        v = cell.text
-                      elsif tmp_type == :formula
-                        vt = :formula
-                        v = cell.text.to_f #TODO: !!!!
-                      else
-                        vt = :float
-                        v = cell.text
-                      end
-                      x,y = split_coordinate(row.attributes['r'])
-                      tr=nil #TODO: ???s
-                      set_cell_values(sheet,x,y,0,v,vt,formula,tr,str_v)
-                    end
-                  end
-                end
-              end
-            end
-          end
-        end
-      end
-    end
-    sheet_found = true #TODO:
-    if !sheet_found
-      raise RangeError
+  def format2type(format)
+    if FORMATS.has_key? format
+      FORMATS[format]
+    else
+      :float
     end
-    @cells_read[sheet] = true
   end
 
-# read all cells in the selected sheet
+  # read all cells in the selected sheet
   def read_cells(sheet=nil)
     sheet = @default_sheet unless sheet
     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
+    raise RangeError unless self.sheets.include? sheet
     n = self.sheets.index(sheet)
     @sheet_doc[n].each_element do |worksheet|
       worksheet.each_element do |elem|
@@ -350,34 +379,24 @@ class Excelx < GenericSpreadsheet
                 if row.name == 'c'
                   if row.attributes['t'] == 's'
                     tmp_type = :shared
-                  end
-                  if row.attributes['s'] == '2' #or
-                     #row.attributes['s'] == '1' # 2008-07-26
-                    tmp_type = :time
-                  elsif row.attributes['s'] == '1' and row.attributes['t'] == nil # and ergaenzt 2008-07-03
-                    tmp_type = :formula
+                  else
+                    s_attribute = row.attributes['s']
+                    format = attribute2format(s_attribute)
+                    tmp_type = format2type(format)
                   end
                   formula = nil
-                  f_element_found = false
                   row.each_element do |cell|
                     if cell.name == 'f'
-                      f_element_found = true
                       formula = cell.text
                     end
                     if cell.name == 'v'
-                      if tmp_type == :formula and f_element_found == false
-                        #if cell.text.to_f < 1.0 # 2008-07-26
-                        tmp_type = :time
-                        #else # 2008-07-26
-                        #tmp_type = :date #2008-07-26
-                        #end #2008-07-26
-                      end
                       if tmp_type == :time #2008-07-26
                         if cell.text.to_f >= 1.0 # 2008-07-26
                           tmp_type = :date # 2008-07-26
                         end # 2008-07-26
                       end # 2008-07-26
-                      excelx_type = :numeric_or_formula
+                      excelx_type = [:numeric_or_formula,format]
+                      excelx_value = cell.text
                       if tmp_type == :shared
                         vt = :string
                         str_v = @shared_table[cell.text.to_i]
@@ -397,7 +416,7 @@ class Excelx < GenericSpreadsheet
                       end
                       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)
+                      set_cell_values(sheet,x,y,0,v,vt,formula,tr,str_v,excelx_type,excelx_value,s_attribute)
                     end
                   end
                 end
@@ -413,7 +432,7 @@ class Excelx < GenericSpreadsheet
     end
     @cells_read[sheet] = true
   end
-
+  
   # Checks if the default_sheet exists. If not an RangeError exception is
   # raised
   def check_default_sheet
@@ -435,7 +454,7 @@ class Excelx < GenericSpreadsheet
     end
   end
 
-
+  # extracts all needed files from the zip file
   def process_zipfile(zipfilename, zip, path='')
     @sheet_files = []
     Zip::ZipFile.open(zipfilename) {|zf|
@@ -451,9 +470,13 @@ class Excelx < GenericSpreadsheet
             f << zip.read(entry)
           }
         end
+        if entry.to_s.end_with?('styles.xml')
+          open(@tmpdir+'/'+@file_nr.to_s+'_roo_styles.xml','wb') {|f|
+            f << zip.read(entry)
+          }
+        end
         if entry.to_s =~ /sheet([0-9]+).xml$/
           nr = $1
-          #  entry.extract("xaxa_sheet#{nr}.xml")
           open(@tmpdir+'/'+@file_nr.to_s+"_roo_sheet#{nr}",'wb') {|f|
             f << zip.read(entry)
           }
@@ -464,22 +487,26 @@ class Excelx < GenericSpreadsheet
     return
   end
 
+  # extract files from the zip file
   def extract_content(zipfilename)
     Zip::ZipFile.open(@filename) do |zip|
       process_zipfile(zipfilename,zip)
     end
   end
 
+  # sets the value of a cell
   def set_value(row,col,value,sheet=nil)
     sheet = @default_value unless sheet
     @cell[sheet]["#{row},#{col}"] = value
   end
 
+  # sets the type of a cell
   def set_type(row,col,type,sheet=nil)
     sheet = @default_value unless sheet
     @cell_type[sheet]["#{row},#{col}"] = type
   end
 
+  # read the shared strings xml document
   def read_shared_strings(doc)
     doc.each_element do |sst|
       if sst.name == 'sst'
@@ -496,4 +523,51 @@ class Excelx < GenericSpreadsheet
     end
   end
 
+  # read the styles elements of an excelx document
+  def read_styles(doc)
+    @numFmts = []
+    @cellXfs = []
+    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 == "cellXfs"
+            e2.each_element do |e3|
+              if e3.name == 'xf'
+                numFmtId = e3.attributes['numFmtId'] 
+                # p numFmtId
+                @cellXfs << [numFmtId]
+              end
+            end
+          end
+        end
+      end
+    end
+  end
+
+  # convert internal excelx attribute to a format
+  def attribute2format(s)
+    result = nil
+    @numFmts.each {|nf|
+      if nf.first == @cellXfs[s.to_i].first
+        result = nf[1]
+        break
+      end
+    }
+    unless result
+      id = @cellXfs[s.to_i].first.to_i
+      if STANDARD_FORMATS.has_key? id
+        result = STANDARD_FORMATS[id]
+      end
+    end
+    result
+  end
+
 end # class