roo 1.9.7 → 1.10.0

data/History.txt

@@ -1,7 +1,19 @@
+== 1.10.0 2011-10-10
+
+* 4 enhancements
+  * New class Csv.
+  * Openoffice, Libreoffice: new method 'labels'
+  * Excelx: implemented all methods concerning labels
+  * Openoffice, Excelx: new methods concerning comments (comment, comment? and comments)
+  
+* 2 bugfixes
+  * XLSX: some cells were not recognized correctly from a spreadsheet file from a windows mobile phone.
+  * labels: Moved to a separate methode. There were problems if there was an access to a label before read_cells were called.
+
 == 1.9.7 2011-08-27
 
 * 1 bugfix
-  * Openoffice: Better way for extracting formula strings, some characters were deleted at the formula string under some circumstances.
+  * Openoffice: Better way for extracting formula strings, some characters were deleted at the formula string.
 
 == 1.9.6 2011-08-03
 

data/README.txt

@@ -9,6 +9,8 @@ Roo can access the contents of various spreadsheet files. It can handle
 * Excel
 * Google spreadsheets
 * Excelx
+* Libreoffice
+* CSV
 
 
 == FEATURES/PROBLEMS:

data/bin/roo

@@ -26,8 +26,7 @@ Choice.options do
 		long '--info <spreadsheetfile>'
 		desc 'Show information about a spreadsheet file'
 		action do
-			p Choice.choices
-			puts 'Filename: '+Choice.choices['info']
+			puts '#'
 		end
 	end
 end

data/lib/roo.rb

@@ -3,7 +3,7 @@
 module Roo
 
   # :stopdoc:
-  VERSION = '1.9.7'
+  VERSION = '1.10.0'
   LIBPATH = ::File.expand_path(::File.dirname(__FILE__)) + ::File::SEPARATOR
   PATH = ::File.dirname(LIBPATH) + ::File::SEPARATOR
   # :startdoc:
@@ -54,6 +54,8 @@ module Roo
           Excelx.new(file)
         when '.ods'
           Openoffice.new(file)
+        when '.csv'
+          Csv.new(file)
           # when ''
         else
           Google.new(file)
@@ -70,6 +72,7 @@ require 'roo/openoffice'
 require 'roo/excel'
 require 'roo/excelx'
 require 'roo/google'
+require 'roo/csv'
 
 #Roo.require_all_libs_relative_to(__FILE__)
 

data/lib/roo/csv.rb

@@ -0,0 +1,117 @@
+require 'rubygems'
+require 'csv'
+
+# The Csv class can read csv files (must be separated with commas) which then
+# can be handled like spreadsheets. This means you can access cells like A5
+# within these files.
+# The Csv class provides only string objects. If you want conversions to other
+# types you have to do it yourself.
+
+class Csv < GenericSpreadsheet
+  def initialize(filename, packed=nil, file_warning=:error, tmpdir=nil)
+    @filename = filename
+    super()
+    @cell = Hash.new
+    @cell_type = Hash.new
+    @cells_read = Hash.new
+    @first_row = Hash.new
+    @last_row = Hash.new
+    @first_column = Hash.new
+    @last_column = Hash.new
+  end
+
+  # Returns an array with the names of the sheets. In Csv class there is only
+  # one dummy sheet, because a csv file cannot have more than one sheet.
+  def sheets
+    ['default']
+  end
+
+  def cell(row, col, sheet=nil)
+    sheet = @default_sheet unless sheet
+    read_cells(sheet) unless @cells_read[sheet]
+    row,col = normalize(row,col)
+    @cell[[row,col]]
+  end
+
+  def celltype(row, col, sheet=nil)
+    sheet = @default_sheet unless sheet
+    read_cells(sheet) unless @cells_read[sheet]
+    row,col = normalize(row,col)
+    @cell_type[[row,col]]
+  end
+
+  def cell_postprocessing(row,col,value)
+    value
+  end
+
+  private
+
+  def celltype_class(value)
+    return {String => :string,
+      Float => :float,
+      Date => :date,
+      DateTime => :datetime,
+    }[value.class]
+    raise "unknown type for #{value.inspect}"
+  end
+
+  def read_cells(sheet=nil)
+    sheet = @default_sheet unless sheet
+    @cell_type = {} unless @cell_type
+    @cell = {} unless @cell
+    @first_row[sheet] = 1
+    @last_row[sheet] = 0
+    @first_column[sheet] = 1
+    @last_column[sheet] = 1
+    rownum = 1
+    CSV.foreach(@filename) do |row|
+      row.each_with_index do |elem,i|
+        @cell[[rownum,i+1]] = cell_postprocessing rownum,i+1, elem
+        @cell_type[[rownum,i+1]] = celltype_class @cell[[rownum,i+1]]
+        if i+1 > @last_column[sheet]
+          @last_column[sheet] += 1
+        end
+      end
+      rownum += 1
+      @last_row[sheet] += 1
+    end
+    @cells_read[sheet] = true
+    #-- adjust @first_row if neccessary
+    loop do
+      if !row(@first_row[sheet]).any? and @first_row[sheet] < @last_row[sheet]
+        @first_row[sheet] += 1
+      else
+        break
+      end
+    end
+    #-- adjust @last_row if neccessary
+    loop do
+      if !row(@last_row[sheet]).any? and @last_row[sheet] and
+          @last_row[sheet] > @first_row[sheet]
+        @last_row[sheet] -= 1
+      else
+        break
+      end
+    end
+    #-- adjust @first_column if neccessary
+    loop do
+      if !column(@first_column[sheet]).any? and
+          @first_column[sheet] and
+          @first_column[sheet] < @last_column[sheet]
+        @first_column[sheet] += 1
+      else
+        break
+      end
+    end
+    #-- adjust @last_column if neccessary
+    loop do
+      if !column(@last_column[sheet]).any? and
+          @last_column[sheet] and
+          @last_column[sheet] > @first_column[sheet]
+        @last_column[sheet] -= 1
+      else
+        break
+      end
+    end
+  end
+end # class Csv

data/lib/roo/excel.rb

@@ -93,7 +93,7 @@ end
 # Class for handling Excel-Spreadsheets
 class Excel < GenericSpreadsheet 
 
-  EXCEL_NO_FORMULAS = 'formulas are not supported for excel spreadsheets'
+  EXCEL_NO_FORMULAS = 'Formulas are not supported for excel spreadsheets.'
 
   # Creates a new Excel spreadsheet object.
   # Parameter packed: :zip - File is a zip-file
@@ -191,17 +191,17 @@ class Excel < GenericSpreadsheet
 
   # returns NO formula in excel spreadsheets
   def formula(row,col,sheet=nil)
-    raise EXCEL_NO_FORMULAS
+    wait_for_version_080
   end
 
   # raises an exception because formulas are not supported for excel files
   def formula?(row,col,sheet=nil)
-    raise EXCEL_NO_FORMULAS
+    wait_for_version_080
   end
 
   # returns NO formulas in excel spreadsheets
   def formulas(sheet=nil)
-    raise EXCEL_NO_FORMULAS
+    wait_for_version_080
   end
 
   # Given a cell, return the cell's font
@@ -459,5 +459,14 @@ class Excel < GenericSpreadsheet
     return value_type, value
   end
   private :read_cell
-  
+
+  def wait_for_version_080
+    if Spreadsheet::VERSION<='0.8.0'
+      raise EXCEL_NO_FORMULAS+
+        " We have to wait for the 0.8.0 version of the Spreadsheet gem (currently used version is #{Spreadsheet::VERSION})"
+    else
+      raise 'Thomas should implement formulas from Spreadsheet gem'
+    end
+  end
+
 end

data/lib/roo/excelx.rb

@@ -48,6 +48,8 @@ class Excelx < GenericSpreadsheet
     'hh:mm:ss' => :time,
     "dd/mm/yy\\ hh:mm" => :datetime,
     'dd/mmm/yy' => :date, # 2011-05-21
+    'yyyy-mm-dd' => :date, # 2011-09-16
+    # was used in a spreadsheet file from a windows phone
   }
   STANDARD_FORMATS = { 
     0 => 'General',
@@ -102,6 +104,7 @@ class Excelx < GenericSpreadsheet
     end
     @@nr += 1
     @file_nr = @@nr
+    @comments_files = Array.new
     extract_content(@filename)
     file = File.new(File.join(@tmpdir, @file_nr.to_s+"_roo_workbook.xml"))
     @workbook_doc = Nokogiri::XML(file)
@@ -127,6 +130,12 @@ class Excelx < GenericSpreadsheet
       @sheet_doc[i] = Nokogiri::XML(file)
       file.close
     end
+    @comments_doc = []
+    @comments_files.each_with_index do |item, i|
+      file = File.new(item)
+      @comments_doc[i] = Nokogiri::XML(file)
+      file.close
+    end
     FileUtils::rm_r(@tmpdir)
     @default_sheet = self.sheets.first
     @cell = Hash.new
@@ -140,8 +149,26 @@ class Excelx < GenericSpreadsheet
     @excelx_type = Hash.new
     @excelx_value = Hash.new
     @s_attribute = Hash.new # TODO: ggf. wieder entfernen nur lokal benoetigt
+    @label = Hash.new
+    @labels_read = false
+    @comment = Hash.new
+    @comments_read = Hash.new
   end
 
+  def method_missing(m,*args)
+    # is method name a label name
+    read_labels unless @labels_read
+    if @label.has_key?(m.to_s)
+      sheet = @default_sheet unless sheet
+      read_cells(sheet) unless @cells_read[sheet]
+      row,col = label(m.to_s)
+      cell(row,col)
+    else
+      # call super for methods like #a1
+      super
+    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
@@ -183,7 +210,21 @@ class Excelx < GenericSpreadsheet
     row,col = normalize(row,col)
     formula(row,col) != nil
   end
-  
+
+    # returns each formula in the selected sheet as an array of elements
+  # [row, col, formula]
+  def formulas(sheet=nil)
+    sheet = @default_sheet unless sheet
+    read_cells(sheet) unless @cells_read[sheet]
+    if @formula[sheet]
+      @formula[sheet].each.collect do |elem|
+        [elem[0][0], elem[0][1], elem[1]]
+      end
+    else
+      []
+    end
+  end
+
   class Font
     attr_accessor :bold, :italic, :underline
     
@@ -295,13 +336,80 @@ class Excelx < GenericSpreadsheet
     @cell[sheet].inspect
   end
 
+  # returns the row,col values of the labelled cell
+  # (nil,nil) if label is not defined
+  def label(labelname)
+    read_labels unless @labels_read
+    unless @label.size > 0
+      return nil,nil,nil
+    end
+    if @label.has_key? labelname
+      return @label[labelname][1].to_i,
+        GenericSpreadsheet.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, [sheetname,row,col]]
+  def labels
+    # sheet = @default_sheet unless sheet
+    # read_cells(sheet) unless @cells_read[sheet]
+    read_labels unless @labels_read
+    result = []
+    @label.each do |label|
+      result << [ label[0], # name
+        [ label[1][1].to_i, # row
+          GenericSpreadsheet.letter_to_number(label[1][2]), # column
+          label[1][0], # sheet
+        ] ]
+    end
+    result
+  end
+
+  # returns the comment at (row/col)
+  # nil if there is no comment
+  def comment(row,col,sheet=nil)
+    sheet = @default_sheet unless sheet
+    #read_cells(sheet) unless @cells_read[sheet]
+    read_comments(sheet) unless @comments_read[sheet]
+    row,col = normalize(row,col)
+    return nil unless @comment[sheet]
+    @comment[sheet][[row,col]]
+  end
+
+  # true, if there is a comment
+  def comment?(row,col,sheet=nil)
+    sheet = @default_sheet unless sheet
+    # read_cells(sheet) unless @cells_read[sheet]
+    read_comments(sheet) unless @comments_read[sheet]
+    row,col = normalize(row,col)
+    comment(row,col) != nil
+  end
+
+  # returns each comment in the selected sheet as an array of elements
+  # [row, col, comment]
+  def comments(sheet=nil)
+    sheet = @default_sheet unless sheet
+    read_comments(sheet) unless @comments_read[sheet]
+    if @comment[sheet]
+      @comment[sheet].each.collect do |elem|
+        [elem[0][0],elem[0][1],elem[1]]
+      end
+    else
+      []
+    end
+  end
+
   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,
-    excelx_value=nil,
-    s_attribute=nil)
+      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
@@ -332,32 +440,6 @@ class Excelx < GenericSpreadsheet
     @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
-    i = 0
-    while i<s.length and "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz".include?(s[i,1])
-      letter += s[i,1]
-      i+=1
-    end
-    while i<s.length and "0123456789".include?(s[i,1])
-      number = number*10 + s[i,1].to_i
-      i+=1
-    end
-    if letter=="" or number==0
-      raise ArgumentError
-    end
-    return letter,number
-  end
-
-  def split_coordinate(str)
-    letter,number = split_coord(str)
-    x = GenericSpreadsheet.letter_to_number(letter)
-    y = number
-    return x,y
-  end
-
   def format2type(format)
     format = format.to_s # weil von Typ Nokogiri::XML::Attr
     if FORMATS.has_key? format
@@ -387,6 +469,10 @@ class Excelx < GenericSpreadsheet
       elsif c['t'] == 'str'
         tmp_type = :string
         # 2011-02-25 END
+        # 2011-09-15 BEGIN
+      elsif c['t'] == 'inlineStr'
+	      tmp_type = :inlinestr
+        # 2011-09-15 END
       else
         s_attribute = c['s'].to_i
         format = attribute2format(s_attribute)
@@ -394,6 +480,23 @@ class Excelx < GenericSpreadsheet
       end
       formula = nil
       c.children.each do |cell|
+	      # 2011-09-15 BEGIN
+        if cell.name == 'is'
+          cell.children.each do |is|
+            if is.name == 't'
+              inlinestr_content = is.content
+              vt = :string
+              str_v = inlinestr_content
+              excelx_type = :string
+              y, x = GenericSpreadsheet.split_coordinate(c['r'])
+              v = nil
+              tr=nil #TODO: ???s
+              excelx_value = inlinestr_content #cell.content
+              set_cell_values(sheet,x,y,0,v,vt,formula,tr,str_v,excelx_type,excelx_value,s_attribute)
+            end
+          end
+        end
+	      # 2011-09-15 END
         if cell.name == 'f'
           formula = cell.content
         end
@@ -439,7 +542,7 @@ class Excelx < GenericSpreadsheet
             vt = :float
             v = cell.content
           end
-          x,y = split_coordinate(c['r'])
+          y, x = GenericSpreadsheet.split_coordinate(c['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
@@ -450,27 +553,115 @@ class Excelx < GenericSpreadsheet
       raise RangeError
     end
     @cells_read[sheet] = true
+    # begin comments
+=begin
+Datei xl/comments1.xml
+  <?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
+  <comments xmlns="http://schemas.openxmlformats.org/spreadsheetml/2006/main">
+    <authors>
+      <author />
+    </authors>
+    <commentList>
+      <comment ref="B4" authorId="0">
+        <text>
+          <r>
+            <rPr>
+              <sz val="10" />
+              <rFont val="Arial" />
+              <family val="2" />
+            </rPr>
+            <t>Kommentar fuer B4</t>
+          </r>
+        </text>
+      </comment>
+      <comment ref="B5" authorId="0">
+        <text>
+          <r>
+            <rPr>
+            <sz val="10" />
+            <rFont val="Arial" />
+            <family val="2" />
+          </rPr>
+          <t>Kommentar fuer B5</t>
+        </r>
+      </text>
+    </comment>
+  </commentList>
+  </comments>
+=end
+=begin
+    if @comments_doc[self.sheets.index(sheet)]
+      read_comments(sheet)
+    end
+=end
+    #end comments
   end
-  
+
+  # Reads all comments from a sheet
+  def read_comments(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)
+    return unless @comments_doc[n] #>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
+    @comments_doc[n].xpath("//*[local-name()='comments']").each do |comment|
+      comment.children.each do |cc|
+        if cc.name == 'commentList'
+          cc.children.each do |commentlist|
+            if commentlist.name == 'comment'
+              ref = commentlist.attributes['ref'].to_s
+              row,col = GenericSpreadsheet.split_coordinate(ref)
+              commentlist.children.each do |clc|
+                if clc.name == 'text'
+                  clc.children.each do |text|
+                    if text.name == 'r'
+                      text.children.each do |r|
+                        if r.name == 't'
+                          comment = r.text
+                          @comment[sheet] = Hash.new unless @comment[sheet]
+                          @comment[sheet][[row,col]] = comment
+                        end
+                      end
+                    end
+                  end
+                end
+              end
+            end
+          end
+        end
+      end
+    end
+    @comments_read[sheet] = true
+  end
+
+  def read_labels
+    @workbook_doc.xpath("//*[local-name()='definedName']").each do |defined_name|
+	    # "Sheet1!$C$5"
+      sheet = defined_name.text.split('!').first
+      coordinates = defined_name.text.split('!')[1]
+      dummy,col,row = coordinates.split('$')
+      @label[defined_name['name']] = [sheet,row,col]
+    end
+    @labels_read = true
+  end
+
   # Checks if the default_sheet exists. If not an RangeError exception is
   # raised
   def check_default_sheet
     sheet_found = false
     raise ArgumentError, "Error: default_sheet not set" if @default_sheet == nil
-    
     sheet_found = true if sheets.include?(@default_sheet)
-    
     if ! sheet_found
       raise RangeError, "sheet '#{@default_sheet}' not found"
     end
   end
 
-  # extracts all needed files from the zip file
+  # Extracts all needed files from the zip file
   def process_zipfile(zipfilename, zip, path='')
     @sheet_files = []
     Zip::ZipFile.open(zipfilename) {|zf|
       zf.entries.each {|entry|
-        #entry.extract
         if entry.to_s.end_with?('workbook.xml')
           open(@tmpdir+'/'+@file_nr.to_s+'_roo_workbook.xml','wb') {|f|
             f << zip.read(entry)
@@ -493,9 +684,16 @@ class Excelx < GenericSpreadsheet
           }
           @sheet_files[nr.to_i-1] = @tmpdir+'/'+@file_nr.to_s+"_roo_sheet#{nr}"
         end
+        if entry.to_s =~ /comments([0-9]+).xml$/
+          nr = $1
+          open(@tmpdir+'/'+@file_nr.to_s+"_roo_comments#{nr}",'wb') {|f|
+            f << zip.read(entry)
+          }
+          @comments_files[nr.to_i-1] = @tmpdir+'/'+@file_nr.to_s+"_roo_comments#{nr}"
+        end
       }
     }
-    return
+    # return
   end
 
   # extract files from the zip file