spreadsheet 0.6.1.3 → 0.6.1.4

data/GUIDE.txt

@@ -50,8 +50,8 @@ To access the values stored in a Row, treat the Row like an Array.
 
   row[0]
 
--> this will return a String, a Float, an Integer, a Formula or a Date or
-DateTime object - or nil if the cell is empty.
+-> this will return a String, a Float, an Integer, a Formula, a Link or a Date
+or DateTime object - or nil if the cell is empty.
 
 More information about the formatting of a cell can be found in the Format
 with the equivalent index

data/History.txt

@@ -1,3 +1,14 @@
+=== 0.6.1.4 / 2008-10-23
+
+* 1 Bugfix
+
+  * Biff8#wide now works properly even if $KCODE=='UTF-8'
+    (Thanks to Bjørn Hjelle for the Bugreport)
+
+* 1 minor enhancement
+
+  * Read/Write functionality for Links (only URLs can be written as of now)
+
 === 0.6.1.3 / 2008-10-21
 
 * 2 Bugfixes

data/Manifest.txt

@@ -32,6 +32,7 @@ lib/spreadsheet/excel/writer/worksheet.rb
 lib/spreadsheet/font.rb
 lib/spreadsheet/format.rb
 lib/spreadsheet/formula.rb
+lib/spreadsheet/link.rb
 lib/spreadsheet/row.rb
 lib/spreadsheet/workbook.rb
 lib/spreadsheet/worksheet.rb

data/lib/spreadsheet.rb

@@ -42,7 +42,7 @@ module Spreadsheet
 
   ##
   # The version of Spreadsheet you are using.
-  VERSION = '0.6.1.3'
+  VERSION = '0.6.1.4'
 
   ##
   # Default client Encoding. Change this value if your application uses a

data/lib/spreadsheet/excel.rb

@@ -81,5 +81,8 @@ module Spreadsheet
         write row, col, data, format
       end
     end
+    def write_url row, col, url, string=url, format=nil
+      row(row)[col] = Link.new url, string
+    end
   end
 end

data/lib/spreadsheet/excel/internals.rb

@@ -211,6 +211,7 @@ module Internals
     :font         => 0x0031, # ●● FONT ➜ 6.43
     :format       => 0x041e, # ○○ FORMAT (Number Format) ➜ 6.45
     :formula      => 0x0006, #    FORMULA ➜ 6.46
+    :hlink        => 0x01b8, #    HLINK ➜ 6.52 (BIFF8 only)
     :label        => 0x0204, #    LABEL ➜ 6.59 (BIFF2-BIFF7)
     :labelsst     => 0x00fd, #    LABELSST ➜ 6.61 (BIFF8 only)
     :mulblank     => 0x00be, #    MULBLANK ➜ 6.64 (BIFF5-BIFF8)

data/lib/spreadsheet/excel/reader.rb

@@ -1,6 +1,7 @@
 require 'spreadsheet/encodings'
 require 'spreadsheet/font'
 require 'spreadsheet/formula'
+require 'spreadsheet/link'
 require 'spreadsheet/excel/error'
 require 'spreadsheet/excel/internals'
 require 'spreadsheet/excel/sst_entry'
@@ -384,6 +385,176 @@ class Reader
       # leave the Formula value blank
     end
   end
+  def read_hlink worksheet, work, pos, len
+    # 6.53.1 Common Record Contents
+    # Offset  Size  Contents
+    #      0     8  Cell range address of all cells containing this hyperlink
+    #               (➜ 3.13.1)
+    #      8    16  GUID of StdLink:
+    #               D0 C9 EA 79 F9 BA CE 11 8C 82 00 AA 00 4B A9 0B
+    #               (79EAC9D0-BAF9-11CE-8C82-00AA004BA90B)
+    #     24     4  Unknown value: 0x00000002
+    #     28     4  Option flags (see below)
+    #               Bit  Mask        Contents
+    #                 0  0x00000001  0 = No link extant
+    #                                1 = File link or URL
+    #                 1  0x00000002  0 = Relative file path
+    #                                1 = Absolute path or URL
+    #           2 and 4  0x00000014  0 = No description
+    #                                1 (both bits) = Description
+    #                 3  0x00000008  0 = No text mark
+    #                                1 = Text mark
+    #                 7  0x00000080  0 = No target frame
+    #                                1 = Target frame
+    #                 8  0x00000100  0 = File link or URL
+    #                                1 = UNC path (incl. server name)
+    #--------------------------------------------------------------------------
+    #   [32]     4  (optional, see option flags) Character count of description
+    #               text, including trailing zero word (dl)
+    #   [36]  2∙dl  (optional, see option flags) Character array of description
+    #               text, no Unicode string header, always 16-bit characters,
+    #               zero-terminated
+    #--------------------------------------------------------------------------
+    # [var.]     4  (optional, see option flags) Character count of target
+    #               frame, including trailing zero word (fl)
+    # [var.]  2∙fl  (optional, see option flags) Character array of target
+    #               frame, no Unicode string header, always 16-bit characters,
+    #               zero-terminated
+    #--------------------------------------------------------------------------
+    #   var.  var.  Special data (➜ 6.53.2 and following)
+    #--------------------------------------------------------------------------
+    # [var.]     4  (optional, see option flags) Character count of the text
+    #               mark, including trailing zero word (tl)
+    # [var.]  2∙tl  (optional, see option flags) Character array of the text
+    #               mark without “#” sign, no Unicode string header, always
+    #               16-bit characters, zero-terminated
+    firstrow, lastrow, firstcol, lastcol, guid, opts = work.unpack 'v4H32x4V'
+    has_link = opts & 0x0001
+    absolute = opts & 0x0002
+    desc     = opts & 0x0014
+    textmark = opts & 0x0008
+    target   = opts & 0x0080
+    unc      = opts & 0x0100
+    link = Link.new
+    url, description = nil
+    pos = 32
+    if desc > 0
+      description, pos = read_hlink_string work, pos
+      link << description
+    end
+    if target > 0
+      link.target_frame, pos = read_hlink_string work, pos
+    end
+    if unc > 0
+      # 6.53.4 Hyperlink to a File with UNC (Universal Naming Convention) Path
+      # These data fields are for UNC paths containing a server name (for
+      # instance “\\server\path\file.xls”). The lower 9 bits of the option
+      # flags field must be 1.x00x.xx112.
+      # Offset  Size  Contents
+      #      0     4  Character count of the UNC,
+      #               including trailing zero word (fl)
+      #      4  2∙fl  Character array of the UNC, no Unicode string header,
+      #               always 16-bit characters, zeroterminated.
+      link.url, pos = read_hlink_string work, pos
+    elsif has_link > 0
+      uid, = work.unpack "x#{pos}H32"
+      pos += 16
+      if uid == "e0c9ea79f9bace118c8200aa004ba90b"
+        # 6.53.2 Hyperlink containing a URL (Uniform Resource Locator)
+        # These data fields occur for links which are not local files or files
+        # in the local network (for instance HTTP and FTP links and e-mail
+        # addresses). The lower 9 bits of the option flags field must be
+        # 0.x00x.xx112 (x means optional, depending on hyperlink content). The
+        # GUID could be used to distinguish a URL from a file link.
+        # Offset  Size  Contents
+        #      0    16  GUID of URL Moniker:
+        #               E0 C9 EA 79 F9 BA CE 11 8C 82 00 AA 00 4B A9 0B
+        #               (79EAC9E0-BAF9-11CE-8C82-00AA004BA90B)
+        #     16     4  Size of character array of the URL, including trailing
+        #               zero word (us). There are us/2-1 characters in the
+        #               following string.
+        #     20    us  Character array of the URL, no Unicode string header,
+        #               always 16-bit characters, zeroterminated
+        size, = work.unpack "x#{pos}V"
+        pos += 4
+        data = work[pos, size].chomp "\000\000"
+        link.url = client data, 'UTF-16LE'
+        pos += size
+      else
+        # 6.53.3 Hyperlink to a Local File
+        # These data fields are for links to files on local drives. The path of
+        # the file can be complete with drive letter (absolute) or relative to
+        # the location of the workbook. The lower 9 bits of the option flags
+        # field must be 0.x00x.xxx12. The GUID could be used to distinguish a
+        # URL from a file link.
+        # Offset  Size  Contents
+        #      0    16  GUID of File Moniker:
+        #               03 03 00 00 00 00 00 00 C0 00 00 00 00 00 00 46
+        #               (00000303-0000-0000-C000-000000000046)
+        #     16     2  Directory up-level count. Each leading “..\” in the
+        #               file link is deleted and increases this counter.
+        #     18     4  Character count of the shortened file path and name,
+        #               including trailing zero byte (sl)
+        #     22    sl  Character array of the shortened file path and name in
+        #               8.3-DOS-format. This field can be filled with a long
+        #               file name too. No Unicode string header, always 8-bit
+        #               characters, zeroterminated.
+        #  22+sl    24  Unknown byte sequence:
+        #               FF FF AD DE 00 00 00 00
+        #               00 00 00 00 00 00 00 00
+        #               00 00 00 00 00 00 00 00
+        #  46+sl     4  Size of the following file link field including string
+        #               length field and additional data field (sz). If sz is
+        #               zero, nothing will follow (except a text mark).
+        # [50+sl]    4  (optional) Size of character array of the extended file
+        #               path and name (xl). There are xl/2 characters in the
+        #               following string.
+        # [54+sl]    2  (optional) Unknown byte sequence: 03 00
+        # [56+sl]   xl  (optional) Character array of the extended file path
+        #               and name (xl), no Unicode string header, always 16-bit
+        #               characters, not zero-terminated
+        uplevel, count = work.unpack "x#{pos}vV"
+        pos += 6
+        # TODO: short file path may have any of the OEM encodings. Find out which
+        #       and use the #client method to convert the encoding.
+        prefix = internal('..\\', 'UTF-8') * uplevel
+        link.dos = link.url = prefix << work[pos, count].chomp("\000")
+        pos += count + 24
+        total, size = work.unpack "x#{pos}V2"
+        pos += 10
+        if total > 0
+          link.url = client work[pos, size], 'UTF-16LE'
+          pos += size
+        end
+      end
+    else
+      # 6.53.5 Hyperlink to the Current Workbook
+      # In this case only the text mark field is present (optional with
+      # description).
+      # Example: The URL “#Sheet2!B1:C2” refers to the given range in the
+      # current workbook.
+      # The lower 9 bits of the option flags field must be 0.x00x.1x002.
+    end
+    if textmark > 0
+      link.fragment, _ = read_hlink_string work, pos
+    end
+    if link.empty?
+      link << link.href
+    end
+    firstrow.upto lastrow do |row|
+      firstcol.upto lastcol do |col|
+        worksheet.add_link row, col, link
+      end
+    end
+  end
+  def read_hlink_string work, pos
+    count, = work.unpack "x#{pos}V"
+    len = count * 2
+    pos += 4
+    data = work[pos, len].chomp "\000\000"
+    pos += len
+    [client(data, 'UTF-16LE'), pos]
+  end
   def read_index worksheet, work, pos, len
     # Offset  Size  Contents
     #      0     4  Not used
@@ -615,6 +786,8 @@ class Reader
       when :row        # ○○ Row Blocks ➜ 5.7
                        # ●  ROW ➜ 6.83
         set_row_address worksheet, work, pos, len
+      when :hlink
+        read_hlink worksheet, work, pos, len
       end
       previous = op
     end

data/lib/spreadsheet/excel/reader/biff8.rb

@@ -150,7 +150,9 @@ module Biff8
   ##
   # Insert null-characters into a compressed UTF-16 string
   def wide string
-    string.split('').zip(Array.new(string.size, 0.chr)).join
+    data = ''
+    string.each_byte do |byte| data << byte.chr << 0.chr end
+    data
   end
   private
   ##

data/lib/spreadsheet/excel/row.rb

@@ -53,7 +53,9 @@ class Row < Spreadsheet::Row
   end
   def enriched_data idx, data # :nodoc:
     res = nil
-    if fmt = format(idx)
+    if link = @worksheet.links[[@idx, idx]]
+      res = link
+    elsif fmt = format(idx)
       res = if fmt.datetime? || fmt.time?
               _datetime data
             elsif fmt.date?

data/lib/spreadsheet/excel/worksheet.rb

@@ -11,11 +11,15 @@ module Spreadsheet
 class Worksheet < Spreadsheet::Worksheet
   include Spreadsheet::Excel::Offset
   offset :dimensions
-  attr_reader :offset, :ole
+  attr_reader :offset, :ole, :links
   def initialize opts = {}
     super
     @offset, @ole, @reader = opts[:offset], opts[:ole], opts[:reader]
     @dimensions = nil
+    @links = {}
+  end
+  def add_link row, column, link
+    @links.store [row, column], link
   end
   def column idx
     ensure_rows_read

data/lib/spreadsheet/excel/writer/worksheet.rb

@@ -413,6 +413,7 @@ class Worksheet
     # ○  PHONETIC ➜ 6.73
     # ○  Conditional Formatting Table ➜ 5.12
     # ○  Hyperlink Table ➜ 5.13
+    write_hyperlink_table
     # ○  Data Validity Table ➜ 5.14
     # ○  SHEETLAYOUT ➜ 6.91 (BIFF8X only)
     # ○  SHEETPROTECTION Additional protection, ➜ 6.92 (BIFF8X only)
@@ -420,6 +421,85 @@ class Worksheet
     # ●  EOF ➜ 6.36
     write_eof
   end
+  def write_hlink row, col, link
+    # FIXME: only Hyperlinks are supported at present.
+    cell_range = [
+      row, row, # Cell range address of all cells containing this hyperlink
+      col, col, # (➜ 3.13.1)
+    ].pack 'v4'
+    guid = [
+      # GUID of StdLink:
+      # D0 C9 EA 79 F9 BA CE 11 8C 82 00 AA 00 4B A9 0B
+      # (79EAC9D0-BAF9-11CE-8C82-00AA004BA90B)
+      "d0c9ea79f9bace118c8200aa004ba90b",
+    ].pack 'H32'
+    opts  = 0x01
+    opts |= 0x02
+    opts |= 0x14 unless link == link.url
+    opts |= 0x08 if link.fragment
+    opts |= 0x80 if link.target_frame
+    # TODO: UNC support
+    options = [
+      2,        # Unknown value: 0x00000002
+      opts,     # Option flags
+                #     Bit  Mask        Contents
+                #       0  0x00000001  0 = No link extant
+                #                      1 = File link or URL
+                #       1  0x00000002  0 = Relative file path
+                #                      1 = Absolute path or URL
+                # 2 and 4  0x00000014  0 = No description
+                #                      1 (both bits) = Description
+                #       3  0x00000008  0 = No text mark
+                #                      1 = Text mark
+                #       7  0x00000080  0 = No target frame
+                #                      1 = Target frame
+                #       8  0x00000100  0 = File link or URL
+                #                      1 = UNC path (incl. server name)
+
+    ].pack('V2')
+    tail = []
+    unless link == link.url
+      desc = internal(link).dup << "\000\000"
+      tail.push [desc.size / 2].pack('V'), desc
+    end
+    if link.target_frame
+      frme = internal(link.target_frame).dup << "\000\000"
+      tail.push [frme.size / 2].pack('V'), frme
+    end
+    url = internal(link.url).dup << "\000\000"
+    tail.push [
+      # 6.53.2 Hyperlink containing a URL (Uniform Resource Locator)
+      # These data fields occur for links which are not local files or files
+      # in the local network (for instance HTTP and FTP links and e-mail
+      # addresses). The lower 9 bits of the option flags field must be
+      # 0.x00x.xx112 (x means optional, depending on hyperlink content). The
+      # GUID could be used to distinguish a URL from a file link.
+      # GUID of URL Moniker:
+      # E0 C9 EA 79 F9 BA CE 11 8C 82 00 AA 00 4B A9 0B
+      # (79EAC9E0-BAF9-11CE-8C82-00AA004BA90B)
+      'e0c9ea79f9bace118c8200aa004ba90b',
+      url.size  # Size of character array of the URL, including trailing zero
+                # word (us). There are us/2-1 characters in the following
+                # string.
+    ].pack('H32V'), url
+    if link.fragment
+      frag = internal(link.fragment).dup << "\000\000"
+      tail.push [frag.size / 2].pack('V'), frag
+    end
+    write_op opcode(:hlink), cell_range, guid, options, *tail
+  end
+  def write_hyperlink_table
+    # TODO: theoretically it's possible to write fewer records by combining
+    #       identical neighboring links in cell-ranges
+    links = []
+    @worksheet.each do |row|
+      row.each_with_index do |cell, idx|
+        if cell.is_a? Link
+          write_hlink row.idx, idx, cell
+        end
+      end
+    end
+  end
   def write_iteration
     its = 0 # 0 = Iterations off; 1 = Iterations on
     write_op 0x0011, [its].pack('v')

data/lib/spreadsheet/link.rb

@@ -0,0 +1,40 @@
+require 'uri'
+
+module Spreadsheet
+  ##
+  # The Link class. Is a Subclass of String, which lets you treat a Cell that
+  # contains a Link just as if it was a String (containing the link's description
+  # if there is one or the url with fragment otherwise), but gives you access
+  # to the url, fragment and target_frame if you need it.
+  #
+  #
+  # Interesting Attributes
+  # #url          :: The Uniform Resource Location this Link points to.
+  # #fragment     :: Also called text mark: http://example.com/page.html#fragment
+  # #target_frame :: Which frame a Link should be opened in, should also support
+  #                  the special frames _blank, _parent, _self and _top.
+  # #dos          :: Excel may store a DOS-Filename together with the long
+  #                  Filename introduced in VFAT. You probably will not need this,
+  #                  but if you do, here is where you can find it.
+  class Link < String
+    attr_accessor :target_frame, :url, :dos, :fragment
+    def initialize url='', description=url
+      super description
+      @url = url
+    end
+    ##
+    # The Url with the fragment appended if present.
+    def href
+      href = (@url || @dos).to_s.dup
+      if @fragment
+        href << '#' << @fragment
+      end
+      href
+    end
+    ##
+    # Attempts to parse the output of href. May raise a URI::InvalidURIError
+    def to_uri
+      URI.parse href
+    end
+  end
+end

data/lib/spreadsheet/worksheet.rb

@@ -30,6 +30,7 @@ module Spreadsheet
       @workbook = opts[:workbook]
       @rows = []
       @columns = []
+      @links = {}
     end
     ##
     # Add a Format to the Workbook. If you use Row#set_format, you should not
@@ -139,6 +140,14 @@ module Spreadsheet
       sprintf "#<%s:0x%014x %s @rows[%i]>", self.class, object_id,
                                             variables, row_count
     end
+    ## The last Row containing any data
+    def last_row
+      row(last_row_index)
+    end
+    ## The index of the last Row containing any data
+    def last_row_index
+      [dimensions[1] - 1, 0].max
+    end
     ##
     # Replace the Row at _idx_ with the following arguments. Like #update_row,
     # but truncates the Row if there are fewer arguments than Cells in the Row.

data/test/integration.rb

@@ -44,8 +44,8 @@ module Spreadsheet
         enc = Encoding.find enc
       end
       assert_equal enc, book.encoding
-      assert_equal 23, book.formats.size
-      assert_equal 4, book.fonts.size
+      assert_equal 24, book.formats.size
+      assert_equal 5, book.fonts.size
       str1 = book.shared_string 0
       assert_equal @@iconv.iconv('Shared String'), str1
       str2 = book.shared_string 1
@@ -76,7 +76,7 @@ module Spreadsheet
       assert_equal 10, sheet.row_count
       assert_equal 11, sheet.column_count
       useds = [0,0,0,0,0,0,0,1,0,0]
-      unuseds = [2,2,1,1,1,2,1,11,1,1]
+      unuseds = [2,2,1,1,1,2,1,11,1,2]
       sheet.each do |row|
         assert_equal useds.shift, row.first_used
         assert_equal unuseds.shift, row.first_unused
@@ -147,8 +147,8 @@ module Spreadsheet
         enc = Encoding.find enc
       end
       assert_equal enc, book.encoding
-      assert_equal 23, book.formats.size
-      assert_equal 4, book.fonts.size
+      assert_equal 24, book.formats.size
+      assert_equal 5, book.fonts.size
       str1 = book.shared_string 0
       assert_equal 'Shared String', str1
       str2 = book.shared_string 1
@@ -179,7 +179,7 @@ module Spreadsheet
       assert_equal 10, sheet.row_count
       assert_equal 11, sheet.column_count
       useds = [0,0,0,0,0,0,0,1,0,0]
-      unuseds = [2,2,1,1,1,2,1,11,1,1]
+      unuseds = [2,2,1,1,1,2,1,11,1,2]
       sheet.each do |row|
         assert_equal useds.shift, row.first_used
         assert_equal unuseds.shift, row.first_unused
@@ -238,6 +238,11 @@ module Spreadsheet
       assert_equal 0.0001, row[0]
       row = sheet.row 9
       assert_equal 0.00009, row[0]
+      link = row[1]
+      assert_instance_of Link, link
+      assert_equal 'Link-Text', link
+      assert_equal 'http://scm.ywesee.com/spreadsheet', link.url
+      assert_equal 'http://scm.ywesee.com/spreadsheet', link.href
     end
     def test_version_excel95__ooffice__utf16
       Spreadsheet.client_encoding = 'UTF-16LE'
@@ -542,7 +547,7 @@ module Spreadsheet
       assert_equal 10, sheet.row_count
       assert_equal 11, sheet.column_count
       useds = [0,0,0,0,0,0,0,1,0,0]
-      unuseds = [2,2,1,1,1,2,1,11,1,1]
+      unuseds = [2,2,1,1,1,2,1,11,1,2]
       sheet.each do |row|
         assert_equal useds.shift, row.first_used
         assert_equal unuseds.shift, row.first_unused
@@ -651,7 +656,7 @@ module Spreadsheet
       assert_equal 10, sheet.row_count
       assert_equal 11, sheet.column_count
       useds = [0,0,0,0,0,0,0,1,0,0]
-      unuseds = [2,2,1,1,1,2,1,11,1,1]
+      unuseds = [2,2,1,1,1,2,1,11,1,2]
       sheet.each do |row|
         assert_equal useds.shift, row.first_used
         assert_equal unuseds.shift, row.first_unused
@@ -737,6 +742,9 @@ module Spreadsheet
       fmt = Format.new :color => 'aqua'
       sheet1[5,0] = 0.75
       sheet1.row(5).set_format 0, fmt
+      link = Link.new 'http://scm.ywesee.com/?p=spreadsheet;a=summary',
+                      'The Spreadsheet GitWeb'
+      sheet1[5,1] = link
       sheet1[6,0] = 1
       fmt = Format.new :color => 'green'
       sheet1.row(6).set_format 0, fmt
@@ -850,6 +858,11 @@ module Spreadsheet
       assert_equal 0.75, row[0]
       assert_equal 0.75, sheet[5,0]
       assert_equal 0.75, sheet.cell(5,0)
+      link = row[1]
+      assert_instance_of Link, link
+      url = @@iconv.iconv 'http://scm.ywesee.com/?p=spreadsheet;a=summary'
+      assert_equal @@iconv.iconv('The Spreadsheet GitWeb'), link
+      assert_equal url, link.url
       row = sheet.row 6
       assert_equal :green, row.format(0).font.color
       assert_equal 1, row[0]

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: spreadsheet
 version: !ruby/object:Gem::Version 
-  version: 0.6.1.3
+  version: 0.6.1.4
 platform: ruby
 authors: 
 - Hannes Wyss
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2008-10-21 00:00:00 +02:00
+date: 2008-10-23 00:00:00 +02:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -80,6 +80,7 @@ files:
 - lib/spreadsheet/font.rb
 - lib/spreadsheet/format.rb
 - lib/spreadsheet/formula.rb
+- lib/spreadsheet/link.rb
 - lib/spreadsheet/row.rb
 - lib/spreadsheet/workbook.rb
 - lib/spreadsheet/worksheet.rb