roo 2.1.1 → 2.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 53d37d09a786f9536fd5bb00c7cacf9888c8b032
-  data.tar.gz: 284974b328b26668c58e3b952a12b8777439037f
+  metadata.gz: 0a61e2a1f190d2a30554f2bed45557c31573761e
+  data.tar.gz: 3bebd2370044ab8c20911368291932337201bb73
 SHA512:
-  metadata.gz: 4b0e6d58d52600710fc5feb49e7483fc81e4bca65fa9013a3fa1004f5526107392b02f982ae981676544e301056a0246b4b85e38d3959565174c247ce49695a9
-  data.tar.gz: 5c3beb6580bece8aa730d1f29b6316889ebb7ce5a7003d8f654edcb1c9800a9f939021fd0d88749489ffdee6ebf496739664e5fede43aeb0067a2a16ad46f2f5
+  metadata.gz: e57f90c7aad4b738399b3b3ff0e00b06cbe5fd9fd7ae68c8540f6086521d925db17a8219cacceb656012c3e2f10d814218d95a7ac2c188958c83c4c9425e5f4b
+  data.tar.gz: f1fe2348405fcd5097465d761ba60b321cfd52d45957eef61fecf8d6a43a7afd59afeb11e873318f6188dd0b117219b93b87e9ec3d99e9421a2f9defbb195cb2

data/CHANGELOG.md

@@ -1,3 +1,20 @@
+## [2.2.0] - 2015-10-31
+### Added
+- Added support for returning Integers values to Roo::OpenOffice [#258](https://github.com/roo-rb/roo/pull/258)
+- A missing Header Raises `Roo::HeaderRowNotFoundError`  [#247](https://github.com/roo-rb/roo/pull/247)
+- Roo::Excelx::Shared class to pass shared data to Roo::Excelx sheets [#220](https://github.com/roo-rb/roo/pull/220)
+- Proper Type support to Roo::Excelx [#240](https://github.com/roo-rb/roo/pull/240)
+- Added Roo::HeaderRowNotFoundError [#247](https://github.com/roo-rb/roo/pull/247)
+
+### Changed
+- Made spelling/grammar corrections in the README[260](https://github.com/roo-rb/roo/pull/260)
+- Moved Roo::Excelx::Format module [#259](https://github.com/roo-rb/roo/pull/259)
+- Updated README with details about `Roo::Excelx#each_with_streaming` method [#250](https://github.com/roo-rb/roo/pull/250)
+
+### Fixed
+- Fixed Base64 not found issue in Open Office [#267](https://github.com/roo-rb/roo/pull/267)
+- Fixed Regexp to allow method access to cells with multiple digits [#255](https://github.com/roo-rb/roo/pull/255), [#268](https://github.com/roo-rb/roo/pull/268)
+
 ## [2.1.1] - 2015-08-02
 ### Fixed invalid new lines with _x000D_ character[#231](https://github.com/roo-rb/roo/pull/231)
 ### Fixed missing URI issue. [#245](https://github.com/roo-rb/roo/pull/245)

data/Guardfile

@@ -3,7 +3,7 @@
 
 guard :minitest, test_folders: ['test'] do
   watch(%r{^test/(.*)\/?test_(.*)\.rb$})
-  watch(%r{^lib/(.*/)?([^/]+)\.rb$})     { |m| "test/#{m[1]}test_#{m[2]}.rb" }
+  watch(%r{^lib/(.*/)?([^/]+)\.rb$})     { |m| "test/#{m[1].to_s.sub('roo/', '')}test_#{m[2]}.rb" }
   watch(%r{^test/test_helper\.rb$})      { 'test' }
 end
 
@@ -21,4 +21,3 @@ guard :rspec, cmd: 'bundle exec rspec' do
   watch('spec/spec_helper.rb')  { "spec" }
   watch(%r{^spec/support/(.+)\.rb$})                  { "spec" }
 end
-

data/README.md

@@ -151,6 +151,26 @@ xlsx.each_row_streaming do |row|
 end
 ```
 
+By default blank cells will be excluded from the array. To keep them, use the option pad_cells = true. (They will be set to nil in the array)
+```ruby
+xlsx.each_row_streaming(pad_cells: true) do |row|
+  puts row.inspect # Array of Excelx::Cell objects
+end
+```
+
+To stream only some of the rows, you can use the ```max_rows``` and ```offset```options.
+```ruby
+xlsx.each_row_streaming(offset: 1) do |row| # Will exclude first (inevitably header) row
+  puts row.inspect # Array of Excelx::Cell objects
+end
+```
+
+```ruby
+xlsx.each_row_streaming(max_rows: 3) do |row| # Will yield 4 rows (it's automatically incremented by 1) after the supplied offset.
+  puts row.inspect # Array of Excelx::Cell objects
+end
+```
+
 Iterate over each row
 
 ```ruby
@@ -182,7 +202,7 @@ xlsx.formula('A', 2)
 
 ### OpenOffice / LibreOffice Support
 
-Roo::OpenOffice supports for encrypted OpenOffice spreadsheets.
+Roo::OpenOffice has support for encrypted OpenOffice spreadsheets.
 
 ```ruby
 # Load an encrypted OpenOffice Spreadsheet
@@ -210,7 +230,7 @@ ods.formula('A', 2)
 s = Roo::CSV.new("mycsv.csv")
 ```
 
-Because Roo uses the [standard CSV library](), and you can use options available to that library to parse csv files. You can pass options using the ``csv_options`` key.
+Because Roo uses the [standard CSV library](), you can use options available to that library to parse csv files. You can pass options using the ``csv_options`` key.
 
 For instance, you can load tab-delimited files (``.tsv``), and you can use a particular encoding when opening the file.
 

data/lib/roo.rb

@@ -1,4 +1,5 @@
 require 'roo/constants'
+require 'roo/errors'
 require 'roo/spreadsheet'
 require 'roo/base'
 

data/lib/roo/base.rb

@@ -303,7 +303,7 @@ class Roo::Base
   def method_missing(m, *args)
     # #aa42 => #cell('aa',42)
     # #aa42('Sheet1')  => #cell('aa',42,'Sheet1')
-    if m =~ /^([a-z]+)(\d)$/
+    if m =~ /^([a-z]+)(\d+)$/
       col = ::Roo::Utils.letter_to_number(Regexp.last_match[1])
       row = Regexp.last_match[2].to_i
       if args.empty?
@@ -392,10 +392,10 @@ class Roo::Base
         @header_line = line_no
         return return_headers ? headers : line_no
       elsif line_no > 100
-        fail "Couldn't find header row."
+        raise Roo::HeaderRowNotFoundError
       end
     end
-    fail "Couldn't find header row."
+    raise Roo::HeaderRowNotFoundError
   end
 
   protected
@@ -680,47 +680,47 @@ class Roo::Base
 
   # The content of a cell in the csv output
   def cell_to_csv(row, col, sheet)
-    if empty?(row, col, sheet)
-      ''
-    else
-      onecell = cell(row, col, sheet)
-
-      case celltype(row, col, sheet)
-      when :string
+    return '' if empty?(row, col, sheet)
+
+    onecell = cell(row, col, sheet)
+
+    case celltype(row, col, sheet)
+    when :string
+      %("#{onecell.gsub('"', '""')}") unless onecell.empty?
+    when :boolean
+      # TODO: this only works for excelx
+      onecell = self.sheet_for(sheet).cells[[row, col]].formatted_value
+      %("#{onecell.gsub('"', '""').downcase}")
+    when :float, :percentage
+      if onecell == onecell.to_i
+        onecell.to_i.to_s
+      else
+        onecell.to_s
+      end
+    when :formula
+      case onecell
+      when String
         %("#{onecell.gsub('"', '""')}") unless onecell.empty?
-      when :boolean
-        %("#{onecell.gsub('"', '""').downcase}")
-      when :float, :percentage
+      when Float
         if onecell == onecell.to_i
           onecell.to_i.to_s
         else
           onecell.to_s
         end
-      when :formula
-        case onecell
-        when String
-          %("#{onecell.gsub('"', '""')}") unless onecell.empty?
-        when Float
-          if onecell == onecell.to_i
-            onecell.to_i.to_s
-          else
-            onecell.to_s
-          end
-        when DateTime
-          onecell.to_s
-        else
-          fail "unhandled onecell-class #{onecell.class}"
-        end
-      when :date, :datetime
+      when DateTime
         onecell.to_s
-      when :time
-        integer_to_timestring(onecell)
-      when :link
-        %("#{onecell.url.gsub('"', '""')}")
       else
-        fail "unhandled celltype #{celltype(row, col, sheet)}"
-      end || ''
-    end
+        fail "unhandled onecell-class #{onecell.class}"
+      end
+    when :date, :datetime
+      onecell.to_s
+    when :time
+      integer_to_timestring(onecell)
+    when :link
+      %("#{onecell.url.gsub('"', '""')}")
+    else
+      fail "unhandled celltype #{celltype(row, col, sheet)}"
+    end || ''
   end
 
   # converts an integer value to a time string like '02:05:06'

data/lib/roo/errors.rb

@@ -0,0 +1,9 @@
+module Roo
+  # A base error class for Roo. Most errors thrown by Roo should inherit from
+  # this class.
+  class Error < StandardError; end
+
+  # Raised when Roo cannot find a header row that matches the given column
+  # name(s).
+  class HeaderRowNotFoundError < Error; end
+end

data/lib/roo/excelx.rb

@@ -2,9 +2,13 @@ require 'nokogiri'
 require 'zip/filesystem'
 require 'roo/link'
 require 'roo/utils'
+require 'forwardable'
 
 module Roo
   class Excelx < Roo::Base
+    extend Forwardable
+
+    require 'roo/excelx/shared'
     require 'roo/excelx/workbook'
     require 'roo/excelx/shared_strings'
     require 'roo/excelx/styles'
@@ -13,68 +17,10 @@ module Roo
     require 'roo/excelx/relationships'
     require 'roo/excelx/comments'
     require 'roo/excelx/sheet_doc'
-    
-    module Format
-      EXCEPTIONAL_FORMATS = {
-        'h:mm am/pm' => :date,
-        'h:mm:ss am/pm' => :date
-      }
-
-      STANDARD_FORMATS = {
-        0 => 'General'.freeze,
-        1 => '0'.freeze,
-        2 => '0.00'.freeze,
-        3 => '#,##0'.freeze,
-        4 => '#,##0.00'.freeze,
-        9 => '0%'.freeze,
-        10 => '0.00%'.freeze,
-        11 => '0.00E+00'.freeze,
-        12 => '# ?/?'.freeze,
-        13 => '# ??/??'.freeze,
-        14 => 'mm-dd-yy'.freeze,
-        15 => 'd-mmm-yy'.freeze,
-        16 => 'd-mmm'.freeze,
-        17 => 'mmm-yy'.freeze,
-        18 => 'h:mm AM/PM'.freeze,
-        19 => 'h:mm:ss AM/PM'.freeze,
-        20 => 'h:mm'.freeze,
-        21 => 'h:mm:ss'.freeze,
-        22 => 'm/d/yy h:mm'.freeze,
-        37 => '#,##0 ;(#,##0)'.freeze,
-        38 => '#,##0 ;[Red](#,##0)'.freeze,
-        39 => '#,##0.00;(#,##0.00)'.freeze,
-        40 => '#,##0.00;[Red](#,##0.00)'.freeze,
-        45 => 'mm:ss'.freeze,
-        46 => '[h]:mm:ss'.freeze,
-        47 => 'mmss.0'.freeze,
-        48 => '##0.0E+0'.freeze,
-        49 => '@'.freeze
-      }
-
-      def to_type(format)
-        format = format.to_s.downcase
-        if (type = EXCEPTIONAL_FORMATS[format])
-          type
-        elsif format.include?('#')
-          :float
-        elsif !format.match(/d+(?![\]])/).nil? || format.include?('y')
-          if format.include?('h') || format.include?('s')
-            :datetime
-          else
-            :date
-          end
-        elsif format.include?('h') || format.include?('s')
-          :time
-        elsif format.include?('%')
-          :percentage
-        else
-          :float
-        end
-      end
-
-      module_function :to_type
-    end
+    require 'roo/excelx/coordinate'
+    require 'roo/excelx/format'
 
+    delegate [:styles, :workbook, :shared_strings, :rels_files, :sheet_files, :comments_files] => :@shared
     ExceedsMaxError = Class.new(StandardError)
 
     # initialization and opening of a spreadsheet file
@@ -94,9 +40,8 @@ module Roo
       end
 
       @tmpdir = make_tmpdir(basename, options[:tmpdir_root])
+      @shared = Shared.new(@tmpdir)
       @filename = local_filename(filename_or_stream, @tmpdir, packed)
-      @comments_files = []
-      @rels_files = []
       process_zipfile(@filename || filename_or_stream)
 
       @sheet_names = workbook.sheets.map do |sheet|
@@ -106,7 +51,7 @@ module Roo
       end.compact
       @sheets = []
       @sheets_by_name = Hash[@sheet_names.map.with_index do |sheet_name, n|
-        @sheets[n] = Sheet.new(sheet_name, @rels_files[n], @sheet_files[n], @comments_files[n], styles, shared_strings, workbook, sheet_options)
+        @sheets[n] = Sheet.new(sheet_name, @shared, n, sheet_options)
         [sheet_name, @sheets[n]]
       end]
 
@@ -187,7 +132,7 @@ module Roo
     def set(row, col, value, sheet = nil) #:nodoc:
       key = normalize(row, col)
       cell_type = cell_type_by_value(value)
-      sheet_for(sheet).cells[key] = Cell.new(value, cell_type, nil, cell_type, value, nil, nil, nil, Cell::Coordinate.new(row, col))
+      sheet_for(sheet).cells[key] = Cell.new(value, cell_type, nil, cell_type, value, nil, nil, nil, Coordinate.new(row, col))
     end
 
     # Returns the formula at (row,col).
@@ -239,14 +184,14 @@ module Roo
     # Note: this is only available within the Excelx class
     def excelx_type(row, col, sheet = nil)
       key = normalize(row, col)
-      safe_send(sheet_for(sheet).cells[key], :excelx_type)
+      safe_send(sheet_for(sheet).cells[key], :cell_type)
     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)
       key = normalize(row, col)
-      safe_send(sheet_for(sheet).cells[key], :excelx_value)
+      safe_send(sheet_for(sheet).cells[key], :cell_value)
     end
 
     # returns the internal format of an excel cell
@@ -259,8 +204,8 @@ module Roo
       sheet = sheet_for(sheet)
       key = normalize(row, col)
       cell = sheet.cells[key]
-      !cell || !cell.value || (cell.type == :string && cell.value.empty?) \
-      || (row < sheet.first_row || row > sheet.last_row || col < sheet.first_column || col > sheet.last_column)
+      !cell || cell.empty? || (cell.type == :string && cell.value.empty?) ||
+        (row < sheet.first_row || row > sheet.last_row || col < sheet.first_column || col > sheet.last_column)
     end
 
     # shows the internal representation of all cells
@@ -404,6 +349,7 @@ module Roo
         name = sheets[id]
         entry = entries.find { |e| e.name =~ /#{name}$/ }
         path = "#{tmpdir}/roo_sheet#{i + 1}"
+        sheet_files << path
         @sheet_files << path
         entry.extract(path)
       end
@@ -457,19 +403,21 @@ module Roo
           #       sheet's comment file is in the sheet1.xml.rels file. SEE
           #       ECMA-376 12.3.3 in "Ecma Office Open XML Part 1".
           nr = Regexp.last_match[1].to_i
-          @comments_files[nr - 1] = "#{@tmpdir}/roo_comments#{nr}"
+          comments_files[nr - 1] = "#{@tmpdir}/roo_comments#{nr}"
         when /sheet([0-9]+).xml.rels$/
           # FIXME: Roo seems to use sheet[\d].xml.rels for hyperlinks only, but
           #        it also stores the location for sharedStrings, comments,
           #        drawings, etc.
           nr = Regexp.last_match[1].to_i
-          @rels_files[nr - 1] = "#{@tmpdir}/roo_rels#{nr}"
+          rels_files[nr - 1] = "#{@tmpdir}/roo_rels#{nr}"
         end
 
         entry.extract(path) if path
       end
     end
 
+    # NOTE: To reduce memory, styles, shared_strings, workbook can be class
+    #       variables in a Shared module.
     def styles
       @styles ||= Styles.new(File.join(@tmpdir, 'roo_styles.xml'))
     end

data/lib/roo/excelx/cell.rb

@@ -1,4 +1,12 @@
 require 'date'
+require 'roo/excelx/cell/base'
+require 'roo/excelx/cell/boolean'
+require 'roo/excelx/cell/datetime'
+require 'roo/excelx/cell/date'
+require 'roo/excelx/cell/empty'
+require 'roo/excelx/cell/number'
+require 'roo/excelx/cell/string'
+require 'roo/excelx/cell/time'
 
 module Roo
   class Excelx
@@ -6,7 +14,9 @@ module Roo
       attr_reader :type, :formula, :value, :excelx_type, :excelx_value, :style, :hyperlink, :coordinate
       attr_writer :value
 
+      # DEPRECATED: Please use Cell.create_cell instead.
       def initialize(value, type, formula, excelx_type, excelx_value, style, hyperlink, base_date, coordinate)
+        warn '[DEPRECATION] `Cell.new` is deprecated.  Please use `Cell.create_cell` instead.'
         @type = type
         @formula = formula
         @base_date = base_date if [:date, :datetime].include?(@type)
@@ -29,10 +39,29 @@ module Roo
         end
       end
 
+      def self.create_cell(type, *values)
+        case type
+        when :string
+          Cell::String.new(*values)
+        when :boolean
+          Cell::Boolean.new(*values)
+        when :number
+          Cell::Number.new(*values)
+        when :date
+          Cell::Date.new(*values)
+        when :datetime
+          Cell::DateTime.new(*values)
+        when :time
+          Cell::Time.new(*values)
+        end
+      end
+
+      # Deprecated: use Roo::Excelx::Coordinate instead.
       class Coordinate
         attr_accessor :row, :column
 
         def initialize(row, column)
+          warn '[DEPRECATION] `Roo::Excel::Cell::Coordinate` is deprecated.  Please use `Roo::Excelx::Coordinate` instead.'
           @row, @column = row, column
         end
       end
@@ -57,20 +86,20 @@ module Roo
       def create_date(date)
         yyyy, mm, dd = date.strftime('%Y-%m-%d').split('-')
 
-        Date.new(yyyy.to_i, mm.to_i, dd.to_i)
+        ::Date.new(yyyy.to_i, mm.to_i, dd.to_i)
       end
 
       def create_datetime(date)
         datetime_string = date.strftime('%Y-%m-%d %H:%M:%S.%N')
         t = round_datetime(datetime_string)
 
-        DateTime.civil(t.year, t.month, t.day, t.hour, t.min, t.sec)
+        ::DateTime.civil(t.year, t.month, t.day, t.hour, t.min, t.sec)
       end
 
       def round_datetime(datetime_string)
         /(?<yyyy>\d+)-(?<mm>\d+)-(?<dd>\d+) (?<hh>\d+):(?<mi>\d+):(?<ss>\d+.\d+)/ =~ datetime_string
 
-        Time.new(yyyy.to_i, mm.to_i, dd.to_i, hh.to_i, mi.to_i, ss.to_r).round(0)
+        ::Time.new(yyyy.to_i, mm.to_i, dd.to_i, hh.to_i, mi.to_i, ss.to_r).round(0)
       end
     end
   end