roo 2.0.1 → 2.1.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: db86b4586034783940303cfd17a7ee5c60277a51
-  data.tar.gz: 2cf9222f2f2759f164164f745c72abe6b901df98
+  metadata.gz: 53d37d09a786f9536fd5bb00c7cacf9888c8b032
+  data.tar.gz: 284974b328b26668c58e3b952a12b8777439037f
 SHA512:
-  metadata.gz: 442e05a8bffdc839c45a8684e8882f4327098387617b76027b6b5c0543c2a7b4be284c511c35f8333c9bc1147e31350d1f6f67174fc62e5b7fc45520ab0ad307
-  data.tar.gz: 4ba67602e5772cae52f41ea0357ddd6835d25c35be57797307a6194a46ad977d7922b884b4fa826c91efd8533c6f3e56995e8cb0a13146a32e4a3ee7d5e42ae9
+  metadata.gz: 4b0e6d58d52600710fc5feb49e7483fc81e4bca65fa9013a3fa1004f5526107392b02f982ae981676544e301056a0246b4b85e38d3959565174c247ce49695a9
+  data.tar.gz: 5c3beb6580bece8aa730d1f29b6316889ebb7ce5a7003d8f654edcb1c9800a9f939021fd0d88749489ffdee6ebf496739664e5fede43aeb0067a2a16ad46f2f5

data/.travis.yml

@@ -10,4 +10,5 @@ matrix:
   allow_failures:
     - rvm: ruby-head
     - rvm: jruby-19mode
+    - rvm: rbx-2
 bundler_args: --without local_development

data/CHANGELOG.md

@@ -1,3 +1,18 @@
+## [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)
+
+## [2.1.0] - 2015-07-18
+### Added
+- Added support for Excel 2007 `xlsm` files. [#232](https://github.com/roo-rb/roo/pull/232)
+- Roo::Excelx returns an enumerator when calling each_row_streaming without a block. [#224](https://github.com/roo-rb/roo/pull/224)
+- Returns an enumerator when calling `each` without a block. [#219](https://github.com/roo-rb/roo/pull/219)
+
+### Fixed
+- Removed tabs and windows CRLF. [#235](https://github.com/roo-rb/roo/pull/235), [#234](https://github.com/roo-rb/roo/pull/234)
+- Fixed Regexp to only check for valid URI's when opening a spreadsheet. [#229](https://github.com/roo-rb/roo/pull/228)
+- Open streams in Roo:Excelx correctly. [#222](https://github.com/roo-rb/roo/pull/222)
+
 ## [2.0.1] - 2015-06-01
 ### Added
 - Return an enumerator when calling '#each' without a block [#219](https://github.com/roo-rb/roo/pull/219)
@@ -464,7 +479,7 @@
 
 ## [0.2.4] - 2007-06-16
 ### Fixed
-- ID 11605	Two cols with same value: crash roo (openoffice version only)
+- ID 11605  Two cols with same value: crash roo (openoffice version only)
 
 ## [0.2.3] - 2007-06-02
 ### Changed / Added

data/README.md

@@ -3,16 +3,13 @@
 [![Build Status](https://img.shields.io/travis/roo-rb/roo.svg?style=flat-square)](https://travis-ci.org/roo-rb/roo) [![Code Climate](https://img.shields.io/codeclimate/github/roo-rb/roo.svg?style=flat-square)](https://codeclimate.com/github/roo-rb/roo) [![Coverage Status](https://img.shields.io/coveralls/roo-rb/roo.svg?style=flat-square)](https://coveralls.io/r/roo-rb/roo) [![Gem Version](https://img.shields.io/gem/v/roo.svg?style=flat-square)](https://rubygems.org/gems/roo)
 
 Roo implements read access for all common spreadsheet types. It can handle:
-
-* Excelx
-* OpenOffice / LibreOffice
+* Excel 2007 - 2013 formats (xlsx, xlsm)
+* LibreOffice / OpenOffice.org formats (ods)
 * CSV
+* Excel 97, Excel 2002 XML, and Excel 2003 XML formats when using the [roo-xls](https://github.com/roo-rb/roo-xls) gem (xls, xml)
+* Google spreadsheets with read/write access when using [roo-google](https://github.com/roo-rb/roo-google)
 
-## Additional Libraries
-
-In addition, the [roo-xls](https://github.com/roo-rb/roo-xls) and [roo-google](https://github.com/roo-rb/roo-google) gems exist to extend Roo to support reading classic Excel formats (i.e. `.xls` and ``Excel2003XML``) and read/write access for Google spreadsheets.
-
-# #Installation
+## Installation
 
 Install as a gem
 
@@ -21,7 +18,7 @@ Install as a gem
 Or add it to your Gemfile
 
 ```ruby
-gem 'roo', '~> 2.0.0'
+gem 'roo', '~> 2.1.0'
 ```
 ## Usage
 
@@ -102,7 +99,7 @@ s.cell(1,'A',s.sheets[1])
 ```
 
 #### Querying a spreadsheet
-Use ``each`` with a ``block`` to iterate over each row.
+Use ``each`` to iterate over each row.
 
 If each is given a hash with the names of some columns, then each will generate a hash with the columns supplied for each row.
 
@@ -143,7 +140,7 @@ sheet.to_xml
 sheet.to_yaml
 ```
 
-### Excel (xlsx) Support
+### Excel (xlsx and xlsm) Support
 
 Stream rows from an Excelx spreadsheet.
 

data/lib/roo/base.rb

@@ -91,7 +91,7 @@ class Roo::Base
       first_column = [first_column, key.last.to_i].min
       last_column = [last_column, key.last.to_i].max
     end if @cell[sheet]
-    {first_row: first_row, first_column: first_column, last_row: last_row, last_column: last_column}
+    { first_row: first_row, first_column: first_column, last_row: last_row, last_column: last_column }
   end
 
   %w(first_row last_row first_column last_column).each do |key|
@@ -117,22 +117,23 @@ class Roo::Base
     result = "--- \n"
     from_row.upto(to_row) do |row|
       from_column.upto(to_column) do |col|
-        unless empty?(row, col, sheet)
-          result << "cell_#{row}_#{col}: \n"
-          prefix.each do|k, v|
-            result << "  #{k}: #{v} \n"
-          end
-          result << "  row: #{row} \n"
-          result << "  col: #{col} \n"
-          result << "  celltype: #{celltype(row, col, sheet)} \n"
-          value = cell(row, col, sheet)
-          if celltype(row, col, sheet) == :time
-            value = integer_to_timestring(value)
-          end
-          result << "  value: #{value} \n"
+        next if empty?(row, col, sheet)
+
+        result << "cell_#{row}_#{col}: \n"
+        prefix.each do|k, v|
+          result << "  #{k}: #{v} \n"
+        end
+        result << "  row: #{row} \n"
+        result << "  col: #{col} \n"
+        result << "  celltype: #{celltype(row, col, sheet)} \n"
+        value = cell(row, col, sheet)
+        if celltype(row, col, sheet) == :time
+          value = integer_to_timestring(value)
         end
+        result << "  value: #{value} \n"
       end
     end
+
     result
   end
 
@@ -170,7 +171,7 @@ class Roo::Base
   end
 
   def inspect
-    "<##{ self.class }:#{ self.object_id.to_s(8) } #{ self.instance_variables.join(' ') }>"
+    "<##{self.class}:#{object_id.to_s(8)} #{instance_variables.join(' ')}>"
   end
 
   # find a row either by row number or a condition
@@ -217,7 +218,7 @@ class Roo::Base
     row, col = normalize(row, col)
     cell_type = cell_type_by_value(value)
     set_value(row, col, value, sheet)
-    set_type(row, col, cell_type , sheet)
+    set_type(row, col, cell_type, sheet)
   end
 
   def cell_type_by_value(value)
@@ -225,7 +226,7 @@ class Roo::Base
     when Fixnum then :float
     when String, Float then :string
     else
-      raise ArgumentError, "Type for #{value} not set"
+      fail ArgumentError, "Type for #{value} not set"
     end
   end
 
@@ -256,13 +257,13 @@ class Roo::Base
       sheets.each do|sheet|
         self.default_sheet = sheet
         result << 'Sheet ' + n.to_s + ":\n"
-        unless first_row
-          result << '  - empty -'
-        else
+        if first_row
           result << "  First row: #{first_row}\n"
           result << "  Last row: #{last_row}\n"
           result << "  First column: #{::Roo::Utils.number_to_letter(first_column)}\n"
           result << "  Last column: #{::Roo::Utils.number_to_letter(last_column)}"
+        else
+          result << '  - empty -'
         end
         result << "\n" if sheet != sheets.last
         n += 1
@@ -282,12 +283,12 @@ class Roo::Base
               # sonst gibt es Fehler bei leeren Blaettern
               first_row.upto(last_row) do |row|
                 first_column.upto(last_column) do |col|
-                  unless empty?(row, col)
-                    x.cell(cell(row, col),
+                  next if empty?(row, col)
+
+                  x.cell(cell(row, col),
                            row: row,
                            column: col,
                            type: celltype(row, col))
-                  end
                 end
               end
             end
@@ -318,7 +319,7 @@ class Roo::Base
   # access different worksheets by calling spreadsheet.sheet(1)
   # or spreadsheet.sheet('SHEETNAME')
   def sheet(index, name = false)
-    self.default_sheet = String === index ? index : sheets[index]
+    self.default_sheet = index.is_a?(::String) ? index : sheets[index]
     name ? [default_sheet, self] : self
   end
 
@@ -352,25 +353,23 @@ class Roo::Base
   # control characters and white spaces around columns
 
   def each(options = {})
-    if block_given?
-      if options.empty?
-        1.upto(last_row) do |line|
-          yield row(line)
-        end
-      else
-        clean_sheet_if_need(options)
-        search_or_set_header(options)
-        headers = @headers ||
-                  Hash[(first_column..last_column).map do |col|
-                    [cell(@header_line, col), col]
-                  end]
-
-        @header_line.upto(last_row) do |line|
-          yield(Hash[headers.map { |k, v| [k, cell(line, v)] }])
-        end
+    return to_enum(:each, options) unless block_given?
+
+    if options.empty?
+      1.upto(last_row) do |line|
+        yield row(line)
       end
     else
-      to_enum(:each, options)
+      clean_sheet_if_need(options)
+      search_or_set_header(options)
+      headers = @headers ||
+                Hash[(first_column..last_column).map do |col|
+                  [cell(@header_line, col), col]
+                end]
+
+      @header_line.upto(last_row) do |line|
+        yield(Hash[headers.map { |k, v| [k, cell(line, v)] }])
+      end
     end
   end
 
@@ -401,30 +400,32 @@ class Roo::Base
 
   protected
 
-  def file_type_check(filename, ext, name, warning_level, packed = nil)
+  def file_type_check(filename, exts, name, warning_level, packed = nil)
     if packed == :zip
-      # lalala.ods.zip => lalala.ods
-      # hier wird KEIN unzip gemacht, sondern nur der Name der Datei
-      # getestet, falls es eine gepackte Datei ist.
+      # spreadsheet.ods.zip => spreadsheet.ods
+      # Decompression is not performed here, only the 'zip' extension
+      # is removed from the file.
       filename = File.basename(filename, File.extname(filename))
     end
 
-    if uri?(filename) && qs_begin = filename.rindex('?')
+    if uri?(filename) && (qs_begin = filename.rindex('?'))
       filename = filename[0..qs_begin - 1]
     end
-    if File.extname(filename).downcase != ext
-      case warning_level
-      when :error
-        warn file_type_warning_message(filename, ext)
-        fail TypeError, "#{filename} is not #{name} file"
-      when :warning
-        warn "are you sure, this is #{name} spreadsheet file?"
-        warn file_type_warning_message(filename, ext)
-      when :ignore
-        # ignore
-      else
-        fail "#{warning_level} illegal state of file_warning"
-      end
+    exts = Array(exts)
+
+    return if exts.include?(File.extname(filename).downcase)
+
+    case warning_level
+    when :error
+      warn file_type_warning_message(filename, exts)
+      fail TypeError, "#{filename} is not #{name} file"
+    when :warning
+      warn "are you sure, this is #{name} spreadsheet file?"
+      warn file_type_warning_message(filename, exts)
+    when :ignore
+      # ignore
+    else
+      fail "#{warning_level} illegal state of file_warning"
     end
   end
 
@@ -475,16 +476,18 @@ class Roo::Base
     return if is_stream?(filename)
     filename = download_uri(filename, tmpdir) if uri?(filename)
     filename = unzip(filename, tmpdir) if packed == :zip
-    unless File.file?(filename)
-      fail IOError, "file #{filename} does not exist"
-    end
+
+    fail IOError, "file #{filename} does not exist" unless File.file?(filename)
+
     filename
   end
 
-  def file_type_warning_message(filename, ext)
-    "use #{Roo::CLASS_FOR_EXTENSION.fetch(ext.sub('.', '').to_sym)}.new to handle #{ext} spreadsheet files. This has #{File.extname(filename).downcase}"
+  def file_type_warning_message(filename, exts)
+    *rest, last_ext = exts
+    ext_list = rest.any? ? "#{rest.join(', ')} or #{last_ext}" : last_ext
+    "use #{Roo::CLASS_FOR_EXTENSION.fetch(last_ext.sub('.', '').to_sym)}.new to handle #{ext_list} spreadsheet files. This has #{File.extname(filename).downcase}"
   rescue KeyError
-    raise "unknown file type: #{ext}"
+    raise "unknown file types: #{ext_list}"
   end
 
   def find_by_row(row_index)
@@ -533,11 +536,8 @@ class Roo::Base
   end
 
   def make_tmpdir(prefix = nil, root = nil, &block)
-    prefix = if prefix
-      TEMP_PREFIX + prefix
-    else
-      TEMP_PREFIX
-    end
+    prefix = "#{TEMP_PREFIX}#{prefix}"
+
     ::Dir.mktmpdir(prefix, root || ENV['ROO_TMP'], &block).tap do |result|
       block_given? || track_tmpdir!(result)
     end
@@ -585,9 +585,9 @@ class Roo::Base
         fail ArgumentError
       end
     end
-    if col.is_a?(::String)
-      col = ::Roo::Utils.letter_to_number(col)
-    end
+
+    col = ::Roo::Utils.letter_to_number(col) if col.is_a?(::String)
+
     [row, col]
   end
 
@@ -638,7 +638,7 @@ class Roo::Base
         fail RangeError, "sheet index #{sheet} not found"
       end
     when String
-      unless sheets.include? sheet
+      unless sheets.include?(sheet)
         fail RangeError, "sheet '#{sheet}' not found"
       end
     else
@@ -667,14 +667,14 @@ class Roo::Base
   # parameter is nil the output goes to STDOUT
   def write_csv_content(file = nil, sheet = nil, separator = ',')
     file ||= STDOUT
-    if first_row(sheet) # sheet is not empty
-      1.upto(last_row(sheet)) do |row|
-        1.upto(last_column(sheet)) do |col|
-          file.print(separator) if col > 1
-          file.print cell_to_csv(row, col, sheet)
-        end
-        file.print("\n")
-      end # sheet not empty
+    return unless first_row(sheet) # The sheet is empty
+
+    1.upto(last_row(sheet)) do |row|
+      1.upto(last_column(sheet)) do |col|
+        file.print(separator) if col > 1
+        file.print cell_to_csv(row, col, sheet)
+      end
+      file.print("\n")
     end
   end
 
@@ -726,9 +726,9 @@ class Roo::Base
   # converts an integer value to a time string like '02:05:06'
   def integer_to_timestring(content)
     h = (content / 3600.0).floor
-    content = content - h * 3600
+    content -= h * 3600
     m = (content / 60.0).floor
-    content = content - m * 60
+    content -= m * 60
     s = content
     sprintf('%02d:%02d:%02d', h, m, s)
   end

data/lib/roo/constants.rb

@@ -0,0 +1,5 @@
+module Roo
+  ROO_EXCEL_NOTICE   = "Excel support has been extracted to roo-xls due to its dependency on the GPL'd spreadsheet gem. Install roo-xls to use Roo::Excel.".freeze
+  ROO_EXCELML_NOTICE = "Excel SpreadsheetML support has been extracted to roo-xls. Install roo-xls to use Roo::Excel2003XML.".freeze
+  ROO_GOOGLE_NOTICE = "Google support has been extracted to roo-google. Install roo-google to use Roo::Google.".freeze
+end

data/lib/roo/excelx/shared_strings.rb

@@ -13,9 +13,19 @@ module Roo
 
       private
 
+      def fix_invalid_shared_strings(doc)
+        invalid = { '_x000D_'  => "\n" }
+        xml = doc.to_s
+
+        if xml[/#{invalid.keys.join('|')}/]
+          @doc = ::Nokogiri::XML(xml.gsub(/#{invalid.keys.join('|')}/, invalid))
+        end
+      end
+
       def extract_shared_strings
         return [] unless doc_exists?
 
+        fix_invalid_shared_strings(doc)
         # read the shared strings xml document
         doc.xpath('/sst/si').map do |si|
           shared_string = ''

data/lib/roo/excelx.rb

@@ -89,7 +89,7 @@ module Roo
       sheet_options[:expand_merged_ranges] = (options[:expand_merged_ranges] || false)
 
       unless is_stream?(filename_or_stream)
-        file_type_check(filename_or_stream, '.xlsx', 'an Excel-xlsx', file_warning, packed)
+        file_type_check(filename_or_stream, %w[.xlsx .xlsm], 'an Excel 2007', file_warning, packed)
         basename = File.basename(filename_or_stream)
       end
 
@@ -321,7 +321,12 @@ module Roo
     # Yield an array of Excelx::Cell
     # Takes options for sheet, pad_cells, and max_rows
     def each_row_streaming(options = {})
-      sheet_for(options.delete(:sheet)).each_row(options) { |row| yield row }
+      sheet = sheet_for(options.delete(:sheet))
+      if block_given?
+        sheet.each_row(options) { |row| yield row }
+      else
+        sheet.to_enum(:each_row, options)
+      end
     end
 
     private
@@ -409,19 +414,13 @@ module Roo
       @sheet_files = []
 
       unless is_stream?(zipfilename_or_stream)
-        process_zipfile_entries Zip::File.open(zipfilename_or_stream).to_a.sort_by(&:name)
+        zip_file = Zip::File.open(zipfilename_or_stream)
       else
-        stream = Zip::InputStream.open zipfilename_or_stream
-        begin
-          entries = []
-          while (entry = stream.get_next_entry)
-            entries << entry
-          end
-          process_zipfile_entries entries
-        ensure
-          stream.close
-        end
+        zip_file = Zip::CentralDirectory.new
+        zip_file.read_from_stream zipfilename_or_stream
       end
+
+      process_zipfile_entries zip_file.to_a.sort_by(&:name)
     end
 
     def process_zipfile_entries(entries)

data/lib/roo/libre_office.rb

@@ -1,5 +1,4 @@
 require 'roo/open_office'
 
 # LibreOffice is just an alias for Roo::OpenOffice class
-class Roo::LibreOffice < Roo::OpenOffice
-end
+Roo::LibreOffice = Roo::OpenOffice