Merge pull request #45 from ms-ati/excel-2003-xml-support-font-colors

[Excel 2003 XML] Add support for font colors + required fixes

README.md

@@ -10,6 +10,11 @@ This library extends Roo to add support for handling class Excel files, includin
 There is no support for formulas in Roo for .xls files - you can get the result
 of a formula but not the formula itself.
 
+## Limitations
+
+Roo::Xls currently doesn't provide support for the following features in Roo:
+* [Option `:expand_merged_ranged => true`](https://github.com/roo-rb/roo#expand_merged_ranges)
+
 ## License
 
 While Roo and Roo::Xls are licensed under the MIT / Expat license, please note that the `spreadsheet` gem [is released under](https://github.com/zdavatz/spreadsheet/blob/master/LICENSE.txt) the GPLv3 license. Please be aware that the author of the `spreadsheet` gem [claims you need a commercial license](http://spreadsheet.ch/2014/10/24/using-ruby-spreadsheet-on-heroku-with-dynos/) to use it as part of a public-facing, closed-source service, an interpretation [at odds with the FSF's intent and interpretation of the license](http://www.gnu.org/licenses/gpl-faq.html#UnreleasedMods). 

lib/roo/xls/excel_2003_xml.rb

@@ -21,7 +21,7 @@ def initialize(filename, options = {})
       end
       @doc = ::Roo::Utils.load_xml(@filename)
     end
-    namespace = @doc.namespaces.select{|xmlns, urn| urn == 'urn:schemas-microsoft-com:office:spreadsheet'}.keys.last
+    namespace = @doc.namespaces.select { |_, urn| urn == 'urn:schemas-microsoft-com:office:spreadsheet' }.keys.last
     @namespace = (namespace.nil? || namespace.empty?) ? 'ss' : namespace.split(':').last
     super(filename, options)
     @formula = {}
@@ -58,7 +58,7 @@ def formula(row, col, sheet = nil)
   alias_method :formula?, :formula
 
   class Font
-    attr_accessor :bold, :italic, :underline
+    attr_accessor :bold, :italic, :underline, :color, :name, :size
 
     def bold?
       @bold == '1'
@@ -191,31 +191,49 @@ def read_cells(sheet = nil)
     validate_sheet!(sheet)
     return if @cells_read[sheet]
     sheet_found = false
+
     @doc.xpath("/#{@namespace}:Workbook/#{@namespace}:Worksheet[@#{@namespace}:Name='#{sheet}']").each do |ws|
       sheet_found = true
-      row = 1
+      column_styles = {}
+
+      # Column Styles
       col = 1
-      column_attributes = {}
-      idx = 0
       ws.xpath("./#{@namespace}:Table/#{@namespace}:Column").each do |c|
-        column_attributes[(idx += 1).to_s] = c['StyleID']
+        skip_to_col = c["#{@namespace}:Index"].to_i
+        col = skip_to_col if skip_to_col > 0
+        col_style_name = c["#{@namespace}:StyleID"]
+        column_styles[col] = col_style_name unless col_style_name.nil?
+        col += 1
       end
+
+      # Rows
+      row = 1
       ws.xpath("./#{@namespace}:Table/#{@namespace}:Row").each do |r|
-        skip_to_row = r['Index'].to_i
+        skip_to_row = r["#{@namespace}:Index"].to_i
         row = skip_to_row if skip_to_row > 0
-        style_name = r['StyleID'] if r['StyleID']
+
+        # Excel uses a 'Span' attribute on a 'Row' to indicate the presence of
+        # empty rows to skip.
+        skip_next_rows = r["#{@namespace}:Span"].to_i
+
+        row_style_name = r["#{@namespace}:StyleID"]
+
+        # Cells
+        col = 1
         r.xpath("./#{@namespace}:Cell").each do |c|
-          skip_to_col = c['Index'].to_i
+          skip_to_col = c["#{@namespace}:Index"].to_i
           col = skip_to_col if skip_to_col > 0
-          if c['StyleID']
-            style_name = c['StyleID']
-          elsif
-            style_name ||= column_attributes[c['Index']]
-          end
+
+          skip_next_cols = c["#{@namespace}:MergeAcross"].to_i
+
+          cell_style_name = c["#{@namespace}:StyleID"]
+          style_name = cell_style_name || row_style_name || column_styles[col]
+
+          # Cell Data
           c.xpath("./#{@namespace}:Data").each do |cell|
             formula = cell['Formula']
             value_type = cell["#{@namespace}:Type"].downcase.to_sym
-            v =  cell.content
+            v = cell.content
             str_v = v
             case value_type
             when :number
@@ -234,10 +252,9 @@ def read_cells(sheet = nil)
             end
             set_cell_values(sheet, col, row, 0, v, value_type, formula, cell, str_v, style_name)
           end
-          col += 1
+          col += (skip_next_cols + 1)
         end
-        row += 1
-        col = 1
+        row += (skip_next_rows + 1)
       end
     end
     unless sheet_found
@@ -248,12 +265,16 @@ def read_cells(sheet = nil)
 
   def read_styles
     @doc.xpath("/#{@namespace}:Workbook/#{@namespace}:Styles/#{@namespace}:Style").each do |style|
-      style_id = style['ID']
-      @style_definitions[style_id] = Roo::Excel2003XML::Font.new
-      if font = style.at_xpath("./#{@namespace}:Font")
-        @style_definitions[style_id].bold = font['Bold']
-        @style_definitions[style_id].italic = font['Italic']
-        @style_definitions[style_id].underline = font['Underline']
+      style_id = style["#{@namespace}:ID"]
+      font = style.at_xpath("./#{@namespace}:Font")
+      unless font.nil?
+        @style_definitions[style_id] = Roo::Excel2003XML::Font.new
+        @style_definitions[style_id].bold      = font["#{@namespace}:Bold"]
+        @style_definitions[style_id].italic    = font["#{@namespace}:Italic"]
+        @style_definitions[style_id].underline = font["#{@namespace}:Underline"]
+        @style_definitions[style_id].color     = font["#{@namespace}:Color"]
+        @style_definitions[style_id].name      = font["#{@namespace}:FontName"]
+        @style_definitions[style_id].size      = font["#{@namespace}:Size"]
       end
     end
   end

spec/lib/roo/xls/excel2003xml_spec.rb

@@ -1,14 +1,204 @@
 require 'spec_helper'
 
-describe Roo::Excel2003XML do
+RSpec.describe Roo::Excel2003XML do
+  subject { instance }
+
+  let(:instance) { described_class.new(path) }
+  let(:path) { File.join(test_files_dir, test_file) }
+  let(:test_files_dir) { 'test/files' }
+
   describe '.new' do
     context 'with an xml file' do
-      let(:path) { 'test/files/datetime.xml' }
+      let(:test_file) { 'datetime.xml' }
+
+      it 'loads the file without error' do
+        expect { subject }.to_not raise_error
+      end
+    end
+  end
+
+  describe '#cell' do
+    subject { super().cell(cell[:row], cell[:col]) }
+
+    before(:each) { instance.default_sheet = instance.sheets.first }
+
+    context 'with merged cells' do
+      # See font_colors_screenshot_in_Mac_Excel_16.10.png for a screenshot of
+      # this how this file looks in Mac Excel 16.10.
+      let(:test_file) { 'font_colors.xml' }
+
+      context 'the top-left cell in a merged cell' do
+        let(:cell) { { :row => 1, :col => 'A' } }
+
+        it 'returns the contents of the merged cell' do
+          is_expected.to eq 'Roo::Xls Test of Font Colors'
+        end
+      end
+
+      context 'the cell to the right of the top-left cell in a merged cell' do
+        let(:cell) { { :row => 2, :col => 'A' } }
+        it { is_expected.to be_nil }
+      end
+
+      context 'the cell below the top-left cell in a merged cell' do
+        let(:cell) { { :row => 2, :col => 'B' } }
+        it { is_expected.to be_nil }
+      end
+
+      context 'the first cell to the right of an entire merged cell' do
+        let(:cell) { { :row => 1, :col => 'K' } }
+
+        it 'returns the expected contents' do
+          is_expected.to eq 'This entire COLUMN should be ITALIC and GREEN'
+        end
+      end
+
+      context 'the first cell below an entire merged cell' do
+        let(:cell) { { :row => 6, :col => 'A' } }
+
+        it 'returns the expected contents' do
+          is_expected.to eq '(The above should be font "Courier New", size 24)'
+        end
+      end
+    end
+  end
+
+  describe '#font' do
+    subject { super().font(cell[:row], cell[:col]) }
+
+    before(:each) { instance.default_sheet = instance.sheets.first }
+
+    # See font_colors_screenshot_in_Mac_Excel_16.10.png for a screenshot of
+    # this how this file looks in Mac Excel 16.10.
+    let(:test_file) { 'font_colors.xml' }
+
+    let(:default_attrs) do
+      {
+        :name       => 'Arial',
+        :size       => '12',
+        :color      => '#000000',
+        :bold?      => false,
+        :italic?    => false,
+        :underline? => false
+      }
+    end
+
+    let(:expected_attrs) { default_attrs }
+
+    context 'with no font styling' do
+      let(:cell) { { :row => 6, :col => 'A' } }
+
+      it 'returns default font attributes' do
+        is_expected.to have_attributes(default_attrs)
+      end
+    end
+
+    context 'with styling set on an individual cell' do
+      context 'when set font name and size' do
+        let(:cell) { { :row => 1, :col => 'A' } }
+
+        it 'returns expected font attributes including name and size' do
+          expects = default_attrs.merge({ :name => 'Courier New', :size => '24' })
+          is_expected.to have_attributes(expects)
+        end
+      end
+
+      context 'when colored BLACK' do
+        let(:cell) { { :row => 7, :col => 'A' } }
+
+        it 'returns default font attributes (which include black)' do
+          is_expected.to have_attributes(default_attrs)
+        end
+      end
+
+      context 'when colored RED' do
+        let(:cell) { { :row => 8, :col => 'A' } }
+
+        it 'returns defaults plus red color' do
+          expects = default_attrs.merge({ :color => '#FF0000' })
+          is_expected.to have_attributes(expects)
+        end
+      end
+
+      context 'when colored BLUE' do
+        let(:cell) { { :row => 9, :col => 'A' } }
+
+        it 'returns defaults plus blue color' do
+          expects = default_attrs.merge({ :color => '#0066CC' })
+          is_expected.to have_attributes(expects)
+        end
+      end
+
+      context 'when BOLD' do
+        let(:cell) { { :row => 10, :col => 'A' } }
+
+        it 'returns defaults plus bold style' do
+          # somehow in Excel, this ended up "no color" rather than black...
+          expects = default_attrs.merge({ :bold? => true, :color => nil })
+          is_expected.to have_attributes(expects)
+        end
+      end
+
+      context 'when ITALIC' do
+        let(:cell) { { :row => 11, :col => 'A' } }
+
+        it 'returns defaults plus italic style' do
+          # somehow in Excel, this ended up "no color" rather than black...
+          expects = default_attrs.merge({ :italic? => true, :color => nil })
+          is_expected.to have_attributes(expects)
+        end
+      end
+
+      context 'when UNDERLINED' do
+        let(:cell) { { :row => 12, :col => 'A' } }
+
+        it 'returns defaults plus underlined style' do
+          # somehow in Excel, this ended up "no color" rather than black...
+          expects = default_attrs.merge({ :underline? => true, :color => nil })
+          is_expected.to have_attributes(expects)
+        end
+      end
+
+      context 'when BOLD, ITALIC, UNDERLINED, and colored PURPLE' do
+        let(:cell) { { :row => 13, :col => 'A' } }
+
+        it 'returns defaults plus bold, italic, underlined, and purple color' do
+          expects = default_attrs.merge({
+                                          :color      => '#666699',
+                                          :bold?      => true,
+                                          :italic?    => true,
+                                          :underline? => true
+                                        })
+          is_expected.to have_attributes(expects)
+        end
+      end
+    end
+
+    context 'with styling set on an entire row' do
+      let(:row_style) do
+        default_attrs.merge({ :color => '#ED7D31', :bold? => true })
+      end
+
+      context 'when no cell styling' do
+        let(:cell) { { :row => 14, :col => 'L' } }
+
+        it 'returns the row style' do
+          is_expected.to have_attributes(row_style)
+        end
+      end
+    end
+
+    context 'with styling set on an entire column' do
+      let(:col_style) do
+        default_attrs.merge({ :color => '#00FF00', :italic? => true })
+      end
+
+      context 'when no cell styling' do
+        let(:cell) { { :row => 20, :col => 'K' } }
 
-      it 'loads the file' do
-        expect {
-          Roo::Excel2003XML.new(path)
-        }.to_not raise_error
+        it 'returns the column style' do
+          is_expected.to have_attributes(col_style)
+        end
       end
     end
   end