All the fixes from the Creek forks out there, and more (#47)

* Update nokogiri version

Needed to upgrade to Ruby 2.4

* Updated rspec version so tests can pass.

* Change to use rspec 3 syntax

* Convert specs to RSpec 3.6.0 syntax with Transpec

This conversion is done by Transpec 3.3.0 with the following command:
    transpec

* 25 conversions
    from: obj.should
      to: expect(obj).to

* 22 conversions
    from: == expected
      to: eq(expected)

* 1 conversion
    from: obj.should_not
      to: expect(obj).not_to

For more details: https://github.com/yujinakayama/transpec#supported-conversions

* Added logic to contend with the default namespace having a prefix

Added logic to contend with the default namespace having a prefix. Only
addressed on the 'book'

* Handle sheets files already containing xl/ prefix

Sheets generated by certain libraries already prefix xl/

* Bump httparty to 0.15 and Ruby to >= 2.0.0

Plenty of bugfixes were committed to httparty since 0.14
Some of those fixes address very serious bugs and memory leaks
Httparty 0.15 requires Ruby >= 2.0.0, and now so does this gem

* support for inlineStr cell type

Author: Simon Grondin

creek.gemspec

@@ -18,14 +18,14 @@ Gem::Specification.new do |spec|
   spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
   spec.require_paths = ["lib"]
 
-  spec.required_ruby_version = '>= 1.9.2'
+  spec.required_ruby_version = '>= 2.0.0'
 
   spec.add_development_dependency "bundler", "~> 1.3"
   spec.add_development_dependency "rake"
-  spec.add_development_dependency 'rspec', '~> 2.13.0'
+  spec.add_development_dependency 'rspec', '~> 3.6.0'
   spec.add_development_dependency 'pry'
 
-  spec.add_dependency 'nokogiri', '~> 1.6.0'
+  spec.add_dependency 'nokogiri', '~> 1.7.0'
   spec.add_dependency 'rubyzip', '>= 1.0.0'
-  spec.add_dependency 'httparty', '~> 0.14.0'
+  spec.add_dependency 'httparty', '~> 0.15.5'
 end

lib/creek/book.rb

@@ -34,9 +34,18 @@ def initialize path, options = {}
     def sheets
       doc = @files.file.open "xl/workbook.xml"
       xml = Nokogiri::XML::Document.parse doc
+      namespaces = xml.namespaces
+
+      cssPrefix = ''
+      namespaces.each do |namespace|
+        if namespace[1] == 'http://schemas.openxmlformats.org/spreadsheetml/2006/main' && namespace[0] != 'xmlns' then
+          cssPrefix = namespace[0].split(':')[1]+'|'
+        end
+      end
+
       rels_doc = @files.file.open "xl/_rels/workbook.xml.rels"
       rels = Nokogiri::XML::Document.parse(rels_doc).css("Relationship")
-      @sheets = xml.css('sheet').map do |sheet|
+      @sheets = xml.css(cssPrefix+'sheet').map do |sheet|
         sheetfile = rels.find { |el| sheet.attr("r:id") == el.attr("Id") }.attr("Target")
         Sheet.new(self, sheet.attr("name"), sheet.attr("sheetid"),  sheet.attr("state"), sheet.attr("visible"), sheet.attr("r:id"), sheetfile)
       end

lib/creek/sheet.rb

@@ -66,7 +66,7 @@ def rows_with_meta_data
     # Returns a hash per row that includes the cell ids and values.
     # Empty cells will be also included in the hash with a nil value.
     def rows_generator include_meta_data=false
-      path = "xl/#{@sheetfile}"
+      path = if @sheetfile.start_with? "/xl/" or @sheetfile.start_with? "xl/" then @sheetfile else "xl/#{@sheetfile}" end
       if @book.files.file.exist?(path)
         # SAX parsing, Each element in the stream comes through as two events:
         # one to open the element and one to close it.
@@ -103,6 +103,10 @@ def rows_generator include_meta_data=false
                 unless cell.nil?
                   cells[cell] = convert(node.inner_xml, cell_type, cell_style_idx)
                 end
+              elsif (node.name.eql? 't') and (node.node_type.eql? opener)
+                unless cell.nil?
+                  cells[cell] = convert(node.inner_xml, cell_type, cell_style_idx)
+                end
               end
             end
           end

spec/shared_string_spec.rb

@@ -7,12 +7,12 @@
     doc = Nokogiri::XML(shared_strings_xml_file)
     dictionary = Creek::SharedStrings.parse_shared_string_from_document(doc)
 
-    dictionary.keys.size.should == 5
-    dictionary[0].should == 'Cell A1'
-    dictionary[1].should == 'Cell B1'
-    dictionary[2].should == 'My Cell'
-    dictionary[3].should == 'Cell A2'
-    dictionary[4].should == 'Cell B2'
+    expect(dictionary.keys.size).to eq(5)
+    expect(dictionary[0]).to eq('Cell A1')
+    expect(dictionary[1]).to eq('Cell B1')
+    expect(dictionary[2]).to eq('My Cell')
+    expect(dictionary[3]).to eq('Cell A2')
+    expect(dictionary[4]).to eq('Cell B2')
   end
 
 end

spec/styles/converter_spec.rb

@@ -9,7 +9,7 @@ def convert(value, type, style)
 
     describe :date_time do
       it "works" do
-        convert('41275', 'n', :date_time).should == Date.new(2013,01,01)
+        expect(convert('41275', 'n', :date_time)).to eq(Date.new(2013,01,01))
       end
     end
   end

spec/styles/style_types_spec.rb

@@ -7,9 +7,9 @@
       xml_file = File.open('spec/fixtures/styles/first.xml')
       doc      = Nokogiri::XML(xml_file)
       res      = Creek::Styles::StyleTypes.new(doc).call
-      res.size.should == 8
-      res[3].should == :date_time
-      res.should == [:unsupported, :unsupported, :unsupported, :date_time, :unsupported, :unsupported, :unsupported, :unsupported]
+      expect(res.size).to eq(8)
+      expect(res[3]).to eq(:date_time)
+      expect(res).to eq([:unsupported, :unsupported, :unsupported, :date_time, :unsupported, :unsupported, :unsupported, :unsupported])
     end
   end
 end

spec/test_spec.rb

@@ -2,33 +2,38 @@
 
 describe 'Creek trying to parsing an invalid file.' do
   it 'Fail to open a legacy xls file.' do
-    lambda { Creek::Book.new 'spec/fixtures/invalid.xls' }.should raise_error 'Not a valid file format.'
+    expect { Creek::Book.new 'spec/fixtures/invalid.xls' }
+      .to raise_error 'Not a valid file format.'
   end
 
   it 'Ignore file extensions on request.' do
     path = 'spec/fixtures/sample-as-zip.zip'
-    lambda { Creek::Book.new path, :check_file_extension => false }.should_not raise_error
+    expect { Creek::Book.new path, check_file_extension: false }
+      .not_to raise_error
   end
 
   it 'Check file extension when requested.' do
-    open_book = lambda { Creek::Book.new 'spec/fixtures/invalid.xls', :check_file_extension => true }
-    open_book.should raise_error 'Not a valid file format.'
+    expect { Creek::Book.new 'spec/fixtures/invalid.xls', check_file_extension: true }
+      .to raise_error 'Not a valid file format.'
   end
 
   it 'Fail to open remote file' do
-    lambda { Creek::Book.new 'http://dev-builds.libreoffice.org/tmp/test.xlsx' }.should raise_error
+    expect { Creek::Book.new 'http://dev-builds.libreoffice.org/tmp/test.xlsx' }
+      .to raise_error(Zip::Error, /not found/)
   end
 
   it 'Opens remote file with remote flag' do
-    lambda { Creek::Book.new 'http://dev-builds.libreoffice.org/tmp/test.xlsx', remote: true }.should_not raise_error
+    expect { Creek::Book.new 'http://dev-builds.libreoffice.org/tmp/test.xlsx', remote: true }
+      .not_to raise_error
   end
 
   it 'Check file extension of original_filename if passed.' do
     path = 'spec/fixtures/temp_string_io_file_path_with_no_extension'
-    lambda { Creek::Book.new path, :original_filename => 'invalid.xls' }.should raise_error 'Not a valid file format.'
-    lambda { Creek::Book.new path, :original_filename => 'valid.xlsx' }.should_not raise_error
+    expect { Creek::Book.new path, :original_filename => 'invalid.xls' }
+      .to raise_error 'Not a valid file format.'
+    expect { Creek::Book.new path, :original_filename => 'valid.xlsx' }
+      .not_to raise_error
   end
-
 end
 
 describe 'Creek parsing a sample XLSX file' do
@@ -49,15 +54,15 @@
   end
 
   it 'open an XLSX file successfully.' do
-    @creek.should_not be_nil
+    expect(@creek).not_to be_nil
   end
 
   it 'find sheets successfully.' do
-    @creek.sheets.count.should == 1
+    expect(@creek.sheets.count).to eq(1)
     sheet = @creek.sheets.first
-    sheet.state.should eql nil
-    sheet.name.should eql 'Sheet1'
-    sheet.rid.should eql 'rId1'
+    expect(sheet.state).to eql nil
+    expect(sheet.name).to eql 'Sheet1'
+    expect(sheet.rid).to eql 'rId1'
   end
 
   it 'Parse rows with empty cells successfully.' do
@@ -68,16 +73,16 @@
       row_count += 1
     end
 
-    rows[0].should == @expected_rows[0]
-    rows[1].should == @expected_rows[1]
-    rows[2].should == @expected_rows[2]
-    rows[3].should == @expected_rows[3]
-    rows[4].should == @expected_rows[4]
-    rows[5].should == @expected_rows[5]
-    rows[6].should == @expected_rows[6]
-    rows[7].should == @expected_rows[7]
-    rows[8].should == @expected_rows[8]
-    row_count.should == 9
+    expect(rows[0]).to eq(@expected_rows[0])
+    expect(rows[1]).to eq(@expected_rows[1])
+    expect(rows[2]).to eq(@expected_rows[2])
+    expect(rows[3]).to eq(@expected_rows[3])
+    expect(rows[4]).to eq(@expected_rows[4])
+    expect(rows[5]).to eq(@expected_rows[5])
+    expect(rows[6]).to eq(@expected_rows[6])
+    expect(rows[7]).to eq(@expected_rows[7])
+    expect(rows[8]).to eq(@expected_rows[8])
+    expect(row_count).to eq(9)
   end
 
   it 'Parse rows with empty cells and meta data successfully.' do
@@ -87,6 +92,6 @@
       rows << row
       row_count += 1
     end
-    rows.map{|r| r['cells']}.should == @expected_rows
+    expect(rows.map{|r| r['cells']}).to eq(@expected_rows)
   end
 end