Merge branch 'master' into issue_389

.travis.yml

@@ -1,20 +1,21 @@
 language: ruby
 rvm:
-  - 2.2
-  - 2.3
-  - 2.4
   - 2.5
+  - 2.6
+  - 2.7
   - ruby-head
   - jruby-9.1.6.0
 env:
   - LONG_RUN=true
 matrix:
   include:
-    - rvm: 2.0
-      gemfile: Gemfile_ruby2
-    - rvm: 2.1
-      gemfile: Gemfile_ruby2
+    - rvm: 2.6
+      env: RUBYOPT=--jit LONG_RUN=true
+    - rvm: ruby-head
+      env: RUBYOPT=--jit LONG_RUN=true
   allow_failures:
     - rvm: ruby-head
+    - rvm: ruby-head
+      env: RUBYOPT=--jit LONG_RUN=true
     - rvm: jruby-9.1.6.0
 bundler_args: --without local_development

CHANGELOG.md

@@ -1,8 +1,51 @@
-## Unreleased
+##  Unreleased
+
+##  [2.8.3] 2020-02-03
+### Changed/Added
+- Updated rubyzip version. Now minimal version is 1.3.0 [515](https://github.com/roo-rb/roo/pull/515) - [CVE-2019-16892](https://github.com/rubyzip/rubyzip/pull/403)
+
+##  [2.8.2] 2019-02-01
+### Changed/Added
+- Support range cell for Excelx's links [490](https://github.com/roo-rb/roo/pull/490)
+- Skip `extract_hyperlinks` if not required [488](https://github.com/roo-rb/roo/pull/488)
+
+### Fixed
+- Fixed error for invalid link [492](https://github.com/roo-rb/roo/pull/492)
+
+##  [2.8.1] 2019-01-21
+### Fixed
+- Fixed error if excelx's cell have empty children [487](https://github.com/roo-rb/roo/pull/487)
+
+##  [2.8.0] 2019-01-18
+### Fixed
+- Fixed inconsistent column length for CSV [375](https://github.com/roo-rb/roo/pull/375)
+- Fixed formatted_value with `%` for Excelx [416](https://github.com/roo-rb/roo/pull/416)
+- Improved Memory consumption and performance [434](https://github.com/roo-rb/roo/pull/434) [449](https://github.com/roo-rb/roo/pull/449) [454](https://github.com/roo-rb/roo/pull/454) [456](https://github.com/roo-rb/roo/pull/456) [458](https://github.com/roo-rb/roo/pull/458) [462](https://github.com/roo-rb/roo/pull/462) [466](https://github.com/roo-rb/roo/pull/466)
+- Accept both Transitional and Strict Type for Excelx's worksheets [441](https://github.com/roo-rb/roo/pull/441)
+- Fixed ruby warnings [442](https://github.com/roo-rb/roo/pull/442) [476](https://github.com/roo-rb/roo/pull/476)
+- Restore support for URL as file identifier for CSV [462](https://github.com/roo-rb/roo/pull/462)
+- Fixed missing location for Excelx's links [482](https://github.com/roo-rb/roo/pull/482)
+
+### Changed / Added
+- Drop support for ruby 2.2.x and lower
 - Updated rubyzip version for fixing security issue. Now minimal version is 1.2.1
+- Roo::Excelx::Coordinate now inherits Array [458](https://github.com/roo-rb/roo/pull/458)
+- Improved Roo::HeaderRowNotFoundError exception's message [461](https://github.com/roo-rb/roo/pull/461)
+- Added `empty_cell` option which by default disable allocation for Roo::Excelx::Cell::Empty [464](https://github.com/roo-rb/roo/pull/464)
+- Added support for variable number of decimals for Excelx's formatted_value [387](https://github.com/roo-rb/roo/pull/387)
+- Added `disable_html_injection` option to disable html injection for shared string in `Roo::Excelx` [392](https://github.com/roo-rb/roo/pull/392)
+- Added image extraction for Excelx [414](https://github.com/roo-rb/roo/pull/414) [397](https://github.com/roo-rb/roo/pull/397)
+- Added support for `1e6` as scientific notation for Excelx [433](https://github.com/roo-rb/roo/pull/433)
+- Added support for Integer as 0 based index for Excelx's `sheet_for` [455](https://github.com/roo-rb/roo/pull/455)
+- Extended `no_hyperlinks` option for non streaming Excelx methods [459](https://github.com/roo-rb/roo/pull/459)
+- Added `empty_cell` option to disable Roo::Excelx::Cell::Empty allocation for Excelx [464](https://github.com/roo-rb/roo/pull/464)
+- Added support for Integer with leading zero for Roo:Excelx [479](https://github.com/roo-rb/roo/pull/479)
+- Refactored Excelx code [453](https://github.com/roo-rb/roo/pull/453) [477](https://github.com/roo-rb/roo/pull/477) [483](https://github.com/roo-rb/roo/pull/483) [484](https://github.com/roo-rb/roo/pull/484)
 
 ### Deprecations
 - Roo::Excelx::Sheet#present_cells is deprecated [454](https://github.com/roo-rb/roo/pull/454)
+- Roo::Utils.split_coordinate is deprecated [458](https://github.com/roo-rb/roo/pull/458)
+- Roo::Excelx::Cell::Base#link is deprecated [457](https://github.com/roo-rb/roo/pull/457)
 
 ## [2.7.1] 2017-01-03
 ### Fixed

README.md

@@ -18,7 +18,7 @@ Install as a gem
 Or add it to your Gemfile
 
 ```ruby
-gem "roo", "~> 2.7.0"
+gem "roo", "~> 2.8.0"
 ```
 ## Usage
 
@@ -261,7 +261,7 @@ ods.formula('A', 2)
 csv = Roo::CSV.new("mycsv.csv")
 ```
 
-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.
+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.
 
@@ -274,6 +274,18 @@ csv = Roo::CSV.new("mytsv.tsv", csv_options: {col_sep: "\t"})
 csv = Roo::CSV.new("mycsv.csv", csv_options: {encoding: Encoding::ISO_8859_1})
 ```
 
+You can also open csv files through the Roo::Spreadsheet class (useful if you accept both CSV and Excel types from a user file upload, for example).
+
+```ruby
+# Load a spreadsheet from a file path
+# Roo figures out the right parser based on file extension
+spreadsheet = Roo::Spreadsheet.open(csv_or_xlsx_file)
+
+# Load a csv and auto-strip the BOM (byte order mark)
+# csv files saved from MS Excel typically have the BOM marker at the beginning of the file
+spreadsheet = Roo::Spreadsheet.open("mycsv.csv", { csv_options: { encoding: 'bom|utf-8' } })
+```
+
 ## Upgrading from Roo 1.13.x
 If you use ``.xls`` or Google spreadsheets, you will need to install ``roo-xls`` or ``roo-google`` to continue using that functionality.
 
@@ -283,7 +295,7 @@ Roo's public methods have stayed relatively consistent between 1.13.x and 2.0.0,
 
 ## Contributing
 ### Features
-1. Fork it ( https://github.com/[my-github-username]/roo/fork )
+1. Fork it ( https://github.com/roo-rb/roo/fork )
 2. Install it (`bundle install --with local_development`)
 3. Create your feature branch (`git checkout -b my-new-feature`)
 4. Commit your changes (`git commit -am 'My new feature'`)
@@ -300,9 +312,6 @@ You can run the tests/examples with Rspec like reporters by running
 Roo also has a few tests that take a long time (5+ seconds). To run these, use
 `LONG_RUN=true bundle exec rake`
 
-When testing using Ruby 2.0 or 2.1, use this command:
-`BUNDLE_GEMFILE=Gemfile_ruby2 bundle exec rake`
-
 ### Issues
 
 If you find an issue, please create a gist and refer to it in an issue ([sample gist](https://gist.github.com/stevendaniels/98a05849036e99bb8b3c)). Here are some instructions for creating such a gist.

lib/roo/base.rb

@@ -288,12 +288,12 @@ def 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]
+                (first_column..last_column).each_with_object({}) do |col, hash|
+                  hash[cell(@header_line, col)] = col
+                end
 
       @header_line.upto(last_row) do |line|
-        yield(Hash[headers.map { |k, v| [k, cell(line, v)] }])
+        yield(headers.each_with_object({}) { |(k, v), hash| hash[k] = cell(line, v) })
       end
     end
   end
@@ -424,9 +424,9 @@ def find_by_row(row_index)
 
   def find_by_conditions(options)
     rows = first_row.upto(last_row)
-    header_for = Hash[1.upto(last_column).map do |col|
-      [col, cell(@header_line, col)]
-    end]
+    header_for = 1.upto(last_column).each_with_object({}) do |col, hash|
+      hash[col] = cell(@header_line, col)
+    end
 
     # are all conditions met?
     conditions = options[:conditions]
@@ -441,9 +441,9 @@ def find_by_conditions(options)
       rows.map { |i| row(i) }
     else
       rows.map do |i|
-        Hash[1.upto(row(i).size).map do |j|
-          [header_for.fetch(j), cell(i, j)]
-        end]
+        1.upto(row(i).size).each_with_object({}) do |j, hash|
+          hash[header_for.fetch(j)] = cell(i, j)
+        end
       end
     end
   end
@@ -497,8 +497,11 @@ def sanitize_value(v)
   def set_headers(hash = {})
     # try to find header row with all values or give an error
     # then create new hash by indexing strings and keeping integers for header array
-    @headers = row_with(hash.values, true)
-    @headers = Hash[hash.keys.zip(@headers.map { |x| header_index(x) })]
+    header_row = row_with(hash.values, true)
+    @headers = {}
+    hash.each_with_index do |(key, _), index|
+      @headers[key] = header_index(header_row[index])
+    end
   end
 
   def header_index(query)
@@ -541,7 +544,7 @@ def download_uri(uri, tmpdir)
     tempfilename = File.join(tmpdir, find_basename(uri))
     begin
       File.open(tempfilename, "wb") do |file|
-        open(uri, "User-Agent" => "Ruby/#{RUBY_VERSION}") do |net|
+        URI.open(uri, "User-Agent" => "Ruby/#{RUBY_VERSION}") do |net|
           file.write(net.read)
         end
       end

lib/roo/csv.rb

@@ -90,17 +90,23 @@ def read_cells(sheet = default_sheet)
     def each_row(options, &block)
       if uri?(filename)
         each_row_using_tempdir(options, &block)
-      elsif is_stream?(filename_or_stream)
-        ::CSV.new(filename_or_stream, options).each(&block)
       else
-        ::CSV.foreach(filename, options, &block)
+        csv_foreach(filename_or_stream, options, &block)
       end
     end
 
     def each_row_using_tempdir(options, &block)
       ::Dir.mktmpdir(Roo::TEMP_PREFIX, ENV["ROO_TMP"]) do |tmpdir|
         tmp_filename = download_uri(filename, tmpdir)
-        ::CSV.foreach(tmp_filename, options, &block)
+        csv_foreach(tmp_filename, options, &block)
+      end
+    end
+
+    def csv_foreach(path_or_io, options, &block)
+      if is_stream?(path_or_io)
+        ::CSV.new(path_or_io, **options).each(&block)
+      else
+        ::CSV.foreach(path_or_io, **options, &block)
       end
     end
 

lib/roo/excelx.rb

@@ -40,8 +40,9 @@ def initialize(filename_or_stream, options = {})
       sheet_options = {}
       sheet_options[:expand_merged_ranges] = (options[:expand_merged_ranges] || false)
       sheet_options[:no_hyperlinks] = (options[:no_hyperlinks] || false)
+      sheet_options[:empty_cell] = (options[:empty_cell] || false)
       shared_options = {}
-        
+
       shared_options[:disable_html_wrapper] = (options[:disable_html_wrapper] || false)
       unless is_stream?(filename_or_stream)
         file_type_check(filename_or_stream, %w[.xlsx .xlsm], 'an Excel 2007', file_warning, packed)
@@ -59,16 +60,17 @@ def initialize(filename_or_stream, options = {})
       @filename = local_filename(filename_or_stream, @tmpdir, packed)
       process_zipfile(@filename || filename_or_stream)
 
-      @sheet_names = workbook.sheets.map do |sheet|
-        unless options[:only_visible_sheets] && sheet['state'] == 'hidden'
-          sheet['name']
-        end
-      end.compact
+      @sheet_names = []
       @sheets = []
-      @sheets_by_name = Hash[@sheet_names.map.with_index do |sheet_name, n|
-        @sheets[n] = Sheet.new(sheet_name, @shared, n, sheet_options)
-        [sheet_name, @sheets[n]]
-      end]
+      @sheets_by_name = {}
+
+      workbook.sheets.each_with_index do |sheet, index|
+        next if options[:only_visible_sheets] && sheet['state'] == 'hidden'
+
+        sheet_name = sheet['name']
+        @sheet_names << sheet_name
+        @sheets_by_name[sheet_name] = @sheets[index] = Sheet.new(sheet_name, @shared, index, sheet_options)
+      end
 
       if cell_max
         cell_count = ::Roo::Utils.num_cells_in_range(sheet_for(options.delete(:sheet)).dimensions)
@@ -333,7 +335,7 @@ def extract_worksheet_ids(entries, path)
 
       wb.extract(path)
       workbook_doc = Roo::Utils.load_xml(path).remove_namespaces!
-      workbook_doc.xpath('//sheet').map { |s| s.attributes['id'].value }
+      workbook_doc.xpath('//sheet').map { |s| s['id'] }
     end
 
     # Internal
@@ -359,14 +361,11 @@ def extract_worksheet_rels(entries, path)
       rels_doc = Roo::Utils.load_xml(path).remove_namespaces!
 
       relationships = rels_doc.xpath('//Relationship').select do |relationship|
-        worksheet_types.include? relationship.attributes['Type'].value
+        worksheet_types.include? relationship['Type']
       end
 
-      relationships.inject({}) do |hash, relationship|
-        attributes = relationship.attributes
-        id = attributes['Id']
-        hash[id.value] = attributes['Target'].value
-        hash
+      relationships.each_with_object({}) do |relationship, hash|
+        hash[relationship['Id']] = relationship['Target']
       end
     end
 
@@ -463,7 +462,7 @@ def process_zipfile_entries(entries)
     end
 
     def safe_send(object, method, *args)
-      object.send(method, *args) if object && object.respond_to?(method)
+      object.send(method, *args) if object&.respond_to?(method)
     end
 
     def worksheet_types

lib/roo/excelx/cell.rb

@@ -40,8 +40,7 @@ def type
       end
 
       def self.create_cell(type, *values)
-        type_class = cell_class(type)
-        type_class && type_class.new(*values)
+        cell_class(type)&.new(*values)
       end
 
       def self.cell_class(type)

lib/roo/excelx/cell/base.rb

@@ -58,7 +58,7 @@ def type
         end
 
         def formula?
-          !!@formula
+          !!(defined?(@formula) && @formula)
         end
 
         def link?

lib/roo/excelx/cell/number.rb

@@ -24,7 +24,7 @@ def create_numeric(number)
           when /\.0/
             Float(number)
           else
-            (number.include?('.') || (/\A[-+]?\d+E[-+]?\d+\z/i =~ number)) ? Float(number) : Integer(number)
+            (number.include?('.') || (/\A[-+]?\d+E[-+]?\d+\z/i =~ number)) ? Float(number) : Integer(number, 10)
           end
         end
 
@@ -48,7 +48,7 @@ def generate_formatter(format)
           when /^(0+)$/ then "%0#{$1.size}d"
           when /^0\.(0+)$/ then "%.#{$1.size}f"
           when '#,##0' then number_format('%.0f')
-          when '#,##0.00' then number_format('%.2f')
+          when /^#,##0.(0+)$/ then number_format("%.#{$1.size}f")
           when '0%'
             proc do |number|
               Kernel.format('%d%%', number.to_f * 100)
@@ -64,6 +64,7 @@ def generate_formatter(format)
           when '#,##0.00;[Red](#,##0.00)' then number_format('%.2f', '[Red](%.2f)')
             # FIXME: not quite sure what the format should look like in this case.
           when '##0.0E+0' then '%.1E'
+          when "_-* #,##0.00\\ _€_-;\\-* #,##0.00\\ _€_-;_-* \"-\"??\\ _€_-;_-@_-" then number_format('%.2f', '-%.2f')
           when '@' then proc { |number| number }
           else
             raise "Unknown format: #{format.inspect}"

lib/roo/excelx/cell/time.rb

@@ -13,7 +13,7 @@ def initialize(value, formula, excelx_type, style, link, base_date, coordinate)
           super
           @format = excelx_type.last
           @datetime = create_datetime(base_date, value)
-          @value = link ? Roo::Link.new(link, value) : (value.to_f * 86_400).to_i
+          @value = link ? Roo::Link.new(link, value) : (value.to_f * 86_400).round.to_i
         end
 
         def formatted_value