saxlsx 0.3.0 → 1.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 5b0cbd91196c90e22a5dc4ede2bf372792cc5e3f
-  data.tar.gz: 96bec50a0662ab5741c54e329c80c073c9c202e2
+  metadata.gz: 48288886c8097438a2f3527b42c31f0db2a63c34
+  data.tar.gz: 3386ac7f891cc701d09697e5620d4d5aafee216b
 SHA512:
-  metadata.gz: e55094b6131ae28e50afb8bed91d91f15b200596f94cb6215f9bea6c5e9303d158f281ec6495d8cbb481528f3893ea614748d3fd3c9130b06555df6d98930e10
-  data.tar.gz: fbd741be1e3b91784bc11502351a4195d31d0459de3765bd7de6acc64cebf4b5580a9eb0fc19bab0ad844f27ab316c84085ff9b4155bf76d52e5cef2f2d05738
+  metadata.gz: e8ca7c015a5b34310b2008f70ea9eec36abdcae99dd4fd29ba70711de286f8dfb37a1cbf4ff7e295b98bd1a2d2cdb700f9b83f9cd790df16d9818a943792c005
+  data.tar.gz: 7dea47237742b543733e9aab3ffdd2c8d11cb9a8a7826e63b244dd20b782dc5661c16bccbef23485cec81fb9855dd1069799d351f67555f35dcb0509a8f30763

data/.travis.yml

@@ -1,5 +1,5 @@
 rvm:
-  - 2.0.0
-  - 2.1.0
-  - rbx
+  - 2.0
+  - 2.1
+  - 2.2
   - ruby-head

data/Gemfile

@@ -2,3 +2,8 @@ source 'https://rubygems.org'
 
 # Specify your gem's dependencies in saxlsx.gemspec
 gemspec
+
+gem 'axlsx', '~> 2.1.0.pre'
+gem 'rubyXL'
+gem 'simple_xlsx_reader'
+gem 'creek'

data/README.md

@@ -2,7 +2,12 @@
 
 [![Build Status](https://travis-ci.org/mak-it/saxlsx.png?branch=master)](https://travis-ci.org/mak-it/saxlsx)
 
-Fast XLSX reader on top of Ox SAX parser.
+**Fast** and memory efficient XLSX reader on top of Ox SAX parser.
+
+It reads row by row and doesn't store the whole sheet in memory, so this
+approach is more suitable when parsing big files. This also means that functions
+and references will not work, as this style of parsing doesn't know
+anything about other rows.
 
 ## Installation
 
@@ -36,6 +41,19 @@ Saxlsx::Workbook.open filename do |w|
 end
 ```
 
+## How fast is it?
+
+```bash
+$ rake bench
+```
+
+```
+creek                  2.610000   0.060000   2.670000 (  2.704594)
+rubyXL                 3.830000   0.130000   3.960000 (  3.985651)
+saxlsx                 0.750000   0.010000   0.760000 (  0.785445)
+simple_xlsx_reader     1.870000   0.040000   1.910000 (  1.940999)
+```
+
 ## Contributing
 
 1. Fork it

data/Rakefile

@@ -4,3 +4,71 @@ require "rspec/core/rake_task"
 Bundler::GemHelper.install_tasks
 RSpec::Core::RakeTask.new(:spec)
 task :default => :spec
+
+
+task :bench do
+  require 'benchmark'
+  require 'axlsx'
+  require 'saxlsx'
+  require 'rubyXL'
+  require 'simple_xlsx_reader'
+  require 'creek'
+
+  path = "tmp/bench.xlsx"
+  unless File.exists?(path)
+    puts "* Generating #{path}"
+    FileUtils.mkdir_p File.dirname(path)
+    Axlsx::Package.new do |p|
+      money_style = p.workbook.styles.add_style(
+        num_fmt: 5, format_code: "€0.000"
+      )
+      p.workbook.add_worksheet(:name => "Pie Chart") do |sheet|
+        10000.times do
+          sheet.add_row(
+            [Date.today, Time.now, 1000, 3.14, "Long" * 100],
+            types: [:date, :time, :integer, :float, :string],
+            style: [nil, nil, nil, money_style, nil]
+          )
+        end
+      end
+      p.use_shared_strings = true
+      p.serialize(path)
+    end
+  end
+
+  Benchmark.benchmark('', 20) do |x|
+    x.report "creek" do
+      w = Creek::Book.new path
+      w.sheets.each do |s|
+        s.rows.each do |r|
+          r.values.inspect
+        end
+      end
+    end
+    x.report "rubyXL" do
+      w = RubyXL::Parser.parse path
+      w.worksheets.each do |s|
+        s.each do |r|
+          r.cells.map(&:value).inspect
+        end
+      end
+    end
+    x.report "saxlsx" do
+      Saxlsx::Workbook.open path do |w|
+        w.sheets.each do |s|
+          s.rows.each do |r|
+            r.to_a.inspect
+          end
+        end
+      end
+    end
+    x.report "simple_xlsx_reader" do
+      w = SimpleXlsxReader.open path
+      w.sheets.each do |s|
+        s.rows.each do |r|
+          r.to_a.inspect
+        end
+      end
+    end
+  end
+end

data/lib/saxlsx/file_system.rb

@@ -19,11 +19,12 @@ module Saxlsx
     end
 
     def workbook
-      @zip.read('xl/workbook.xml').match(/<sheets>.*<\/sheets>/).to_s
+      @zip.read('xl/workbook.xml')
     end
 
     def shared_strings
-      @zip.read('xl/sharedStrings.xml')
+      file = @zip.glob('xl/shared[Ss]trings.xml').first
+      @zip.read(file) if file
     end
 
     def styles

data/lib/saxlsx/rows_collection.rb

@@ -3,15 +3,15 @@ module Saxlsx
 
     include Enumerable
 
-    def initialize(index, file_system, shared_strings)
+    def initialize(index, file_system, workbook)
       @index = index
       @file_system = file_system
-      @shared_strings = shared_strings
+      @workbook = workbook
       @sheet = file_system.sheet(index)
     end
 
     def each(&block)
-      RowsCollectionParser.parse @index, @sheet, @shared_strings, &block
+      RowsCollectionParser.parse @index, @sheet, @workbook, &block
     end
 
     def count

data/lib/saxlsx/rows_collection_parser.rb

@@ -35,13 +35,12 @@ module Saxlsx
       49 => :unsupported     # @
     }
 
-    DATE_SYSTEM_1900 = DateTime.new(1899, 12, 30)
-
     def self.parse(index, data, workbook, &block)
       SaxParser.parse self.new(workbook, &block), data
     end
 
     def initialize(workbook, &block)
+      @base_date      = workbook.base_date
       @shared_strings = workbook.shared_strings
       @number_formats = workbook.number_formats
       @block = block
@@ -97,16 +96,16 @@ module Saxlsx
     def value_of(text)
       case @current_type
       when 's'
-        @shared_strings[text.to_i]
+        @shared_strings[text.to_i] || text
       when 'b'
         BooleanParser.parse text
       else
         case @current_number_format
         when :date
-          DATE_SYSTEM_1900 + text.to_i
+          @base_date + text.to_i
         when :date_time
           # Round time to seconds
-          date = DATE_SYSTEM_1900 + (text.to_f * 86400).round.fdiv(86400)
+          date = @base_date + (text.to_f * 86400).round.fdiv(86400)
           DateTime.new(date.year, date.month, date.day, date.hour, date.minute, date.second)
         when :fixnum
           text.to_i

data/lib/saxlsx/shared_string_collection_parser.rb

@@ -2,7 +2,12 @@ module Saxlsx
   class SharedStringCollectionParser < Ox::Sax
 
     def self.parse(file_system, &block)
-      SaxParser.parse self.new(&block), file_system.shared_strings
+      shared_strings = file_system.shared_strings
+      if shared_strings
+        SaxParser.parse self.new(&block), shared_strings
+      else
+        []
+      end
     end
 
     def initialize(&block)

data/lib/saxlsx/sheet.rb

@@ -3,15 +3,15 @@ module Saxlsx
 
     attr_reader :name
 
-    def initialize(name, index, file_system, shared_strings)
+    def initialize(name, index, file_system, workbook)
       @name = name
       @index = index
       @file_system = file_system
-      @shared_strings = shared_strings
+      @workbook = workbook
     end
 
     def rows
-      @rows ||= RowsCollection.new(@index, @file_system, @shared_strings)
+      @rows ||= RowsCollection.new(@index, @file_system, @workbook)
     end
 
     def to_csv(path)

data/lib/saxlsx/sheet_collection.rb

@@ -3,13 +3,13 @@ module Saxlsx
 
     include Enumerable
 
-    def initialize(file_system, shared_strings)
+    def initialize(file_system, workbook)
       @file_system = file_system
-      @shared_strings = shared_strings
+      @workbook = workbook
     end
 
     def each(&block)
-      SheetCollectionParser.parse @file_system, @shared_strings, &block
+      SheetCollectionParser.parse @file_system, @workbook, &block
     end
 
   end

data/lib/saxlsx/sheet_collection_parser.rb

@@ -3,30 +3,55 @@ module Saxlsx
 
     CurrentSheet = Struct.new :index, :name
 
-    def self.parse(file_system, shared_strings, &block)
-      SaxParser.parse self.new(file_system, shared_strings, &block), file_system.workbook
+    def self.parse(file_system, workbook, &block)
+      SaxParser.parse(
+        self.new(file_system, workbook, &block),
+        file_system.workbook
+      )
     end
 
-    def initialize(file_system, shared_strings, &block)
+    def initialize(file_system, workbook, &block)
       @file_system = file_system
-      @shared_strings = shared_strings
+      @workbook = workbook
       @block = block
       @index = -1
+      @workbook_pr = false
     end
 
     def start_element(name)
-      @current_sheet = CurrentSheet.new(@index += 1) if name == :sheet
+      case name
+      when :sheet
+        @current_sheet = CurrentSheet.new(@index += 1)
+      when :workbookPr
+        @workbook_pr = true
+      end
     end
 
     def end_element(name)
-      if name == :sheet
-        @block.call Sheet.new(@current_sheet.name, @current_sheet.index, @file_system, @shared_strings)
+      case name
+      when :sheet
+        @block.call Sheet.new(
+          @current_sheet.name,
+          @current_sheet.index,
+          @file_system,
+          @workbook
+        )
         @current_sheet = nil
+      when :workbookPr
+        @workbook_pr = false
       end
     end
 
     def attr(name, value)
-      @current_sheet.name = value if @current_sheet && name == :name
+      if @current_sheet
+        if name == :name
+          @current_sheet.name = value
+        end
+      elsif @workbook_pr
+        if name == :date1904 && value =~ /true|1/i
+          @workbook.date1904 = true
+        end
+      end
     end
 
   end

data/lib/saxlsx/version.rb

@@ -1,3 +1,3 @@
 module Saxlsx
-  VERSION = '0.3.0'
+  VERSION = '1.0.0'
 end

data/lib/saxlsx/workbook.rb

@@ -1,5 +1,9 @@
 module Saxlsx
   class Workbook
+    DATE_SYSTEM_1900 = DateTime.new(1899, 12, 30)
+    DATE_SYSTEM_1904 = DateTime.new(1904, 1, 1)
+
+    attr_accessor :date1904
 
     def self.open(filename)
       begin
@@ -35,6 +39,10 @@ module Saxlsx
       @number_formats ||= StyleCollection.new(@file_system).to_a
     end
 
+    def base_date
+      @base_date ||= date1904 ? DATE_SYSTEM_1904 : DATE_SYSTEM_1900
+    end
+
     def to_csv(path)
       sheets.each { |s| s.to_csv path }
     end

data/spec/sheet_spec.rb

@@ -90,4 +90,18 @@ describe Sheet do
     end
   end
 
+  context 'with 1904 date system' do
+    let(:filename) { "#{File.dirname(__FILE__)}/data/Spec1904.xlsx" }
+
+    it 'should use 1904 date system when converting dates' do
+      Workbook.open filename do |w|
+        w.sheets[0].tap do |s|
+          s.rows[0].should eq [
+            DateTime.new(1970, 1, 1, 1, 0, 0),
+            DateTime.new(1970, 1, 1)
+          ]
+        end
+      end
+    end
+  end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: saxlsx
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 1.0.0
 platform: ruby
 authors:
 - Edgars Beigarts
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-03-25 00:00:00.000000000 Z
+date: 2015-03-26 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rubyzip
@@ -126,6 +126,7 @@ files:
 - saxlsx.gemspec
 - spec/column_name_generator_spec.rb
 - spec/data/Spec.xlsx
+- spec/data/Spec1904.xlsx
 - spec/sheet_spec.rb
 - spec/spec_helper.rb
 - spec/workbook_spec.rb
@@ -156,6 +157,7 @@ summary: Fast xlsx reader on top of Ox SAX parser
 test_files:
 - spec/column_name_generator_spec.rb
 - spec/data/Spec.xlsx
+- spec/data/Spec1904.xlsx
 - spec/sheet_spec.rb
 - spec/spec_helper.rb
 - spec/workbook_spec.rb