saxlsx 1.1.0 → 1.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 4b88b50d9148929625a2acdf54bd7d8818f7b03c
-  data.tar.gz: ef9c524a126a202c9ca6db5b04e468c5d8054fbf
+  metadata.gz: 8846f61615fb9bef698366383838cca50957a242
+  data.tar.gz: 5fd15e93f3cac8c82eb292abe9202eeea69259d4
 SHA512:
-  metadata.gz: 80c590e046b649a9907fb2ec0501114fe5a2bc79407f2e3311e0794ecf290e50e43f85c911552513d526fc3133a8f1ae194d6034e53bf18dcaa9eaecb0fe71aa
-  data.tar.gz: cfaf63a640c451923a82a4522b0a15ea480bbfa29842a3260f2cf4b25bfd0a98b6e964252b3510b234e599fbc905c91e9d1a3d1991e77003ab7cb9d340b1cea8
+  metadata.gz: 2450c4efabb247287a16e9daf331f6be1c6ddc8c3f3abdd47b556c8e09e6234a86fb1bbad533fa0a7f18631cdd8dd7207931d78e127eb30045b5a2b95b37844c
+  data.tar.gz: d272886859b32e9e851a52163d8c86f4fcfa5b5f5395e42687eebaaa61525f1672357f009a15bca982c9b8d89df75028aa358a32d9691189db1c32e31f9c72c8

data/Gemfile

@@ -7,3 +7,6 @@ gem 'axlsx', '~> 2.1.0.pre'
 gem 'rubyXL'
 gem 'simple_xlsx_reader'
 gem 'creek'
+gem 'oxcelix'
+gem 'roo'
+gem 'dullard'

data/README.md

@@ -1,13 +1,13 @@
 # Saxlsx
 
-[![Build Status](https://travis-ci.org/mak-it/saxlsx.png?branch=master)](https://travis-ci.org/mak-it/saxlsx)
+[![Join the chat at https://gitter.im/mak-it/saxlsx](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/mak-it/saxlsx?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
+
+[![Build Status](https://travis-ci.org/mak-it/saxlsx.svg?branch=master)](https://travis-ci.org/mak-it/saxlsx)
 
 **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.
+approach is more suitable when parsing big files.
 
 ## Installation
 
@@ -34,6 +34,7 @@ $ gem install saxlsx
 ```ruby
 Saxlsx::Workbook.open filename do |w|
   w.sheets.each do |s|
+    puts s.rows.count
     s.rows.each do |r|
       puts r.inspect
     end
@@ -48,10 +49,25 @@ $ 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)
+Shared Strings
+
+creek                  3.200000   0.060000   3.260000 (  3.392446)
+dullard                3.020000   0.070000   3.090000 (  3.154582)
+oxcelix                1.720000   0.060000   1.780000 (  1.824003)
+roo                   14.580000   0.490000  15.070000 ( 15.311664)
+rubyXL                 5.180000   0.160000   5.340000 (  5.432973)
+saxlsx                 1.020000   0.010000   1.030000 (  1.044249)
+simple_xlsx_reader     2.580000   0.050000   2.630000 (  2.660065)
+
+Inline Strings
+
+creek                  3.970000   0.330000   4.300000 (  4.395139)
+dullard                3.700000   0.360000   4.060000 (  4.174130)
+oxcelix                ERROR
+roo                   14.700000   0.550000  15.250000 ( 15.644152)
+rubyXL                 5.830000   0.440000   6.270000 (  6.365362)
+saxlsx                 1.500000   0.040000   1.540000 (  1.679372)
+simple_xlsx_reader     2.790000   0.070000   2.860000 (  2.916020)
 ```
 
 ## Contributing

data/Rakefile

@@ -7,68 +7,6 @@ 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
+  require './spec/benchmarks.rb'
+  Saxlsx::Benchmarks.new.run
+end

data/lib/saxlsx/file_system.rb

@@ -32,8 +32,7 @@ module Saxlsx
     end
 
     def sheet(i)
-      f = @zip.glob('xl/worksheets/sheet*.xml').sort[i]
-      @zip.read(f)
+      @zip.read("xl/worksheets/sheet#{i+1}.xml")
     end
 
   end

data/lib/saxlsx/version.rb

@@ -1,3 +1,3 @@
 module Saxlsx
-  VERSION = '1.1.0'
+  VERSION = '1.2.0'
 end

data/spec/benchmarks.rb

@@ -0,0 +1,143 @@
+require 'benchmark'
+require 'axlsx'
+require 'saxlsx'
+require 'rubyXL'
+require 'simple_xlsx_reader'
+require 'creek'
+require 'oxcelix'
+require 'roo'
+require 'dullard'
+
+module Saxlsx
+  class Benchmarks
+    def run
+      path = "tmp/bench_shared_strings.xlsx"
+      generate path, true
+      benchmark "Shared Strings", path
+
+      path = "tmp/bench_inline_strings.xlsx"
+      generate path, false
+      benchmark "Inline Strings", path
+    end
+
+    private
+
+    def generate(path, shared_strings)
+      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 => "Sheet 1") 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 = shared_strings
+          p.serialize(path)
+        end
+      end
+    end
+
+    def benchmark(title, path)
+      puts
+      puts title
+      puts
+      Benchmark.benchmark('', 20) do |x|
+        x.report "creek" do
+          run_creek(path)
+        end
+        x.report "dullard" do
+          run_dullard(path)
+        end
+        x.report "oxcelix" do
+          run_oxcelix(path)
+        end
+        x.report "roo" do
+          run_roo(path)
+        end
+        x.report "rubyXL" do
+          run_rubyxl(path)
+        end
+        x.report "saxlsx" do
+          run_saxlsx(path)
+        end
+        x.report "simple_xlsx_reader" do
+          run_simple_xlsx_reader(path)
+        end
+      end
+    end
+
+    def run_creek(path)
+      w = Creek::Book.new path
+      w.sheets.each do |s|
+        s.rows.each do |r|
+          r.values.inspect
+        end
+      end
+    end
+
+    def run_oxcelix(path)
+      w = Oxcelix::Workbook.new(path)
+      w.sheets.each do |s|
+        s.to_ru.to_a.each do |r|
+          r.inspect
+        end
+      end
+    rescue
+      puts "ERROR"
+    end
+
+    def run_rubyxl(path)
+      w = RubyXL::Parser.parse path
+      w.worksheets.each do |s|
+        s.each do |r|
+          r.cells.map(&:value).inspect
+        end
+      end
+    end
+
+    def run_saxlsx(path)
+      Saxlsx::Workbook.open path do |w|
+        w.sheets.each do |s|
+          s.rows.each do |r|
+            r.to_a.inspect
+          end
+        end
+      end
+    end
+
+    def run_simple_xlsx_reader(path)
+      w = SimpleXlsxReader.open path
+      w.sheets.each do |s|
+        s.rows.each do |r|
+          r.to_a.inspect
+        end
+      end
+    end
+
+    def run_roo(path)
+      w = Roo::Excelx.new path
+      w.each_with_pagename do |_, s|
+        s.each do |r|
+          r.to_a.inspect
+        end
+      end
+    end
+
+    def run_dullard(path)
+      w = Dullard::Workbook.new path
+      w.sheets.each do |s|
+        s.rows.each do |r|
+          r.to_a.inspect
+        end
+      end
+    end
+  end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: saxlsx
 version: !ruby/object:Gem::Version
-  version: 1.1.0
+  version: 1.2.0
 platform: ruby
 authors:
 - Edgars Beigarts
@@ -124,6 +124,7 @@ files:
 - lib/saxlsx/version.rb
 - lib/saxlsx/workbook.rb
 - saxlsx.gemspec
+- spec/benchmarks.rb
 - spec/column_name_generator_spec.rb
 - spec/data/Spec.xlsx
 - spec/data/Spec1904.xlsx
@@ -156,6 +157,7 @@ signing_key:
 specification_version: 4
 summary: Fast xlsx reader on top of Ox SAX parser
 test_files:
+- spec/benchmarks.rb
 - spec/column_name_generator_spec.rb
 - spec/data/Spec.xlsx
 - spec/data/Spec1904.xlsx