rbcsv 0.1.8 → 0.2.0

data/examples/benchmarks/benchmark.rb

@@ -0,0 +1,372 @@
+#!/usr/bin/env ruby
+# frozen_string_literal: true
+
+require 'csv'
+require 'benchmark'
+require 'fileutils'
+require 'time'
+require_relative '../../lib/rbcsv'
+
+# ベンチマーク設定
+ITERATIONS = 1000
+LARGE_ITERATIONS = 10
+CSV_FILE = 'sample.csv'
+LARGE_CSV_FILE = 'large_sample.csv'
+
+puts "=" * 60
+puts "RbCsv vs Ruby CSV ベンチマーク比較"
+puts "=" * 60
+puts "Ruby version: #{RUBY_VERSION}"
+puts "RbCsv version: #{RbCsv::VERSION}"
+puts "Date: #{Time.now.strftime('%Y-%m-%d %H:%M:%S')}"
+puts
+
+class BenchmarkRunner
+  def initialize
+    @csv_content = nil
+    @large_csv_content = nil
+  end
+
+  def setup_sample_data
+    # サンプルデータが存在しない場合は作成
+    unless File.exist?(CSV_FILE)
+      create_sample_csv_file
+    end
+
+    @csv_content = File.read(CSV_FILE)
+    puts "使用ファイル: #{CSV_FILE}"
+    puts "ファイルサイズ: #{@csv_content.bytesize} bytes"
+
+    # レコード数を確認
+    records_count = CSV.read(CSV_FILE).length - 1  # ヘッダーを除く
+    puts "レコード数: #{records_count}"
+    puts
+  end
+
+  def create_sample_csv_file
+    puts "サンプルCSVファイルを作成中..."
+    CSV.open(CSV_FILE, "w") do |csv|
+      # ヘッダー
+      csv << %w[id name age score department salary active_date]
+
+      # データ行を生成（1000レコード）
+      1000.times do |i|
+        csv << [
+          i + 1,
+          "User#{i + 1}",
+          rand(20..65),
+          rand(60.0..100.0).round(2),
+          %w[Engineering Sales Marketing HR][rand(4)],
+          rand(40000..120000),
+          (Date.today - rand(365)).to_s
+        ]
+      end
+    end
+    puts "サンプルファイル作成完了: #{CSV_FILE}"
+  end
+
+  def create_large_sample_data
+    if File.exist?(LARGE_CSV_FILE)
+      puts "既存のファイルを使用: #{LARGE_CSV_FILE}"
+    else
+      puts "テストデータを作成中..."
+      original_data = CSV.read(CSV_FILE)
+
+      CSV.open(LARGE_CSV_FILE, "w") do |csv|
+        csv << original_data.first
+
+        50.times do |batch|
+          original_data[1..-1].each_with_index do |row, index|
+            new_row = row.dup
+            new_row[0] = (batch * 1000 + index + 1).to_s  # IDを更新
+            csv << new_row
+          end
+        end
+      end
+
+      puts "ファイル作成完了: #{LARGE_CSV_FILE} (#{File.size(LARGE_CSV_FILE)} bytes)"
+    end
+
+    # 既存または新規に関わらず、コンテンツを読み込み
+    @large_csv_content = File.read(LARGE_CSV_FILE)
+    puts "データサイズ: #{@large_csv_content.bytesize} bytes"
+    puts
+  end
+
+  def run_basic_parsing_benchmark
+    puts "🚀 基本パース性能比較 (#{ITERATIONS}回実行)"
+    puts "-" * 50
+
+    Benchmark.bm(40) do |x|
+      # Ruby標準CSV
+      x.report("Ruby CSV.parse") do
+        ITERATIONS.times do
+          CSV.parse(@csv_content)
+        end
+      end
+
+      x.report("Ruby CSV.parse (headers: true)") do
+        ITERATIONS.times do
+          CSV.parse(@csv_content, headers: true)
+        end
+      end
+
+      # RbCsv 基本機能
+      x.report("RbCsv.parse") do
+        ITERATIONS.times do
+          RbCsv.parse(@csv_content)
+        end
+      end
+
+      x.report("RbCsv.parse! (with trim)") do
+        ITERATIONS.times do
+          RbCsv.parse!(@csv_content)
+        end
+      end
+
+      # RbCsv 型認識機能
+      x.report("RbCsv.parse_typed") do
+        ITERATIONS.times do
+          RbCsv.parse_typed(@csv_content)
+        end
+      end
+
+      x.report("RbCsv.parse_typed! (typed + trim)") do
+        ITERATIONS.times do
+          RbCsv.parse_typed!(@csv_content)
+        end
+      end
+    end
+    puts
+  end
+
+  def run_file_reading_benchmark
+    puts "📁 ファイル読み込み性能比較 (#{ITERATIONS}回実行)"
+    puts "-" * 50
+
+    Benchmark.bm(40) do |x|
+      # Ruby標準CSV
+      x.report("Ruby CSV.read") do
+        ITERATIONS.times do
+          CSV.read(CSV_FILE)
+        end
+      end
+
+      x.report("Ruby CSV.read (headers: true)") do
+        ITERATIONS.times do
+          CSV.read(CSV_FILE, headers: true)
+        end
+      end
+
+      # RbCsv
+      x.report("RbCsv.read") do
+        ITERATIONS.times do
+          RbCsv.read(CSV_FILE)
+        end
+      end
+
+      x.report("RbCsv.read! (with trim)") do
+        ITERATIONS.times do
+          RbCsv.read!(CSV_FILE)
+        end
+      end
+
+      x.report("RbCsv.read_typed") do
+        ITERATIONS.times do
+          RbCsv.read_typed(CSV_FILE)
+        end
+      end
+
+      x.report("RbCsv.read_typed! (typed + trim)") do
+        ITERATIONS.times do
+          RbCsv.read_typed!(CSV_FILE)
+        end
+      end
+    end
+    puts
+  end
+
+  def run_large_data_benchmark
+    create_large_sample_data
+
+    puts "💪 データ性能比較 (#{LARGE_ITERATIONS}回実行)"
+    puts "-" * 50
+
+    # データが正しく読み込まれているかチェック
+    if @large_csv_content.nil? || @large_csv_content.empty?
+      puts "エラー: データが読み込まれていません"
+      return
+    end
+
+    Benchmark.bm(40) do |x|
+      # パース性能比較
+      x.report("Ruby CSV.parse (large)") do
+        LARGE_ITERATIONS.times do
+          CSV.parse(@large_csv_content)
+        end
+      end
+
+      x.report("RbCsv.parse (large)") do
+        LARGE_ITERATIONS.times do
+          RbCsv.parse(@large_csv_content)
+        end
+      end
+
+      x.report("RbCsv.parse_typed (large)") do
+        LARGE_ITERATIONS.times do
+          RbCsv.parse_typed(@large_csv_content)
+        end
+      end
+
+      # ファイル読み込み性能比較
+      x.report("Ruby CSV.read (large file)") do
+        LARGE_ITERATIONS.times do
+          CSV.read(LARGE_CSV_FILE)
+        end
+      end
+
+      x.report("RbCsv.read (large file)") do
+        LARGE_ITERATIONS.times do
+          RbCsv.read(LARGE_CSV_FILE)
+        end
+      end
+
+      x.report("RbCsv.read_typed (large file)") do
+        LARGE_ITERATIONS.times do
+          RbCsv.read_typed(LARGE_CSV_FILE)
+        end
+      end
+    end
+    puts
+  end
+
+  def run_writing_benchmark
+    puts "✏️ ファイル書き込み性能比較 (#{ITERATIONS}回実行)"
+    puts "-" * 50
+
+    # テスト用の出力ファイル名（絶対パスに修正）
+    csv_out = File.join(Dir.pwd, 'benchmark_csv_output.csv')
+    rbcsv_out = File.join(Dir.pwd, 'benchmark_rbcsv_output.csv')
+
+    # 書き込み用のテストデータを準備（文字列に変換）
+    test_data = []
+    100.times do |i|
+      test_data << [
+        (i + 1).to_s,
+        "TestUser#{i + 1}",
+        rand(20..65).to_s,
+        rand(60.0..100.0).round(2).to_s,
+        %w[Engineering Sales Marketing HR][rand(4)],
+        rand(40000..120000).to_s,
+        (Date.today - rand(365)).to_s
+      ]
+    end
+
+    Benchmark.bm(40) do |x|
+      # Ruby標準CSV書き込み
+      x.report("Ruby CSV.open (write)") do
+        ITERATIONS.times do
+          CSV.open(csv_out, "w") do |csv|
+            csv << %w[id name age score department salary active_date]
+            test_data.each { |row| csv << row }
+          end
+        end
+      end
+
+      # RbCsv書き込み
+      x.report("RbCsv.write") do
+        ITERATIONS.times do
+          write_data = [%w[id name age score department salary active_date]] + test_data
+          RbCsv.write(rbcsv_out, write_data)
+        end
+      end
+    end
+
+    # テストファイルをクリーンアップ
+    [csv_out, rbcsv_out].each do |file|
+      File.delete(file) if File.exist?(file)
+    end
+    puts
+  end
+
+
+  def run_type_conversion_comparison
+    puts "🔢 型変換処理の比較 (#{ITERATIONS}回実行)"
+    puts "-" * 50
+
+    csv_data = CSV.parse(@csv_content)
+    rbcsv_data = RbCsv.parse(@csv_content)
+    rbcsv_typed_data = RbCsv.parse_typed(@csv_content)
+
+    Benchmark.bm(40) do |x|
+      # 手動型変換 vs 自動型変換
+      x.report("Manual conversion (CSV)") do
+        ITERATIONS.times do
+          csv_data[1..-1].map do |row|
+            [
+              row[0].to_i,      # id to integer
+              row[1],           # title (keep as string)
+              row[2],           # description (keep as string)
+              row[3],           # category (keep as string)
+              row[4],           # status (keep as string)
+              row[5],           # location (keep as string)
+              Time.parse(row[6]), # start_date to time
+              Time.parse(row[7]), # end_date to time
+              row[8].to_i,      # max_participants to integer
+              Time.parse(row[9]), # created_at to time
+              Time.parse(row[10]) # updated_at to time
+            ]
+          end
+        end
+      end
+
+      x.report("Manual conversion (RbCsv)") do
+        ITERATIONS.times do
+          rbcsv_data[1..-1].map do |row|
+            [
+              row[0].to_i,      # id to integer
+              row[1],           # title (keep as string)
+              row[2],           # description (keep as string)
+              row[3],           # category (keep as string)
+              row[4],           # status (keep as string)
+              row[5],           # location (keep as string)
+              Time.parse(row[6]), # start_date to time
+              Time.parse(row[7]), # end_date to time
+              row[8].to_i,      # max_participants to integer
+              Time.parse(row[9]), # created_at to time
+              Time.parse(row[10]) # updated_at to time
+            ]
+          end
+        end
+      end
+
+      x.report("Automatic conversion (RbCsv typed)") do
+        ITERATIONS.times do
+          rbcsv_typed_data[1..-1]  # すでに型変換済み
+        end
+      end
+    end
+    puts
+  end
+
+  def cleanup
+    [LARGE_CSV_FILE].each do |file|
+      File.delete(file) if File.exist?(file)
+    end
+  end
+  def run_all_benchmarks
+    setup_sample_data
+    run_basic_parsing_benchmark
+    run_file_reading_benchmark
+    run_writing_benchmark
+    run_large_data_benchmark
+    run_type_conversion_comparison
+    cleanup
+  end
+end
+
+# ベンチマーク実行
+if __FILE__ == $0
+  runner = BenchmarkRunner.new
+  runner.run_all_benchmarks
+end

data/{output_comparison.rb → examples/benchmarks/output_comparison.rb}

@@ -2,7 +2,7 @@
 # encoding: utf-8
 
 require 'csv'
-require_relative 'lib/r_csv'
+require_relative '../../lib/rbcsv'
 
 CSV_FILE = 'sample.csv'
 csv_content = File.read(CSV_FILE)
@@ -48,8 +48,8 @@ puts "Data:"
 csv_table.each_with_index { |row, i| puts "  [#{i}] #{row.to_h}" }
 puts
 
-puts "=== 4. RCsv.parse (current - raw arrays) ==="
-rcv_parse = RCsv.parse(small_csv)
+puts "=== 4. RbCsv.parse (current - raw arrays) ==="
+rcv_parse = RbCsv.parse(small_csv)
 puts "Type: #{rcv_parse.class}"
 puts "Rows: #{rcv_parse.length}"
 puts "Data:"
@@ -73,11 +73,11 @@ puts "   - Excludes header from data rows"
 puts "   - Example: #<CSV::Row id:\"1\" name:\"Alice\" age:\"30\">"
 puts
 
-puts "3. RCsv.parse (current):"
+puts "3. RbCsv.parse (current):"
 puts "   - Returns: Array<Array<String>>"
-puts "   - Excludes header row (data only)"
+puts "   - Includes header row as first element"
 puts "   - Raw string arrays"
-puts "   - Example: [[\"1\", \"Alice\", \"30\"], [\"2\", \"Bob\", \"25\"]]"
+puts "   - Example: [[\"id\", \"name\", \"age\"], [\"1\", \"Alice\", \"30\"]]"
 puts
 
 puts "=== Key Differences ==="
@@ -86,54 +86,69 @@ puts
 puts "Header handling:"
 puts "  CSV.parse (default):     Includes header as row[0]"
 puts "  CSV.parse (headers=true): Excludes header, provides row['column'] access"
-puts "  RCsv.parse:              Excludes header, raw arrays only"
+puts "  RbCsv.parse:              Includes header as row[0] (same as CSV.parse default)"
 puts
 
 puts "Data structure:"
 puts "  CSV.parse:               Can return CSV::Row objects with named access"
-puts "  RCsv.parse:              Always returns raw Array<String>"
+puts "  RbCsv.parse:              Always returns raw Array<String>"
 puts
 
 puts "Row count difference:"
 csv_default = CSV.parse(small_csv)
 csv_headers = CSV.parse(small_csv, headers: true)
-rcv_data = RCsv.parse(small_csv)
+rcv_data = RbCsv.parse(small_csv)
 
 puts "  CSV.parse (default):     #{csv_default.length} rows (includes header)"
 puts "  CSV.parse (headers=true): #{csv_headers.length} rows (data only)"
-puts "  RCsv.parse:              #{rcv_data.length} rows (data only)"
+puts "  RbCsv.parse:              #{rcv_data.length} rows (includes header)"
 puts
 
 puts "=== Compatibility Recommendations ==="
 puts
 
-puts "To match CSV.parse (default behavior):"
-puts "  - RCsv should include header row as first element"
-puts "  - Return format: [[\"id\", \"name\"], [\"1\", \"Alice\"], [\"2\", \"Bob\"]]"
+puts "RbCsv already matches CSV.parse (default behavior):"
+puts "  - Both include header row as first element"
+puts "  - Both return format: [[\"id\", \"name\"], [\"1\", \"Alice\"], [\"2\", \"Bob\"]]"
 puts
 
-puts "To match CSV.parse (headers: true):"
-puts "  - More complex: need CSV::Row-like objects"
+puts "To add CSV.parse (headers: true) compatibility:"
+puts "  - Would need CSV::Row-like objects"
 puts "  - Alternative: return {headers: [...], data: [[...], [...]]} structure"
 puts
 
-puts "Current RCsv matches:"
-puts "  - CSV.parse(content, headers: true) # data rows only"
-puts "  - But returns raw arrays instead of CSV::Row objects"
+puts "Current RbCsv compatibility:"
+puts "  - Matches CSV.parse(content) exactly - both return Array<Array<String>> with headers"
+puts "  - Does not yet support CSV.parse(content, headers: true) which returns CSV::Table"
 puts
 
 puts "=== Specific Implementation Suggestions ==="
 puts
 
-puts "Option 1: Match CSV.parse default (simplest):"
-puts "  RCsv.parse(content) → includes header row"
+puts "Option 1: Keep current behavior (already compatible):"
+puts "  RbCsv.parse(content) → already includes header row ✅"
 puts
 
-puts "Option 2: Add options parameter:"
-puts "  RCsv.parse(content, headers: false) → includes header"
-puts "  RCsv.parse(content, headers: true) → excludes header"
+puts "Option 2: Add options parameter for structured data:"
+puts "  RbCsv.parse(content, headers: false) → current behavior (default)"
+puts "  RbCsv.parse(content, headers: true) → future: structured data like CSV::Table"
 puts
 
-puts "Option 3: Multiple methods:"
-puts "  RCsv.parse_raw(content) → raw arrays with header"
-puts "  RCsv.parse(content) → structured data"
+puts "Option 3: Add convenience methods:"
+puts "  RbCsv.table(content) → future: structured data with named access"
+puts "  RbCsv.foreach(file) → future: streaming with block"
+puts
+puts "=== Actual Compatibility Status ==="
+puts
+puts "✅ RbCsv.parse(content) == CSV.parse(content)"
+puts "   Both return: Array<Array<String>> with header row included"
+puts
+puts "❌ RbCsv does not yet support:"
+puts "   - CSV.parse(content, headers: true) → CSV::Table with CSV::Row objects"
+puts "   - Named column access like row['name']"
+puts "   - Header converters (e.g., :symbol)"
+puts
+puts "Future improvements could add:"
+puts "   - RbCsv.parse(content, headers: true) for structured data"
+puts "   - RbCsv::Row and RbCsv::Table classes for compatibility"
+puts "   - Integration with streaming API for large files"