file_bsearch 1.0.2 → 1.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 47cd4336ef5b84ecf2abcc8ec596ad43245ae679
-  data.tar.gz: 0fab662f8d7ebda542d9d3e7eca1b53b9f7a69d6
+  metadata.gz: 789ee1ff24c686a8a3a55d6217bd5b20de1dec49
+  data.tar.gz: 003864105b6c297bb57688e64346cc5d8ffc85cb
 SHA512:
-  metadata.gz: e567fbb1af6fd494739b30d4e101c901ef16c53ba08a2b6bd035c2028b14e1bb593ea36460283fcb8bb7012fa5099b5994d219cf221022694b9ce37f841afc7a
-  data.tar.gz: 880234a6190ee6448db8d9fed3d372bd4e7d1812063d8ed11733062d2bbd0aa0712c03bef56d720facab666740c5d0a9e3f83c402399ecd9035a8d0075c3dfce
+  metadata.gz: 1a506a023c0eff005ecba65ff699775bf14a35b4474077f1becb9f5d6f8ccf11b755ce553097db3abd48886c848118f10460892b09ba669fe11f8e57549e8bcf
+  data.tar.gz: cad31db3261cc1e89f2846665627a7c0f9efacab7b2a72aed34b66c81ec826876bdeaa6208b33eef1267fd418ad9c8861fe82fa69d41b9881b5c357df3df80ca

data/README.ja.md

@@ -13,44 +13,66 @@ gem install file_bsearch
 
 ## Usage
 
+### 引数
+
+Usage のドキュメント内で使用する共通の変数について、以下に解説を記します。
+
+#### encoding
+
+文字コードの指定。
+
+utf8, eucjp, jis, sjis か、それぞれの接頭辞を渡してください。省略するとアスキー文字で検索をおこないます。
+
+#### path
+
+ソート済みのテキストファイルのパス。
+
+#### string
+
+検索に用いる文字列。
+
+#### prefix
+
+検索に用いる、接頭辞を表す文字列。
+
 ### 指定文字列の行が、ファイル内に存在するか
 
 ```ruby
-FileBsearch.exist?(path, string)
+FileBsearch.exist?(path, string, encoding)
 
-File.bsearch?(path, string)
+File.bsearch?(path, string, encoding)
 
 file = open(path)
-file.bsearch?(string)
+file.bsearch?(string, encoding)
 ```
 
-存在する場合は true 、存在しない場合は false が返ります。
+存在する場合は true 、存在しない場合は false を返します。
 
 ### 指定文字列の行が、ファイル内のどの位置にあるか
 
 ```ruby
-FileBsearch.index(path, string)
+FileBsearch.index(path, string, encoding)
 
-File.bsearch(path, string)
+File.bsearch(path, string, encoding)
 
 file = open(path)
-file.bsearch(string)
+file.bsearch(string, encoding)
 ```
 
-存在する場合はその行の開始位置、存在しない場合は false が返ります。
+存在する場合はその行の先頭のファイルポインタ位置の数値、存在しない場合は false を返します。
 
 ### 指定文字列から始まる行の取得
 
 ```ruby
-FileBsearch.get_lines(path, prefix)
+FileBsearch.get_lines(path, prefix, encoding)
 
-File.bsearch_lines(path, prefix)
+File.bsearch_lines(path, prefix, encoding)
 
 file = open(path)
-file.bsearch_lines(prefix)
+file.bsearch_lines(prefix, encoding)
 ```
 
-存在する場合はそれらの行の文字列を含んだ配列、存在しない場合は空の配列が返ります。
+存在する場合はそれらの行の文字列を含んだ配列、存在しない場合は空の配列を返します。
 
 ## Contributing
 

data/README.md

@@ -13,15 +13,35 @@ gem install file_bsearch
 
 ## Usage
 
+### common arguments
+
+common arguments in usage document. mean of each arguments are following.
+
+#### encoding
+
+for character encodings. you can pass String object that is 'utf8', 'eucjp', 'jis', 'sjis' or each prefix. if you pass nil or does not pass, search for ascii character.
+
+#### path
+
+path for sorted text file.
+
+#### string
+
+string for search.
+
+#### prefix
+
+string for prefix search.
+
 ### check to exist line that identical with passed string
 
 ```ruby
-FileBsearch.exist?(path, string)
+FileBsearch.exist?(path, string, encoding)
 
-File.bsearch?(path, string)
+File.bsearch?(path, string, encoding)
 
 file = open(path)
-file.bsearch?(string)
+file.bsearch?(string, encoding)
 ```
 
 if exists, return true.
@@ -30,29 +50,29 @@ if not exists, return false.
 ### search position in file for the line that is identical with passed string
 
 ```ruby
-FileBsearch.index(path, string)
+FileBsearch.index(path, string, encoding)
 
-File.bsearch(path, string)
+File.bsearch(path, string, encoding)
 
 file = open(path)
-file.bsearch(string)
+file.bsearch(string, encoding)
 ```
 
-if exists, return Integer object that is position in file for matched line.
+if exists, return Integer object that has the position of file pointer for head of matched line.
 if not exists, return false.
 
 ### get lines that matched string with passed prefix
 
 ```ruby
-FileBsearch.get_lines(path, prefix)
+FileBsearch.get_lines(path, prefix, encoding)
 
-File.bsearch_lines(path, prefix)
+File.bsearch_lines(path, prefix, encoding)
 
 file = open(path)
-file.bsearch_lines(prefix)
+file.bsearch_lines(prefix, encoding)
 ```
 
-if exists, return Array object that include matched lines.
+if exists, return Array object that has the matched lines.
 if not exists, return empty Array object.
 
 ## Contributing
@@ -65,4 +85,4 @@ if not exists, return empty Array object.
 
 ## Author
 
-[indeep-xyz](http://indeep.xyz/)
+[indeep-xyz](http://indeep.xyz/) (Japanese language)

data/file_bsearch.gemspec

@@ -9,7 +9,9 @@ Gem::Specification.new do |spec|
   spec.authors       = ["indeep-xyz"]
   spec.email         = ["indeep.xyz@gmail.com"]
   spec.summary       = %q{binary search for sorted text file.}
-  spec.description   = %q{binary search for sorted text file. it is effective when file size is bigger.}
+  spec.description   = <<EOS
+binary search for sorted text file. it is effective when file size is bigger.
+EOS
   spec.homepage      = "https://github.com/indeep-xyz/ruby-file-bsearch/"
   spec.license       = "MIT"
 
@@ -18,6 +20,8 @@ Gem::Specification.new do |spec|
   spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
   spec.require_paths = ["lib"]
 
+  spec.add_runtime_dependency "file_char_licker", "~> 0.5"
+
   spec.add_development_dependency "bundler", "~> 1.7"
   spec.add_development_dependency "rake", "~> 10.0"
   spec.add_development_dependency "rspec"

data/lib/file_bsearch/common.rb

@@ -1,87 +1,9 @@
-module FileBsearch
-  class << self
-
-    private
-
-    # get backward lines
-    #
-    # args
-    #   file ... File object
-    #            #pos value should be at SOL (Start Of Line)
-    #   size ... indication of reading size
-    #
-    # returner
-    #   String object as lines
-    def backward_lines(file, size = 512)
-
-      result = ""
-
-      while file.pos > 0
-
-        file.seek(-1, IO::SEEK_CUR)
-        char = file.getc
-
-        if char.match(/[\r\n]/) && result.size > size
-          break
-        else
-          result.insert(0, char)
-          file.seek(-1, IO::SEEK_CUR)
-        end
-      end
-
-      result
-    end
-
-    # get a line string at current position
-    #
-    # args
-    #     file ... File object
-    def current_line(file)
-
-      if file.pos > 0
-
-        # move pointer to before character
-        file.seek(-1, IO::SEEK_CUR)
+# coding: utf-8
 
-        # loop
-        # - move pointer until reach to EOL of before line.
-        until file.getc.match(/[\n\r]/)
+require "file_char_licker"
 
-          # move pointer to before character
-          if file.pos > 1
-            file.seek(-2, IO::SEEK_CUR)
-          else
-
-            # if EOS, break
-            file.rewind
-            break
-          end
-        end
-      end
-
-      file.gets
-    end
-
-    # get forward lines
-    #
-    # args
-    #   file ... File object
-    #            #pos value should be at SOL (Start Of Line)
-    #   size ... indication of reading size
-    #
-    # returner
-    #   String object as lines
-    def forward_lines(file, size = 512)
-
-      result = ""
-
-      while result.size < size && !file.eof?
-
-        result += file.gets
-      end
-
-      result
-    end
+module FileBsearch
+  class << self
 
     # args
     #     file ... File object
@@ -107,7 +29,7 @@ module FileBsearch
       # - if EOF, not found
       # - if pointer did not moved, not found
       while max > file.pos \
-          && (line = current_line(file)) \
+          && (line = file.current_line) \
           && file.pos != old_pos
 
         old_pos = file.pos
@@ -115,8 +37,8 @@ module FileBsearch
 
         # for debug
         # p "-- #{needle}, " + {
-        #   code: code,
         #   text: line.chomp,
+        #   code: code,
         #    min: min,
         #    max: max,
         #    pos: file.pos
@@ -128,8 +50,8 @@ module FileBsearch
         # - if match, return
         case code
         when -1 then min = file.pos - 1
-        when  1 then max = file.pos - line.length
-        else return file.pos - line.length
+        when  1 then max = file.pos - line.bytesize
+        else return file.pos - line.bytesize
         end
 
         # move to mid point
@@ -140,16 +62,18 @@ module FileBsearch
     end
 
     # args
-    #   target ... File object || path for String object
+    #   file ... File object || path for String object
     #
     # returner
     #   File object
-    def to_file(target)
+    def init_file(file, encoding)
 
-      # check the target argument
+      # check the file argument
       # - if not File object, open it as String for file path
-      return target if target.is_a?(File)
-      File.open(target.to_s)
+      file = File.open(file.to_s) unless file.is_a?(File)
+      FileCharLicker.attach(file, encoding)
+
+      file
     end
   end
 end

data/lib/file_bsearch/get_lines.rb

@@ -1,22 +1,23 @@
+# coding: utf-8
+
 module FileBsearch
   class << self
 
     # returner
     #   Array object
-    def get_lines(target, prefix, *args)
+    def get_lines(file, prefix, encoding = nil, *args)
 
-      prefix = prefix.to_s
-      file   = to_file(target)
+      file      = init_file(file, encoding)
+      prefix    = prefix.to_s
 
       # pass to scan method
-      pos = scan(file, nil, *args) do |line|
-        line[0, prefix.size] <=> prefix
-      end
+      pos = scan_around_lines_seed(file, prefix, *args)
 
       if pos.is_a?(Integer)
 
-        needle = Regexp.new("^#{Regexp.escape(prefix)}")
-        result = around_lines(file, pos, needle)
+        file.seek(pos)
+        lines  = file.around_lines(/^#{Regexp.escape(prefix)}/)
+        result = lines.chomp.split(/[\r\n]+/) if lines.bytesize > 0
       end
 
       result || []
@@ -24,124 +25,10 @@ module FileBsearch
 
     private
 
-    # get lines around for passed file#pos
-    #
-    # args
-    #   file   ... File object
-    #   pos    ... starting point for file#pos
-    #              require to be within contiguous range
-    #   needle ... RegExp object for contiguous check
-    # 
-    # returner
-    #   Array object
-    def around_lines(file, pos, needle)
-
-      # scan
-      min = scan_contiguous_min(file, pos, needle) || pos
-      max = scan_contiguous_max(file, pos, needle) || pos
-
-      # read
-      file.seek(min)
-      lines = file.read(max - min).chomp
-
-      # return
-      return lines.split(/[\r\n]+/) if lines.size > 0
-      []
-    end
-
-    # scan min file#pos of contiguous range.
-    #
-    # args
-    #   file   ... File object
-    #   pos    ... starting point for file#pos
-    #              require to be within contiguous range
-    #   needle ... RegExp object for contiguous check
-    #   step   ... buffer size for check processing
-    #
-    # returner
-    #   Integer object for file#pos
-    #   EOS of matched line
-    def scan_contiguous_min(file, pos, needle, step = 512)
-
-      file.pos = pos
-      min      = nil
-
-      loop do
-
-        lines     = backward_lines(file, step)
-        lines_pos = lines.index(needle)
-        file_pos  = file.pos
-
-        # for debug
-        # p [
-        #       lines: lines,
-        #   lines_pos: lines_pos,
-        #    file_pos: file_pos
-        #   ].to_s
-        # sleep 0.05
-
-        if lines_pos.nil?
-          break
-        else
-          min = file_pos + lines_pos
-          break if lines_pos > 0 || file_pos < 1
-        end
-      end
-
-      min
-    end
-
-    # scan max file#pos of contiguous range.
-    #
-    # args
-    #   file   ... File object
-    #   pos    ... starting point for file#pos
-    #              require to be within contiguous range
-    #   needle ... RegExp object for contiguous check
-    #   step   ... buffer size for check processing
-    #
-    # returner
-    #   Integer object for file#pos
-    #   EOL of matched line
-    def scan_contiguous_max(file, pos, needle, step = 512)
-
-      file.pos = pos
-      max      = nil
-
-      loop do
-
-        # file#pos before #forward_lines
-        pos_old   = file.pos
-
-        lines     = forward_lines(file, step)
-        lines_pos = lines.rindex(needle)
-
-        # for debug
-        # p [
-        #   lines: lines,
-        #   lines_pos: lines_pos,
-        #   file_pos: file.pos
-        #   ].to_s
-        # sleep 0.05
-
-        # if did not match needle
-        # - returner is last set value to 'max'
-        break if lines_pos.nil?
-
-        lines_end_pos = lines.index(/([\r\n]+?)/, lines_pos)
-
-        if file.eof?
-          max = (lines_end_pos.nil?) ? file.size : pos_old + lines_end_pos
-          break
-        else
-          max = pos_old + lines_end_pos
-
-          break if lines_end_pos < lines.size - 1
-        end
-
+    def scan_around_lines_seed(file, prefix, *args)
+      scan(file, nil, *args) do |line|
+        line[0, prefix.size] <=> prefix
       end
-
-      max
     end
   end
 end

data/lib/file_bsearch/main.rb

@@ -1,3 +1,5 @@
+# coding: utf-8
+
 module FileBsearch
   class << self
 
@@ -14,16 +16,16 @@ module FileBsearch
     end
 
     # args
-    #   target ... File object || path for String object
+    #   file ... File object || path for String object
     #
     # returner
     #   position in target file
-    def index(target, *args, &block)
+    def index(file, needle, encoding = nil, *args, &block)
 
-      file = to_file(target)
+      file = init_file(file, encoding)
 
       # pass to scan method
-      scan(file, *args, &block)
+      scan(file, needle, *args, &block)
     end
   end
 end

data/lib/file_bsearch/version.rb

@@ -1,3 +1,3 @@
 module FileBsearch
-  VERSION = "1.0.2"
+  VERSION = "1.1.0"
 end

data/spec/file_bsearch_spec.rb

@@ -1,16 +1,19 @@
+# coding: utf-8
+
 require 'spec_helper'
 
 describe FileBsearch do
 
-  let(:path)      { '/tmp/file_bsearch_sample.csv' }
-  let(:correct)   { '9' }
-  let(:incorrect) { '!!!!!' }
-
-  let!(:helper) { MySpecHelper.new(path) }
-
   it 'has a version number' do
     expect(FileBsearch::VERSION).not_to be nil
   end
+end
+
+describe '<ASCII FILE>' do
+
+  let!(:helper)   { MySpecHelper.new }
+  let(:path)      { MySpecHelper::ASCII_SAMPLE_PATH }
+  let(:incorrect) { MySpecHelper::INCORRECT }
 
   describe 'module' do
 
@@ -18,7 +21,8 @@ describe FileBsearch do
 
       it 'when a record is exists' do
 
-        result = FileBsearch.exist?(path, correct)
+        arg    = MySpecHelper::ASCII_ARG
+        result = FileBsearch.exist?(path, arg)
 
         expect(result).to eq(true)
       end
@@ -26,7 +30,6 @@ describe FileBsearch do
       it 'when a record is not exist' do
 
         result = FileBsearch.exist?(path, incorrect)
-
         expect(result).to eq(false)
       end
 
@@ -34,8 +37,8 @@ describe FileBsearch do
 
         File.foreach(path) do |line|
 
-          correct  = line.chomp
-          result = FileBsearch.exist?(path, correct)
+          arg    = line.chomp
+          result = FileBsearch.exist?(path, arg)
 
           expect(result).to eq(true)
         end
@@ -46,10 +49,11 @@ describe FileBsearch do
 
       it 'when returner is number as position in the file' do
 
-        b_result = FileBsearch.index(path, correct)
-        h_result = helper.correct?(correct, b_result)
+        arg    = MySpecHelper::ASCII_ARG
+        pos    = FileBsearch.index(path, arg)
 
-        expect(h_result).to eq(true)
+        result = helper.ascii_correct?(pos, arg)
+        expect(result).to eq(true)
       end
 
       it 'when returner is false' do
@@ -62,15 +66,16 @@ describe FileBsearch do
     describe '#get_lines' do
       it 'when lines is exist that with the prefix' do
 
-        result = FileBsearch.get_lines(path, '1000')
+        arg     = MySpecHelper::ASCII_LIST_ARG
+        correct = MySpecHelper::ASCII_LIST_CORRECT
 
-        expect(result).to eq(%w{1000 10000})
+        result  = FileBsearch.get_lines(path, arg)
+        expect(result).to eq(correct)
       end
 
       it 'when lines is not exist that with the prefix' do
 
-        result = FileBsearch.get_lines(path, '!!!!!')
-
+        result = FileBsearch.get_lines(path, incorrect)
         expect(result).to eq([])
       end
     end
@@ -78,23 +83,25 @@ describe FileBsearch do
 
   describe 'instance method' do
 
-    let(:file) { open(path) }
+    let(:file)  { open(path) }
 
     describe '#bsearch' do
 
       it 'when returner is number as position in the file' do
 
-        b_result = file.bsearch(correct)
-        h_result = helper.correct?(correct, b_result)
+        arg    = MySpecHelper::ASCII_LIST_ARG
+        pos    = file.bsearch(arg)
+        result = helper.ascii_correct?(pos, arg)
 
-        expect(h_result).to eq(true)
+        expect(result).to eq(true)
       end
     end
 
     describe '#bsearch?' do
       it 'when a record is exists' do
 
-        result = file.bsearch?(correct)
+        arg    = MySpecHelper::ASCII_LIST_ARG
+        result = file.bsearch?(arg)
 
         expect(result).to eq(true)
       end
@@ -103,9 +110,11 @@ describe FileBsearch do
     describe '#bsearch_lines' do
       it 'when lines is exist that with the prefix' do
 
-        result = file.bsearch_lines('1000')
+        arg     = MySpecHelper::ASCII_LIST_ARG
+        correct = MySpecHelper::ASCII_LIST_CORRECT
 
-        expect(result).to eq(%w{1000 10000})
+        result = file.bsearch_lines(arg)
+        expect(result).to eq(correct)
       end
     end
   end
@@ -116,17 +125,19 @@ describe FileBsearch do
 
       it 'when returner is number as position in the file' do
 
-        b_result = File.bsearch(path, correct)
-        h_result = helper.correct?(correct, b_result)
+        arg    = MySpecHelper::ASCII_LIST_ARG
+        pos    = File.bsearch(path, arg)
+        result = helper.ascii_correct?(pos, arg)
 
-        expect(h_result).to eq(true)
+        expect(result).to eq(true)
       end
     end
 
     describe '#bsearch?' do
       it 'when a record is exists' do
 
-        result = File.bsearch?(path, correct)
+        arg    = MySpecHelper::ASCII_LIST_ARG
+        result = File.bsearch?(path, arg)
 
         expect(result).to eq(true)
       end
@@ -135,9 +146,80 @@ describe FileBsearch do
     describe '#bsearch_lines' do
       it 'when lines is exist that with the prefix' do
 
-        result = File.bsearch_lines(path, '1000')
+        arg     = MySpecHelper::ASCII_LIST_ARG
+        correct = MySpecHelper::ASCII_LIST_CORRECT
+
+        result = File.bsearch_lines(path, arg)
+        expect(result).to eq(correct)
+      end
+    end
+  end
+end
+
+describe '<UTF-8 FILE>' do
+
+  let!(:helper)   { MySpecHelper.new }
+  let(:path)      { MySpecHelper::MULTIBYTE_SAMPLE_PATH }
+  let(:incorrect) { MySpecHelper::INCORRECT }
+  let(:arg)       { MySpecHelper::MULTIBYTE_ARG }
+  let(:list_arg)  { MySpecHelper::MULTIBYTE_LIST_ARG }
+
+  describe 'module' do
+
+    describe '#exist?' do
+
+      it 'when a record is exists' do
+
+        result = FileBsearch.exist?(path, arg, 'utf-8')
+        expect(result).to eq(true)
+      end
+
+      it 'when a record is not exist' do
+
+        result = FileBsearch.exist?(path, incorrect, 'utf-8')
+        expect(result).to eq(false)
+      end
 
-        expect(result).to eq(%w{1000 10000})
+      it 'when all records are exist' do
+
+        File.foreach(path) do |line|
+
+          arg    = line.chomp
+          result = FileBsearch.exist?(path, arg, 'utf-8')
+
+          expect(result).to eq(true)
+        end
+      end
+    end
+
+    describe '#index' do
+
+      it 'when returner is number as position in the file' do
+
+        pos    = FileBsearch.index(path, arg, 'utf-8')
+        result = helper.multibyte_correct?(pos, arg)
+        expect(result).to eq(true)
+      end
+
+      it 'when returner is false' do
+
+        result = FileBsearch.index(path, incorrect, 'utf-8')
+        expect(result).to eq(false)
+      end
+    end
+
+    describe '#get_lines' do
+      it 'when lines is exist that with the prefix' do
+
+        correct = MySpecHelper::MULTIBYTE_LIST_CORRECT
+        result  = FileBsearch.get_lines(path, list_arg, 'utf-8')
+        expect(result).to eq(correct)
+      end
+
+      it 'when lines is not exist that with the prefix' do
+
+        result = FileBsearch.get_lines(path, incorrect, 'utf-8')
+        expect(result).to eq([])
       end
     end
   end

data/spec/lib/my_spec_helper.rb

@@ -1,32 +1,80 @@
+# coding: utf-8
+
 class MySpecHelper
 
   attr_reader :path
 
-  def initialize(path)
+  ASCII_SAMPLE_PATH     = '/tmp/file_bsearch_sample_ascii.csv'
+  MULTIBYTE_SAMPLE_PATH = '/tmp/file_bsearch_sample_multibyte.csv'
+
+  ASCII_ARG          = '9'
+  ASCII_LIST_ARG     = '1000'
+  ASCII_LIST_CORRECT = %w{1000 10000}
+  INCORRECT          = '!!!!!'
+
+  MULTIBYTE_ARG          = 'ああ'
+  MULTIBYTE_LIST_ARG     = 'あ'
+  MULTIBYTE_LIST_CORRECT = ["あぁ", "ああ", "あぃ", "あい", "あぅ", "あう", "あぇ", "あえ", "あぉ", "あお", "あか", "あが", "あき", "あぎ", "あく", "あぐ", "あけ", "あげ", "あこ", "あご", "あさ", "あざ", "あし", "あじ", "あす", "あず", "あせ", "あぜ", "あそ", "あぞ", "あた", "あだ", "あち", "あぢ", "あっ", "あつ", "あづ", "あて", "あで", "あと", "あど", "あな", "あに", "あぬ", "あね", "あの", "あは", "あば", "あぱ", "あひ", "あび", "あぴ", "あふ", "あぶ", "あぷ", "あへ", "あべ", "あぺ", "あほ", "あぼ", "あぽ", "あま", "あみ", "あむ", "あめ", "あも", "あゃ", "あや", "あゅ", "あゆ", "あょ", "あよ", "あら", "あり", "ある", "あれ", "あろ", "あゎ", "あわ", "あゐ", "あゑ", "あを", "あん", "あゔ", "あゕ", "あゖ"]
 
-    @path = path
+  def initialize
 
-    init_sample_file
+    init_ascii_sample
+    init_multibyte_sample
   end
 
-  def init_sample_file
+  def init_ascii_sample
+
+    path = ASCII_SAMPLE_PATH
 
     # guard
     # - if file is already exists, do not
-    return nil if File.exist?(@path)
+    return nil if File.exist?(path)
 
     # create sorted string
-    str = (0..10_000).to_a.sort { |a, b| a.to_s <=> b.to_s }.join("\n")
+    str = create_lines(0..10_000)
 
-    IO.write(@path, str)
+    IO.write(path, str)
   end
 
-  # check returner
+  def init_multibyte_sample
+
+    path = MULTIBYTE_SAMPLE_PATH
+
+    # guard
+    # - if file is already exists, do not
+    return nil if File.exist?(path)
+
+    # create sorted string
+    str = create_lines('ぁぁ'..'んん')
+
+    IO.write(path, str)
+  end
+
+  # check correct for ascii file
+  #
+  # args
+  #   pos ... Integer object for position in file
+  #           (returner of FileBsearch#index)
+  #   str ... String object for comparison
+  def ascii_correct?(pos, str)
+    str == IO.read(ASCII_SAMPLE_PATH, str.bytesize, pos)
+  end
+
+  # check correct for ascii file
   #
   # args
-  #   str ... string for correct
-  #   pos ... position in the file (FileBsearch#returner)
-  def correct?(str, pos)
-    str == IO.read(@path, str.size, pos)
+  #   pos ... Integer object for position in file
+  #           (returner of FileBsearch#index)
+  #   str ... String object for comparison
+  def multibyte_correct?(pos, str)
+
+    read_str = IO.read(MULTIBYTE_SAMPLE_PATH, str.bytesize, pos)
+    str == read_str.toutf8
+  end
+
+  private
+
+  def create_lines(range)
+    range.to_a.sort { |a, b| a.to_s <=> b.to_s }.join("\n")
   end
 end

metadata

@@ -1,15 +1,29 @@
 --- !ruby/object:Gem::Specification
 name: file_bsearch
 version: !ruby/object:Gem::Version
-  version: 1.0.2
+  version: 1.1.0
 platform: ruby
 authors:
 - indeep-xyz
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-10-02 00:00:00.000000000 Z
+date: 2014-10-09 00:00:00.000000000 Z
 dependencies:
+- !ruby/object:Gem::Dependency
+  name: file_char_licker
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.5'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '0.5'
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
@@ -66,8 +80,8 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-description: binary search for sorted text file. it is effective when file size is
-  bigger.
+description: |
+  binary search for sorted text file. it is effective when file size is bigger.
 email:
 - indeep.xyz@gmail.com
 executables: []