マルチバイト文字列を2バイトとみなし、指定したバイト数でカット。中途半端は切る。

string_util.rb

class StringUtil
  def self.split_by_size(str, max_byte_size)
    length, bytes  = 0, 0
    str.each_char do |char|
      char_size = multi_bytes?(char)? 2 : 1
      break if (bytes + char_size) > max_byte_size
      bytes += char_size
      length += 1
    end
    str[0, length]
  end

  def self.multi_bytes?(char)
    char.bytesize > 1
  end
end

string_util_spec.rb

# -*- coding: UTF-8 -*-
require File.expand_path(File.dirname(__FILE__) + '/string_util.rb')
describe 'StringUtil' do
  describe '#muti_bytes?' do
    context 'with single byte char "a"' do
      it { StringUtil.multi_bytes?("a").should be_false }
    end

    context 'with multi bytes char "あ"' do
      it { StringUtil.multi_bytes?("あ").should be_true }
    end
  end

  describe '#split_by_size' do
    context 'with "abcdefg", 5' do
      it { StringUtil.split_by_size("abcdefg", 5).should == "abcde" }
    end
    context 'with "abcあdefg", 5' do
      it { StringUtil.split_by_size("abcあdefg", 5).should == "abcあ" }
    end
    context 'with "abcdあefg", 5' do
      it { StringUtil.split_by_size("abcdあefg", 5).should == "abcd" }
    end
    context 'with "abcdeあfg", 5' do
      it { StringUtil.split_by_size("abcdeあfg", 5).should == "abcde" }
    end
    context 'with "aあいbcdefg", 5' do
      it { StringUtil.split_by_size("aあいbcdefg", 5).should == "aあい" }
    end

  end

end