Adds unicode utility

diff --git a/lib/unicode.rb b/lib/unicode.rb
new file mode 100644 (file)
index 0000000..13fc0c1
--- /dev/null
+++ b/lib/unicode.rb
@@ -0,0 +1,50 @@
+# Unicodeを扱うライブラリ
+#  コード形式: UTF-8
+
+UCS_FORMAT = /^U\+[0-9a-fA-F]+$/
+CJK_UNIFIED = [0x4E00, 0x9FFF]
+#CJK_UNIFIED_EXT_A = [0x3400, 0x4DBF]
+CJK_UNIFIED_EXT_A = [0x3400, 0x4DFF]
+CJK_UNIFIED_EXT_B = [0x20000, 0x2A6DF]
+CJK_UNIFIED_EXT_C = [0x2A700, 0x2B73F]
+# CJK_UNIFIED_EXT_D = [0x?, 0x?]
+
+# 条件：codespaceの順序に並べる
+CODESPACE = [CJK_UNIFIED_EXT_A,
+             CJK_UNIFIED,
+             CJK_UNIFIED_EXT_B,
+             CJK_UNIFIED_EXT_C,
+            ]
+
+class UnicodeUtility
+  def nextCodepoint(codepoint)
+    CODESPACE.each do |scode, ecode|
+      return scode if codepoint < scode
+      if scode<= codepoint and codepoint < ecode then
+        return codepoint + 1
+      elsif codepoint == ecode then
+        codepoint = ecode
+      end
+    end
+    return nil
+  end
+  def ucs2char(ucs)
+    return nil unless ucs =~ UCS_FORMAT
+    code = ucs.slice(2, ucs.size)
+    [code.hex].pack("U*")
+  end
+  def char2ucs(char)
+    # TODO: Check char in codespace
+    # Unicode codespace is a range of integers from 0 to 10FFFF_hex
+    return format("U+%04X", char.unpack("U")[0])
+  end
+  def codepoint2ucs(codepoint)
+    return format("U+%04X", codepoint)
+  end
+  def ucs2codepoint(ucs)
+#    return nil unless ucs =~ UCS_FORMAT
+    code = ucs.slice(2, ucs.size)
+#    [code.hex].pack("U*")
+    return code.hex
+  end
+end

diff --git a/test/unicode_test.rb b/test/unicode_test.rb
new file mode 100644 (file)
index 0000000..bc08b15
--- /dev/null
+++ b/test/unicode_test.rb
@@ -0,0 +1,42 @@
+require File.join(File.dirname(__FILE__), '..', 'lib', 'unicode')
+require 'test/unit'
+ONE_UCS = "U+4E00"
+
+class UnicodeUtilityTest < Test::Unit::TestCase
+  def setup
+    @unicode = UnicodeUtility.new
+  end
+  def test_next_codepoint
+    first_a = CJK_UNIFIED_EXT_A[0]
+    last_a = CJK_UNIFIED_EXT_A[1]
+    first_cjk = CJK_UNIFIED[0]
+    last_cjk = CJK_UNIFIED[1]
+    first_b = CJK_UNIFIED_EXT_B[0]
+    last_c = CJK_UNIFIED_EXT_C[1]
+
+    assert_equal first_a + 1, @unicode.nextCodepoint(first_a)
+    assert_equal first_cjk, @unicode.nextCodepoint(last_a)
+    assert_equal first_cjk + 1, @unicode.nextCodepoint(first_cjk)
+    assert_equal first_b, @unicode.nextCodepoint(last_cjk)
+    assert_equal nil, @unicode.nextCodepoint(last_c)
+  end
+  def test_ucs2char
+    assert_equal "一", @unicode.ucs2char(ONE_UCS)
+    assert_equal "丁", @unicode.ucs2char("U+4E01")
+    assert_equal nil, @unicode.ucs2char("4E01")
+    assert_equal nil, @unicode.ucs2char("")
+  end
+  def test_char2ucs
+    assert_equal ONE_UCS, @unicode.char2ucs("一")
+    assert_equal "U+4E01", @unicode.char2ucs("丁")
+    assert_equal "U+20000", @unicode.char2ucs("𠀀")
+    # TODO: codespace(a range of integers from 0 to 10FFFF_hex)
+    assert_equal "U+0000", @unicode.char2ucs("")
+  end
+  def test_codepoint2ucs
+    assert_equal ONE_UCS, @unicode.codepoint2ucs(0x4E00)
+  end
+  def test_ucs2codepoint2
+    assert_equal 0x4E00, @unicode.ucs2codepoint(ONE_UCS)
+  end
+end