* include/ruby/oniguruma.h (OnigEncodingTypeST): add precise_ret

  argument for mbc_to_code.
  (ONIGENC_MBC_TO_CODE): provide NULL for precise_ret.
  (ONIGENC_MBC_PRECISE_CODEPOINT): defined.

* include/ruby/encoding.h (rb_enc_mbc_precise_codepoint): defined.

* regenc.h (onigenc_single_byte_mbc_to_code): precise_ret argument
  added.
  (onigenc_mbn_mbc_to_code): ditto.

* regenc.c (onigenc_single_byte_mbc_to_code): precise_ret argument
  added.
  (onigenc_mbn_mbc_to_code): ditto.

* string.c (count_utf8_lead_bytes_with_word): removed.
  (str_utf8_nth): removed.
  (str_utf8_offset): removed.
  (str_strlen): UTF-8 codepoint oriented optimization removed.
  (rb_str_substr): ditto.
  (enc_succ_char): use rb_enc_mbc_precise_codepoint.
  (enc_pred_char): ditto.
  (rb_str_succ): ditto.

* encoding.c (rb_enc_ascget): check length with
  rb_enc_mbc_precise_codepoint.
  (rb_enc_codepoint): use rb_enc_mbc_precise_codepoint.

* regexec.c (string_cmp_ic): add text_end argument.
  (match_at): check end of character after exact string matches.

* enc/utf_8.c (graphme_table): defined for extended graphme cluster
  boundary.
  (grapheme_cmp): defined.
  (get_grapheme_properties): defined.
  (grapheme_boundary_p): defined.
  (MAX_BYTES_LENGTH): defined.
  (comb_char_enc_len): defined.
  (mbc_to_code0): extracted from mbc_to_code.
  (mbc_to_code): use mbc_to_code0.
  (left_adjust_combchar_head): defined.
  (utf_8): use a extended graphme cluster as a unit.

* enc/unicode.c (onigenc_unicode_mbc_case_fold): use
  ONIGENC_MBC_PRECISE_CODEPOINT to extract codepoints.
  (onigenc_unicode_get_case_fold_codes_by_str): ditto.

* enc/euc_jp.c (mbc_to_code): follow mbc_to_code field change.
  use onigenc_mbn_mbc_to_code.

* enc/shift_jis.c (mbc_to_code): ditto.

* enc/emacs_mule.c (mbc_to_code): ditto.

* enc/gbk.c (gbk_mbc_to_code): follow mbc_to_code field and
  onigenc_mbn_mbc_to_code change.

* enc/cp949.c (cp949_mbc_to_code): ditto.

* enc/big5.c (big5_mbc_to_code): ditto.

* enc/euc_tw.c (euctw_mbc_to_code): ditto.

* enc/euc_kr.c (euckr_mbc_to_code): ditto.

* enc/gb18030.c (gb18030_mbc_to_code): ditto.

* enc/utf_32be.c (utf32be_mbc_to_code): follow mbc_to_code field
  change.

* enc/utf_16be.c (utf16be_mbc_to_code): ditto.

* enc/utf_32le.c (utf32le_mbc_to_code): ditto.

* enc/utf_16le.c (utf16le_mbc_to_code): ditto.



git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@19389 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

1 files changed, 19 insertions, 0 deletions

diff --git a/test/ruby/test_m17n.rb b/test/ruby/test_m17n.rb
index 855653fa81..ac68b882a6 100644
--- a/test/ruby/test_m17n.rb
+++ b/test/ruby/test_m17n.rb
@@ -1292,4 +1292,23 @@ class TestM17N < Test::Unit::TestCase
     assert(("".center(1, "\x80".force_encoding("utf-8")); true),
            "moved from btest/knownbug, [ruby-dev:33807]")
   end
+
+  def test_combchar
+    assert_equal(1, "\u{304B 3099}".length)
+    assert_equal("\u{304B 3099}", "\u{304B 3099}"[0])
+    assert_equal(nil, "\u{304B 3099}"[1])
+    assert_equal(nil, "\u{304B 3099}".index("\u3099"))
+  end
+
+  def test_combchar_regexp
+    assert_match(/\A.\z/, "\u304B\u3099")
+    assert_nil(/\u3099/ =~ "\u304B\u3099")
+    assert_nil(/a|b/ =~ "a\u3099")
+    assert_nil(/\u0100|\u0111/ =~ "\u0100\u3099")
+  end
+
+  def test_combchar_codepoint
+    assert_equal([0x30BB, 0x309A], "\u30BB\u309A".codepoints.to_a)
+  end
+
 end