* encoding.c (rb_enc_mbclen): make it never fail.

  (rb_enc_nth): don't check the return value of rb_enc_mbclen.
  (rb_enc_strlen): ditto.
  (rb_enc_precise_mbclen): return needmore(1) if e <= p.
  (rb_enc_get_ascii): new function for extracting ASCII character.

* include/ruby/encoding.h (rb_enc_get_ascii): declared.

* include/ruby/regex.h (ismbchar): removed.

* re.c (rb_reg_expr_str): use rb_enc_get_ascii.
  (unescape_escaped_nonascii): use rb_enc_precise_mbclen to determine
  the termination of escaped non-ASCII character.
  (unescape_nonascii): use rb_enc_precise_mbclen.
  (rb_reg_quote): use rb_enc_get_ascii.
  (rb_reg_regsub): use rb_enc_get_ascii.

* string.c (rb_str_reverse) don't check the return value of
  rb_enc_mbclen.
  (rb_str_split_m): don't call rb_enc_mbclen with e <= p.

* parse.y (is_identchar): use ISASCII.
  (parser_ismbchar): removed.
  (parser_precise_mbclen): new macro.
  (parser_isascii): new macro.
  (parser_tokadd_mbchar): use parser_precise_mbclen to check invalid
  character precisely.
  (parser_tokadd_string): use parser_isascii.
  (parser_yylex): ditto.
  (is_special_global_name): don't call is_identchar with e <= p.
  (rb_enc_symname_p): ditto.

  [ruby-dev:32455]

* ext/tk/sample/tkextlib/vu/canvSticker2.rb: remove coding cookie
  because the encoding is not UTF-8.  [ruby-dev:32475]


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@14131 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

2 files changed, 6 insertions, 6 deletions

diff --git a/test/ruby/test_m17n.rb b/test/ruby/test_m17n.rb
index 02c8dca4d0..df59c88770 100644
--- a/test/ruby/test_m17n.rb
+++ b/test/ruby/test_m17n.rb
@@ -77,8 +77,8 @@ class TestM17N < Test::Unit::TestCase
     assert_raise(SyntaxError) { eval('/\xc2/u') }
     assert_raise(SyntaxError) { eval('/\xe0\x80/u') }
     assert_raise(SyntaxError) { eval('/\xf0\x80\x80/u') }
-    #assert_raise(SyntaxError) { eval('/\xf8\x80\x80\x80/u') }
-    #assert_raise(SyntaxError) { eval('/\xfc\x80\x80\x80\x80/u') }
+    assert_raise(SyntaxError) { eval('/\xf8\x80\x80\x80/u') }
+    assert_raise(SyntaxError) { eval('/\xfc\x80\x80\x80\x80/u') }
 
     # raw 8bit
     assert_raise(SyntaxError) { eval("/\xfe/e") }
@@ -87,7 +87,7 @@ class TestM17N < Test::Unit::TestCase
     # invalid suffix
     assert_raise(SyntaxError) { eval('/\xc2\xff/u') }
     assert_raise(SyntaxError) { eval('/\xc2 /u') }
-    #assert_raise(SyntaxError) { eval('/\xc2\x20/u') }
+    assert_raise(SyntaxError) { eval('/\xc2\x20/u') }
   end
 
   def assert_regexp_generic_encoding(r)
diff --git a/test/ruby/test_regexp.rb b/test/ruby/test_regexp.rb
index 9cb8c4ac1e..84710bb449 100644
--- a/test/ruby/test_regexp.rb
+++ b/test/ruby/test_regexp.rb
@@ -20,7 +20,7 @@ class TestRegexp < Test::Unit::TestCase
 
   def test_yoshidam_net_20041111_2
     assert_raise(RegexpError) do
-      s = "[\xFF-\xFF]"
+      s = "[\xFF-\xFF]".force_encoding("utf-8")
       Regexp.new(s, nil, "u")
     end
   end
@@ -42,8 +42,8 @@ class TestRegexp < Test::Unit::TestCase
     assert_equal :ok, begin
       Regexp.union(
         "a",
-        Regexp.new("\x80".force_encoding("euc-jp")),
-        Regexp.new("\x80".force_encoding("utf-8")))
+        Regexp.new("\xc2\xa1".force_encoding("euc-jp")),
+        Regexp.new("\xc2\xa1".force_encoding("utf-8")))
       :ng
     rescue ArgumentError
       :ok