naruse	2018-11-24 20:53:19 +0900 (Sat, 24 Nov 2018)

  New Revision: 65954

  https://svn.ruby-lang.org/cgi-bin/viewvc.cgi?view=revision&revision=65954

  Log:
    Don't use single byte optimization on grapheme clusters
    
    Unicode Text Segmentation considers CRLF as a character. [Bug #15337]

  Modified files:
    trunk/string.c
    trunk/test/ruby/test_string.rb
Index: test/ruby/test_string.rb
===================================================================
--- test/ruby/test_string.rb	(revision 65953)
+++ test/ruby/test_string.rb	(revision 65954)
@@ -973,6 +973,7 @@ CODE https://github.com/ruby/ruby/blob/trunk/test/ruby/test_string.rb#L973
 
   def test_each_grapheme_cluster
     [
+      "\u{0D 0A}",
       "\u{20 200d}",
       "\u{600 600}",
       "\u{600 20}",
Index: string.c
===================================================================
--- string.c	(revision 65953)
+++ string.c	(revision 65954)
@@ -8459,7 +8459,7 @@ rb_str_each_grapheme_cluster_size(VALUE https://github.com/ruby/ruby/blob/trunk/string.c#L8459
     rb_encoding *enc = rb_enc_from_index(ENCODING_GET(str));
     const char *ptr, *end;
 
-    if (!rb_enc_unicode_p(enc) || single_byte_optimizable(str)) {
+    if (!rb_enc_unicode_p(enc)) {
 	return rb_str_length(str);
     }
 
@@ -8487,7 +8487,7 @@ rb_str_enumerate_grapheme_clusters(VALUE https://github.com/ruby/ruby/blob/trunk/string.c#L8487
     rb_encoding *enc = rb_enc_from_index(ENCODING_GET(str));
     const char *ptr, *end;
 
-    if (!rb_enc_unicode_p(enc) || single_byte_optimizable(str)) {
+    if (!rb_enc_unicode_p(enc)) {
 	return rb_str_enumerate_chars(str, ary);
     }
 

--
ML: ruby-changes@q...
Info: http://www.atdot.net/~ko1/quickml/