nobu	2019-03-05 09:32:15 +0900 (Tue, 05 Mar 2019)

  New Revision: 67167

  https://svn.ruby-lang.org/cgi-bin/viewvc.cgi?view=revision&revision=67167

  Log:
    string.c: respect the actual encoding
    
    * string.c (rb_enc_str_coderange): respect the actual encoding of
      if a BOM presents, and scan for the actual code range.
      [ruby-core:91662] [Bug #15635]

  Modified files:
    trunk/string.c
    trunk/test/ruby/test_m17n.rb
Index: test/ruby/test_m17n.rb
===================================================================
--- test/ruby/test_m17n.rb	(revision 67166)
+++ test/ruby/test_m17n.rb	(revision 67167)
@@ -269,6 +269,13 @@ class TestM17N < Test::Unit::TestCase https://github.com/ruby/ruby/blob/trunk/test/ruby/test_m17n.rb#L269
     assert_empty(encs, bug10598)
   end
 
+  def test_utf_without_bom_valid
+    encs = [Encoding::UTF_16, Encoding::UTF_32].find_all {|enc|
+      !(+"abcd").encode!(enc).force_encoding(enc).valid_encoding?
+    }
+    assert_empty(encs)
+  end
+
   def test_object_utf16_32_inspect
     EnvUtil.suppress_warning do
       begin
Index: string.c
===================================================================
--- string.c	(revision 67166)
+++ string.c	(revision 67167)
@@ -655,12 +655,13 @@ rb_enc_str_coderange(VALUE str) https://github.com/ruby/ruby/blob/trunk/string.c#L655
     if (cr == ENC_CODERANGE_UNKNOWN) {
 	int encidx = ENCODING_GET(str);
 	rb_encoding *enc = rb_enc_from_index(encidx);
-	if (rb_enc_mbminlen(enc) > 1 && rb_enc_dummy_p(enc)) {
+	if (rb_enc_mbminlen(enc) > 1 && rb_enc_dummy_p(enc) &&
+            rb_enc_mbminlen(enc = get_actual_encoding(encidx, str)) == 1) {
 	    cr = ENC_CODERANGE_BROKEN;
 	}
 	else {
 	    cr = coderange_scan(RSTRING_PTR(str), RSTRING_LEN(str),
-				get_actual_encoding(encidx, str));
+                                enc);
 	}
         ENC_CODERANGE_SET(str, cr);
     }

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