merge revision(s) ade5673: [Backport #20190] (ruby#10300)

nurse · web-flow · commit ce372be903e5 · 2024-03-20T13:40:46.000Z
Fix coderange of invalid_encoding_string.&lt;&lt;(ord)

	Appending valid encoding character can change coderange from invalid to valid.
	Example: "\x95".force_encoding('sjis')&lt;&lt;0x5C will be a valid string "\x{955C}"
diff --git a/string.c b/string.c
@@ -3522,8 +3522,12 @@ rb_str_concat(VALUE str1, VALUE str2)
         }
         rb_str_resize(str1, pos+len);
         memcpy(RSTRING_PTR(str1) + pos, buf, len);
-        if (cr == ENC_CODERANGE_7BIT && code > 127)
+        if (cr == ENC_CODERANGE_7BIT && code > 127) {
             cr = ENC_CODERANGE_VALID;
+        }
+        else if (cr == ENC_CODERANGE_BROKEN) {
+            cr = ENC_CODERANGE_UNKNOWN;
+        }
         ENC_CODERANGE_SET(str1, cr);
     }
     return str1;
diff --git a/test/ruby/test_string.rb b/test/ruby/test_string.rb
@@ -301,6 +301,9 @@ def test_LSHIFT # '<<'
     assert_raise(RangeError, bug) {S("a".force_encoding(Encoding::UTF_8)) << -1}
     assert_raise(RangeError, bug) {S("a".force_encoding(Encoding::UTF_8)) << 0x81308130}
     assert_nothing_raised {S("a".force_encoding(Encoding::GB18030)) << 0x81308130}
+
+    s = "\x95".force_encoding(Encoding::SJIS).tap(&:valid_encoding?)
+    assert_predicate(s << 0x5c, :valid_encoding?)
   end
 
   def test_MATCH # '=~'
diff --git a/version.h b/version.h
@@ -11,7 +11,7 @@
 # define RUBY_VERSION_MINOR RUBY_API_VERSION_MINOR
 #define RUBY_VERSION_TEENY 0
 #define RUBY_RELEASE_DATE RUBY_RELEASE_YEAR_STR"-"RUBY_RELEASE_MONTH_STR"-"RUBY_RELEASE_DAY_STR
-#define RUBY_PATCHLEVEL 15
+#define RUBY_PATCHLEVEL 16
 
 #include "ruby/version.h"
 #include "ruby/internal/abi.h"

Original file line number	Diff line number	Diff line change
`@@ -3522,8 +3522,12 @@ rb_str_concat(VALUE str1, VALUE str2)`
`3522`	`3522`	`}`
`3523`	`3523`	`rb_str_resize(str1, pos+len);`
`3524`	`3524`	`memcpy(RSTRING_PTR(str1) + pos, buf, len);`
`3525`		`- if (cr == ENC_CODERANGE_7BIT && code > 127)`
	`3525`	`+ if (cr == ENC_CODERANGE_7BIT && code > 127) {`
`3526`	`3526`	`cr = ENC_CODERANGE_VALID;`
	`3527`	`+ }`
	`3528`	`+ else if (cr == ENC_CODERANGE_BROKEN) {`
	`3529`	`+ cr = ENC_CODERANGE_UNKNOWN;`
	`3530`	`+ }`
`3527`	`3531`	`ENC_CODERANGE_SET(str1, cr);`
`3528`	`3532`	`}`
`3529`	`3533`	`return str1;`