From c97ec525ef38952a83fae7c38ab5c0432cb82d09 Mon Sep 17 00:00:00 2001
From: nobu <nobu@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>
Date: Wed, 6 Sep 2017 13:11:44 +0000
Subject: string.c: fix false coderange

* string.c (rb_enc_str_scrub): enc can differ from the actual
  encoding of the string, the cached coderange is useless then.
  [ruby-core:82674] [Bug #13874]

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59763 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
---
 test/ruby/test_transcode.rb | 27 ++++++++++++++++++++-------
 1 file changed, 20 insertions(+), 7 deletions(-)

(limited to 'test')

diff --git a/test/ruby/test_transcode.rb b/test/ruby/test_transcode.rb
index bdd1788be8..085d84ae21 100644
--- a/test/ruby/test_transcode.rb
+++ b/test/ruby/test_transcode.rb
@@ -2180,17 +2180,19 @@ class TestTranscode < Test::Unit::TestCase
 
   def test_valid_dummy_encoding
     bug9314 = '[ruby-core:59354] [Bug #9314]'
-    assert_separately(%W[- -- #{bug9314}], <<-'end;')
-    bug = ARGV.shift
-    result = assert_nothing_raised(TypeError, bug) {break "test".encode(Encoding::UTF_16)}
-    assert_equal("\xFE\xFF\x00t\x00e\x00s\x00t", result.b, bug)
-    result = assert_nothing_raised(TypeError, bug) {break "test".encode(Encoding::UTF_32)}
-    assert_equal("\x00\x00\xFE\xFF\x00\x00\x00t\x00\x00\x00e\x00\x00\x00s\x00\x00\x00t", result.b, bug)
+    assert_separately(%W[- -- #{bug9314}], "#{<<~"begin;"}\n#{<<~'end;'}")
+    begin;
+      bug = ARGV.shift
+      result = assert_nothing_raised(TypeError, bug) {break "test".encode(Encoding::UTF_16)}
+      assert_equal("\xFE\xFF\x00t\x00e\x00s\x00t", result.b, bug)
+      result = assert_nothing_raised(TypeError, bug) {break "test".encode(Encoding::UTF_32)}
+      assert_equal("\x00\x00\xFE\xFF\x00\x00\x00t\x00\x00\x00e\x00\x00\x00s\x00\x00\x00t", result.b, bug)
     end;
   end
 
   def test_loading_race
-    assert_separately([], <<-'end;') #do
+    assert_separately([], "#{<<~"begin;"}\n#{<<~'end;'}")
+    begin;
       bug11277 = '[ruby-dev:49106] [Bug #11277]'
       num = 2
       th = (0...num).map do |i|
@@ -2207,6 +2209,17 @@ class TestTranscode < Test::Unit::TestCase
     end;
   end
 
+  def test_scrub_encode_with_coderange
+    bug = '[ruby-core:82674] [Bug #13874]'
+    s = "\xe5".b
+    u = Encoding::UTF_8
+    assert_equal("?", s.encode(u, u, invalid: :replace, replace: "?"),
+                 "should replace invalid byte")
+    assert_predicate(s, :valid_encoding?, "any char is valid in binary")
+    assert_equal("?", s.encode(u, u, invalid: :replace, replace: "?"),
+                 "#{bug} coderange should not have side effects")
+  end
+
   def test_universal_newline
     bug11324 = '[ruby-core:69841] [Bug #11324]'
     usascii = Encoding::US_ASCII
-- 
cgit v1.2.3