encoding.c: defer finding encoding

* encoding.c (enc_m_loader): defer finding encoding object not to
  be infected by marshal source.  [ruby-core:71793] [Bug #11760]
* marshal.c (r_object0): enable compatible loader on USERDEF
  class.  the loader function is called with the class itself,
  instead of an allocated object, and the loaded data.
* marshal.c (compat_allocator_table): intialize
  compat_allocator_tbl on demand.
* object.c (rb_undefined_alloc): extract from rb_obj_alloc.

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@52856 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

6 files changed, 59 insertions, 5 deletions

diff --git a/ChangeLog b/ChangeLog
index 73fbb1491f..31ce5bcb42 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,17 @@
+Wed Dec  2 16:27:19 2015  Nobuyoshi Nakada  <nobu@ruby-lang.org>
+
+	* encoding.c (enc_m_loader): defer finding encoding object not to
+	  be infected by marshal source.  [ruby-core:71793] [Bug #11760]
+
+	* marshal.c (r_object0): enable compatible loader on USERDEF
+	  class.  the loader function is called with the class itself,
+	  instead of an allocated object, and the loaded data.
+
+	* marshal.c (compat_allocator_table): intialize
+	  compat_allocator_tbl on demand.
+
+	* object.c (rb_undefined_alloc): extract from rb_obj_alloc.
+
 Wed Dec  2 15:12:43 2015  SHIBATA Hiroshi  <hsbt@ruby-lang.org>
 
 	* configure.in: Fixed double negative comments.
diff --git a/encoding.c b/encoding.c
index 5b9f2a0331..fe720767c2 100644
--- a/encoding.c
+++ b/encoding.c
@@ -1266,6 +1266,14 @@ enc_compatible_p(VALUE klass, VALUE str1, VALUE str2)
 
 /* :nodoc: */
 static VALUE
+enc_s_alloc(VALUE klass)
+{
+    rb_undefined_alloc(klass);
+    return Qnil;
+}
+
+/* :nodoc: */
+static VALUE
 enc_dump(int argc, VALUE *argv, VALUE self)
 {
     rb_check_arity(argc, 0, 1);
@@ -1276,6 +1284,13 @@ enc_dump(int argc, VALUE *argv, VALUE self)
 static VALUE
 enc_load(VALUE klass, VALUE str)
 {
+    return str;
+}
+
+/* :nodoc: */
+static VALUE
+enc_m_loader(VALUE klass, VALUE str)
+{
     return enc_find(klass, str);
 }
 
@@ -1902,7 +1917,7 @@ Init_Encoding(void)
     int i;
 
     rb_cEncoding = rb_define_class("Encoding", rb_cObject);
-    rb_undef_alloc_func(rb_cEncoding);
+    rb_define_alloc_func(rb_cEncoding, enc_s_alloc);
     rb_undef_method(CLASS_OF(rb_cEncoding), "new");
     rb_define_method(rb_cEncoding, "to_s", enc_name, 0);
     rb_define_method(rb_cEncoding, "inspect", enc_inspect, 0);
@@ -1934,6 +1949,8 @@ Init_Encoding(void)
     for (i = 0; i < enc_table.count; ++i) {
 	rb_ary_push(list, enc_new(enc_table.list[i].enc));
     }
+
+    rb_marshal_define_compat(rb_cEncoding, Qnil, NULL, enc_m_loader);
 }
 
 /* locale insensitive ctype functions */
diff --git a/internal.h b/internal.h
index d3db0a6104..1018d0bd30 100644
--- a/internal.h
+++ b/internal.h
@@ -969,6 +969,7 @@ rb_float_new_inline(double d)
 void rb_obj_copy_ivar(VALUE dest, VALUE obj);
 VALUE rb_obj_equal(VALUE obj1, VALUE obj2);
 VALUE rb_class_search_ancestor(VALUE klass, VALUE super);
+NORETURN(void rb_undefined_alloc(VALUE klass));
 double rb_num_to_dbl(VALUE val);
 VALUE rb_obj_dig(int argc, VALUE *argv, VALUE self, VALUE notfound);
 
diff --git a/marshal.c b/marshal.c
index 74d66a208a..3b7f79d614 100644
--- a/marshal.c
+++ b/marshal.c
@@ -128,6 +128,8 @@ mark_marshal_compat_t(void *tbl)
     st_foreach(tbl, mark_marshal_compat_i, 0);
 }
 
+static st_table *compat_allocator_table(void);
+
 void
 rb_marshal_define_compat(VALUE newclass, VALUE oldclass, VALUE (*dumper)(VALUE), VALUE (*loader)(VALUE, VALUE))
 {
@@ -146,7 +148,7 @@ rb_marshal_define_compat(VALUE newclass, VALUE oldclass, VALUE (*dumper)(VALUE),
     compat->dumper = dumper;
     compat->loader = loader;
 
-    st_insert(compat_allocator_tbl, (st_data_t)allocator, (st_data_t)compat);
+    st_insert(compat_allocator_table(), (st_data_t)allocator, (st_data_t)compat);
 }
 
 #define MARSHAL_INFECTION FL_TAINT
@@ -1831,6 +1833,7 @@ r_object0(struct load_arg *arg, int *ivp, VALUE extmod)
 	    VALUE name = r_unique(arg);
 	    VALUE klass = path2class(name);
 	    VALUE data;
+	    st_data_t d;
 
 	    if (!rb_obj_respond_to(klass, s_load, TRUE)) {
 		rb_raise(rb_eTypeError, "class %"PRIsVALUE" needs to have method `_load'",
@@ -1843,7 +1846,11 @@ r_object0(struct load_arg *arg, int *ivp, VALUE extmod)
 	    }
 	    v = load_funcall(arg, klass, s_load, 1, &data);
 	    v = r_entry(v, arg);
-            v = r_leave(v, arg);
+	    if (st_lookup(compat_allocator_tbl, (st_data_t)rb_get_alloc_func(klass), &d)) {
+		marshal_compat_t *compat = (marshal_compat_t*)d;
+		v = compat->loader(klass, v);
+	    }
+	    v = r_post_proc(v, arg);
 	}
         break;
 
@@ -2217,13 +2224,19 @@ Init_marshal(void)
     rb_define_const(rb_mMarshal, "MAJOR_VERSION", INT2FIX(MARSHAL_MAJOR));
     /* minor version */
     rb_define_const(rb_mMarshal, "MINOR_VERSION", INT2FIX(MARSHAL_MINOR));
+}
 
+static st_table *
+compat_allocator_table(void)
+{
+    if (compat_allocator_tbl) return compat_allocator_tbl;
     compat_allocator_tbl = st_init_numtable();
 #undef RUBY_UNTYPED_DATA_WARNING
 #define RUBY_UNTYPED_DATA_WARNING 0
     compat_allocator_tbl_wrapper =
 	Data_Wrap_Struct(rb_cData, mark_marshal_compat_t, 0, compat_allocator_tbl);
     rb_gc_register_mark_object(compat_allocator_tbl_wrapper);
+    return compat_allocator_tbl;
 }
 
 VALUE
diff --git a/object.c b/object.c
index 5023f64b0d..ff2db0b45a 100644
--- a/object.c
+++ b/object.c
@@ -1770,6 +1770,13 @@ rb_class_initialize(int argc, VALUE *argv, VALUE klass)
     return klass;
 }
 
+void
+rb_undefined_alloc(VALUE klass)
+{
+    rb_raise(rb_eTypeError, "allocator undefined for %"PRIsVALUE,
+	     klass);
+}
+
 /*
  *  call-seq:
  *     class.allocate()   ->   obj
@@ -1806,8 +1813,7 @@ rb_obj_alloc(VALUE klass)
     }
     allocator = rb_get_alloc_func(klass);
     if (!allocator) {
-	rb_raise(rb_eTypeError, "allocator undefined for %"PRIsVALUE,
-		 klass);
+	rb_undefined_alloc(klass);
     }
 
     RUBY_DTRACE_CREATE_HOOK(OBJECT, rb_class2name(klass));
diff --git a/test/ruby/test_encoding.rb b/test/ruby/test_encoding.rb
index 63c56e2cd3..abe4317ff6 100644
--- a/test/ruby/test_encoding.rb
+++ b/test/ruby/test_encoding.rb
@@ -33,6 +33,9 @@ class TestEncoding < Test::Unit::TestCase
       assert_raise(TypeError) { e.dup }
       assert_raise(TypeError) { e.clone }
       assert_equal(e.object_id, Marshal.load(Marshal.dump(e)).object_id)
+      assert_not_predicate(e, :tainted?)
+      Marshal.load(Marshal.dump(e).taint)
+      assert_not_predicate(e, :tainted?, '[ruby-core:71793] [Bug #11760]')
     end
   end