[Boost.utility]

[achinkoff@…]

Having considered ​http://www.research.ibm.com/designpatterns/pubs/ph-jun96.txt I propose singleton implementation.

I had attached "singleton.hpp" to this letter.

Below is the test program:

#include <stdio.h>
#include <singleton.hpp>

class A : public boost::Singleton<A>
{
	//
	// Only Singleton class can create and delete us!
	//
	SINGLETON_IS_MY_FRIEND;
public:
	void foo(){ printf("foo!\n"); }
	A(){ printf("A()\n"); }
	~A(){ printf("~A()\n"); }
};

void main()
{
	// It is not compiled:
	//A a = A::Instance();
	// That is exactly as it should be used:
	A::Instance().foo();
}

Test program outputs:

A()
foo!
~A()

I suggest include "singleton.hpp" into "utility.hpp" header that it can be used along with: boost::noncopyable, etc.

[achinkoff@…]

Thread safe "Singleton" pattern implementation.

[achinkoff@…]

Thread safe "Singleton" pattern implementation - corrected.

[achinkoff@…]

@@ -13,7 +13,7 @@
       */
 
     template<bool> class CTAssert;
-    template<> class CTAssert<false>{ };
+    template<> class CTAssert<true>{ };
 
     /**
       See http://www.research.ibm.com/designpatterns/pubs/ph-jun96.txt
@@ -39,8 +39,8 @@
         SingletonDestroyer()
         {
             typedef typename boost::is_array<Doomed>::type DoomedArrayType;
-            CTAssert<DoomedArrayType::value>
-                ____DOOMED_SINGLETON_MUST_NOT_BE_ARRAY____;
+            bool assert_me = (DoomedArrayType::value == false);
+            CTAssert<assert_me> ____DOOMED_SINGLETON_MUST_NOT_BE_ARRAY____;
             /**
              Suppress unused variable warning
              */

[anonymous]

That is what I mean:

@@ -39,7 +39,7 @@

         SingletonDestroyer()
         {
             typedef typename boost::is_array<Doomed>::type DoomedArrayType;
-            bool assert_me = (DoomedArrayType::value == false);
+            const bool assert_me = (DoomedArrayType::value == false);

             CTAssert<assert_me> ____DOOMED_SINGLETON_MUST_NOT_BE_ARRAY____;
             /**
              Suppress unused variable warning

[Steven Watanabe]

New libraries need to go through the review process. Getting singleton right generically is a lot less trivial than it might seem. I believe that one singleton library was rejected a while ago.

A few other points.

1. This doesn't conform to Boost naming conventions.
2. Double checked locking is broken and this has been well known for ages.
3. BOOST_MPL_ASSERT almost always gives better error messages than hand crafted static assertions.