While I don't quite see it as a bomb, I know that probably just indicates short-sightedness on my part. Regardless, I am in the "just make it work right now" boat, so any help you can offer would be much appreciated.
P.S.: I did manage a temporary hack of archive/detail/iserializer.hpp pointer_iserializer::load_object_ptr() that (badly) assumed that if the passed pointer is not NULL, then it is a pointer to an allocated object so there's no need to create a new object. Coupled with a do-nothing load_construct_data() method this worked great for partial-deserialization, but causes memory corruption for normal deserialization since the assumption is not always true (nor is it a good assumption to try an force on other users).
I finally found a half-decent work-around: I added a template specialization of the boost::archive::detail::pointer_iserializer::load_object_ptr() method for each of the classes which I will be creating manually and then deserializing. On top of this, I disabled tracking for them, enabled exporting and added a call to void_cast_register() to enable (de)serialization by a base-class pointer. Here's some code snippets to elaborate the above (hopefully they display fine): // In the header file class MyClass: public cSimpleModule { ... }; BOOST_CLASS_TRACKING(MyClass, boost::serialization::track_never) namespace boost { namespace archive { namespace detail { template<> void pointer_iserializer<boost::archive::text_iarchive, MyClass>::load_object_ptr( basic_iarchive &ar, void * & x, const unsigned int file_version) const { boost::archive::text_iarchive & ar_impl = boost::serialization::smart_cast_reference<boost::archive::text_iarchive &>(ar); ar_impl >> boost::serialization::make_nvp(NULL, *reinterpret_cast<MyClass*&>(x)); } } } } // In the cc file BOOST_CLASS_EXPORT(MyClass) MyClass::MyClass(void) { boost::serialization::void_cast_register<MyClass, cModule>( static_cast<MyClass*>(NULL), static_cast<cModule*>(NULL)); ... } ... -- Adam