This pull request implements the integration of the Garbage Collector with the rest of the code base. NodePtr has been re-aliased to GCNode, and an initialization mechanism has been implemented that allows the create_node() procedure to be dynamically bound to a procedure that creates Data objects that are managed by the garbage collector. All unit tests have been fixed by introducing GC instances into the appropriate test environments and calling init_create_node(). All unit tests should hopefully be passing again. We need to figure out a way to trick the type checker into allowing us to choose whether to bind create_node to the old procedure that returned a std::shared_ptr<Data> or to the new one that returns a GCNode without complaining about the NodePtr aliasing. For now, this is a quick and dirty fix. Please review and let me know if there are any questions/concerns/oversights.
This pull request implements the integration of the Garbage Collector with the rest of the code base.
NodePtrhas been re-aliased toGCNode, and an initialization mechanism has been implemented that allows thecreate_node()procedure to be dynamically bound to a procedure that createsDataobjects that are managed by the garbage collector. All unit tests have been fixed by introducing GC instances into the appropriate test environments and callinginit_create_node(). All unit tests should hopefully be passing again. We need to figure out a way to trick the type checker into allowing us to choose whether to bindcreate_nodeto the old procedure that returned astd::shared_ptr<Data>or to the new one that returns aGCNodewithout complaining about theNodePtraliasing. For now, this is a quick and dirty fix. Please review and let me know if there are any questions/concerns/oversights.