forcing users to use osgDB::readRef*File() methods. The later is preferable as it closes a potential threading bug when using paging databases in conjunction
with the osgDB::Registry Object Cache. This threading bug occurs when one thread gets an object from the Cache via an osgDB::read*File() call where only
a pointer to the object is passed back, so taking a reference to the object is delayed till it gets reassigned to a ref_ptr<>, but at the same time another
thread calls a flush of the Object Cache deleting this object as it's referenceCount is now zero. Using osgDB::readREf*File() makes sure the a ref_ptr<> is
passed back and the referenceCount never goes to zero.
To ensure the OSG builds when OSG_PROVIDE_READFILE is to OFF the many cases of osgDB::read*File() usage had to be replaced with a ref_ptr<> osgDB::readRef*File()
usage. The avoid this change causing lots of other client code to be rewritten to handle the use of ref_ptr<> in place of C pointer I introduced a serious of
templte methods in various class to adapt ref_ptr<> to the underly C pointer to be passed to old OSG API's, example of this is found in include/osg/Group:
bool addChild(Node* child); // old method which can only be used with a Node*
tempalte<class T> bool addChild(const osg::ref_ptr<T>& child) { return addChild(child.get()); } // adapter template method
These changes together cover 149 modified files, so it's a large submission. This extent of changes are warrent to make use of the Object Cache
and multi-threaded loaded more robust.
git-svn-id: http://svn.openscenegraph.org/osg/OpenSceneGraph/trunk@15164 16af8721-9629-0410-8352-f15c8da7e697
with a osg::DefaultUserDataContainer subclassed from this. The user object access methods have now all
been moved from osg::Object into the UserDataContainer class, except for the set/getUserData() methods
that are left in osg::Object for backwards compatibility, and the description list access methods have
been moved back into osg::Node.
main UserObject access methods are now all def
Refactored original UserData and Descriptions strings to be managed alongside the new user object suppport within
a single osg::Object::UserDataContainer.
CID 11447: Unchecked dynamic_cast (FORWARD_NULL)
Dynamic cast to pointer "dynamic_cast <struct osg::NodeCallback *>(nc->clone(this))" can return null.
Assigning null: "first" = "dynamic_cast <struct osg::NodeCallback *>(nc->clone(this))".
The clone() implementation is written using macro's so that it always returns the type of Object
being cloned so it's normally safe to assume that a dynamic_cast<> will always return a valid pointer as long
as the new T that involves creates a valid object. However, if the class being cloned doesn't correctly
implement the clone() method then their potential for the dynamic_cast to fail and will return a NULL and will
result in a memory leak of the object of paraent class that the clone would have defaulted to.
I've tightened up the CopyOp.cpp code to check the return type and added better handling of the clone in the
osg::clone() methods so thay don't have any potential mememory leaks and report warnings to OSG_WARN when
problems are encountered. It may be more apporpriate to throw an exception so will need to ponder this
issue further.
Drawables,StateSet, and osgDB::Registry.
Added cleanup_frame() from to osgProducer::OsgCamerGroup to help with proper
clean of OpenGL objects before exit, and modified osgviewer, osghangglider,
osgwindows examples to do the extra frame call to cleanup_frame() before exit.
which objects have values that vary over the lifetime of the object (DYNAMIC)
and ones that do not vary (STATIC). Removed the equivalent code in
osg::Transform, StateSet and StateAttribute, as these are now encompassed
by the new DataVariance field.
Removed MatrixMode enum from Matrix, and associated fields/parameters from
osg::Transfrom and osg::NodeVisitor, since MatrixMode was not providing
any useful functionality, but made the interface more complex (MatrixMode
was an experimental API)
Added ReferenceFrame field to osg::Transform which allows users to specify
transforms that are relative to their parents (the default, and previous behavior)
or absolute reference frame, which can be used for HUD's, camera relative
light sources etc etc. Note, the view frustum culling for absolute Transform
are disabled, and all their parents up to the root are also automatically
have view frustum culling disabled. However, once passed an absolute Transform
node culling will return to its default state of on, so you can still cull
underneath an absolute transform, its only the culling above which is disabled.
memory manager published at flipcode.com. This can be turned on
with the OSG_USE_MEMORY_MANGER option which then uses custom global
new and delete operators as well as provide osgNew and osgDelete macro's
which add ability to log line and file from which calls are made.
Updated osg,osgUtil,osgDB,osgText and osgPlugins/osg to use osgNew/osgDelete,
and fixed memory leaks highlighted by the new memory manager.
class now combines Cloner and DeepCopy into one class. Cloner and DeepCopy
will be removed in next commit.
Also have added osgcopy app to Demos which shows how the CopyOp have be
subclassed to create users own specific handling of copying.
Have fixed copy constructor problems in GeoSet which fix the deep copy
problem experienced yesterday.
