libogre-1.8-dev 1.8.0+dfsg1-7+b1 / usr / include / OGRE / OgreCommon.h

/*
-----------------------------------------------------------------------------
This source file is part of OGRE
    (Object-oriented Graphics Rendering Engine)
For the latest info, see http://www.ogre3d.org/

Copyright (c) 2000-2012 Torus Knot Software Ltd

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
THE SOFTWARE.
-----------------------------------------------------------------------------
*/
#ifndef __Common_H__
#define __Common_H__
// Common stuff

#include "OgreString.h"

#if defined ( OGRE_GCC_VISIBILITY )
#   pragma GCC visibility push(default)
#endif

#include <utility>
#include <sstream>

#if defined ( OGRE_GCC_VISIBILITY )
#   pragma GCC visibility pop
#endif

namespace Ogre {
	/** \addtogroup Core
	*  @{
	*/
	/** \addtogroup General
	*  @{
	*/

	/// Fast general hashing algorithm
	uint32 _OgreExport FastHash (const char * data, int len, uint32 hashSoFar = 0);
	/// Combine hashes with same style as boost::hash_combine
	template <typename T>
	uint32 HashCombine (uint32 hashSoFar, const T& data)
	{
		return FastHash((const char*)&data, sizeof(T), hashSoFar);
	}


    /** Comparison functions used for the depth/stencil buffer operations and 
		others. */
    enum CompareFunction
    {
        CMPF_ALWAYS_FAIL,
        CMPF_ALWAYS_PASS,
        CMPF_LESS,
        CMPF_LESS_EQUAL,
        CMPF_EQUAL,
        CMPF_NOT_EQUAL,
        CMPF_GREATER_EQUAL,
        CMPF_GREATER
    };

    /** High-level filtering options providing shortcuts to settings the
        minification, magnification and mip filters. */
    enum TextureFilterOptions
    {
        /// Equal to: min=FO_POINT, mag=FO_POINT, mip=FO_NONE
        TFO_NONE,
        /// Equal to: min=FO_LINEAR, mag=FO_LINEAR, mip=FO_POINT
        TFO_BILINEAR,
        /// Equal to: min=FO_LINEAR, mag=FO_LINEAR, mip=FO_LINEAR
        TFO_TRILINEAR,
        /// Equal to: min=FO_ANISOTROPIC, max=FO_ANISOTROPIC, mip=FO_LINEAR
		TFO_ANISOTROPIC
    };

    enum FilterType
    {
        /// The filter used when shrinking a texture
        FT_MIN,
        /// The filter used when magnifying a texture
        FT_MAG,
        /// The filter used when determining the mipmap
        FT_MIP
    };
    /** Filtering options for textures / mipmaps. */
    enum FilterOptions
    {
        /// No filtering, used for FILT_MIP to turn off mipmapping
        FO_NONE,
        /// Use the closest pixel
        FO_POINT,
        /// Average of a 2x2 pixel area, denotes bilinear for MIN and MAG, trilinear for MIP
        FO_LINEAR,
        /// Similar to FO_LINEAR, but compensates for the angle of the texture plane
        FO_ANISOTROPIC
    };

    /** Light shading modes. */
    enum ShadeOptions
    {
        SO_FLAT,
        SO_GOURAUD,
        SO_PHONG
    };

    /** Fog modes. */
    enum FogMode
    {
        /// No fog. Duh.
        FOG_NONE,
        /// Fog density increases  exponentially from the camera (fog = 1/e^(distance * density))
        FOG_EXP,
        /// Fog density increases at the square of FOG_EXP, i.e. even quicker (fog = 1/e^(distance * density)^2)
        FOG_EXP2,
        /// Fog density increases linearly between the start and end distances
        FOG_LINEAR
    };

    /** Hardware culling modes based on vertex winding.
        This setting applies to how the hardware API culls triangles it is sent. */
    enum CullingMode
    {
        /// Hardware never culls triangles and renders everything it receives.
        CULL_NONE = 1,
        /// Hardware culls triangles whose vertices are listed clockwise in the view (default).
        CULL_CLOCKWISE = 2,
        /// Hardware culls triangles whose vertices are listed anticlockwise in the view.
        CULL_ANTICLOCKWISE = 3
    };

    /** Manual culling modes based on vertex normals.
        This setting applies to how the software culls triangles before sending them to the 
		hardware API. This culling mode is used by scene managers which choose to implement it -
		normally those which deal with large amounts of fixed world geometry which is often 
		planar (software culling movable variable geometry is expensive). */
    enum ManualCullingMode
    {
        /// No culling so everything is sent to the hardware.
        MANUAL_CULL_NONE = 1,
        /// Cull triangles whose normal is pointing away from the camera (default).
        MANUAL_CULL_BACK = 2,
        /// Cull triangles whose normal is pointing towards the camera.
        MANUAL_CULL_FRONT = 3
    };

    /** Enumerates the wave types usable with the Ogre engine. */
    enum WaveformType
    {
        /// Standard sine wave which smoothly changes from low to high and back again.
        WFT_SINE,
        /// An angular wave with a constant increase / decrease speed with pointed peaks.
        WFT_TRIANGLE,
        /// Half of the time is spent at the min, half at the max with instant transition between.
        WFT_SQUARE,
        /// Gradual steady increase from min to max over the period with an instant return to min at the end.
        WFT_SAWTOOTH,
        /// Gradual steady decrease from max to min over the period, with an instant return to max at the end.
        WFT_INVERSE_SAWTOOTH,
		/// Pulse Width Modulation. Works like WFT_SQUARE, except the high to low transition is controlled by duty cycle. 
		/// With a duty cycle of 50% (0.5) will give the same output as WFT_SQUARE.
		WFT_PWM
    };

    /** The polygon mode to use when rasterising. */
    enum PolygonMode
    {
		/// Only points are rendered.
        PM_POINTS = 1,
		/// Wireframe models are rendered.
        PM_WIREFRAME = 2,
		/// Solid polygons are rendered.
        PM_SOLID = 3
    };

    /** An enumeration of broad shadow techniques */
    enum ShadowTechnique
    {
        /** No shadows */
        SHADOWTYPE_NONE = 0x00,
		/** Mask for additive shadows (not for direct use, use  SHADOWTYPE_ enum instead)
		*/
		SHADOWDETAILTYPE_ADDITIVE = 0x01,
		/** Mask for modulative shadows (not for direct use, use  SHADOWTYPE_ enum instead)
		*/
		SHADOWDETAILTYPE_MODULATIVE = 0x02,
		/** Mask for integrated shadows (not for direct use, use SHADOWTYPE_ enum instead)
		*/
		SHADOWDETAILTYPE_INTEGRATED = 0x04,
		/** Mask for stencil shadows (not for direct use, use  SHADOWTYPE_ enum instead)
		*/
		SHADOWDETAILTYPE_STENCIL = 0x10,
		/** Mask for texture shadows (not for direct use, use  SHADOWTYPE_ enum instead)
		*/
		SHADOWDETAILTYPE_TEXTURE = 0x20,
		
        /** Stencil shadow technique which renders all shadow volumes as
            a modulation after all the non-transparent areas have been 
            rendered. This technique is considerably less fillrate intensive 
            than the additive stencil shadow approach when there are multiple
            lights, but is not an accurate model. 
        */
        SHADOWTYPE_STENCIL_MODULATIVE = 0x12,
        /** Stencil shadow technique which renders each light as a separate
            additive pass to the scene. This technique can be very fillrate
            intensive because it requires at least 2 passes of the entire
            scene, more if there are multiple lights. However, it is a more
            accurate model than the modulative stencil approach and this is
            especially apparent when using coloured lights or bump mapping.
        */
        SHADOWTYPE_STENCIL_ADDITIVE = 0x11,
        /** Texture-based shadow technique which involves a monochrome render-to-texture
            of the shadow caster and a projection of that texture onto the 
            shadow receivers as a modulative pass. 
        */
        SHADOWTYPE_TEXTURE_MODULATIVE = 0x22,
		
        /** Texture-based shadow technique which involves a render-to-texture
            of the shadow caster and a projection of that texture onto the 
            shadow receivers, built up per light as additive passes. 
			This technique can be very fillrate intensive because it requires numLights + 2 
			passes of the entire scene. However, it is a more accurate model than the 
			modulative approach and this is especially apparent when using coloured lights 
			or bump mapping.
        */
        SHADOWTYPE_TEXTURE_ADDITIVE = 0x21,

		/** Texture-based shadow technique which involves a render-to-texture
		of the shadow caster and a projection of that texture on to the shadow
		receivers, with the usage of those shadow textures completely controlled
		by the materials of the receivers.
		This technique is easily the most flexible of all techniques because 
		the material author is in complete control over how the shadows are
		combined with regular rendering. It can perform shadows as accurately
		as SHADOWTYPE_TEXTURE_ADDITIVE but more efficiently because it requires
		less passes. However it also requires more expertise to use, and 
		in almost all cases, shader capable hardware to really use to the full.
		@note The 'additive' part of this mode means that the colour of
		the rendered shadow texture is by default plain black. It does
		not mean it does the adding on your receivers automatically though, how you
		use that result is up to you.
		*/
		SHADOWTYPE_TEXTURE_ADDITIVE_INTEGRATED = 0x25,
		/** Texture-based shadow technique which involves a render-to-texture
			of the shadow caster and a projection of that texture on to the shadow
			receivers, with the usage of those shadow textures completely controlled
			by the materials of the receivers.
			This technique is easily the most flexible of all techniques because 
			the material author is in complete control over how the shadows are
			combined with regular rendering. It can perform shadows as accurately
			as SHADOWTYPE_TEXTURE_ADDITIVE but more efficiently because it requires
			less passes. However it also requires more expertise to use, and 
			in almost all cases, shader capable hardware to really use to the full.
			@note The 'modulative' part of this mode means that the colour of
			the rendered shadow texture is by default the 'shadow colour'. It does
			not mean it modulates on your receivers automatically though, how you
			use that result is up to you.
		*/
		SHADOWTYPE_TEXTURE_MODULATIVE_INTEGRATED = 0x26
    };

    /** An enumeration describing which material properties should track the vertex colours */
    typedef int TrackVertexColourType;
    enum TrackVertexColourEnum {
        TVC_NONE        = 0x0,
        TVC_AMBIENT     = 0x1,        
        TVC_DIFFUSE     = 0x2,
        TVC_SPECULAR    = 0x4,
        TVC_EMISSIVE    = 0x8
    };

    /** Sort mode for billboard-set and particle-system */
    enum SortMode
    {
        /** Sort by direction of the camera */
        SM_DIRECTION,
        /** Sort by distance from the camera */
        SM_DISTANCE
    };

    /** Defines the frame buffer types. */
    enum FrameBufferType {
        FBT_COLOUR  = 0x1,
        FBT_DEPTH   = 0x2,
        FBT_STENCIL = 0x4
    };

	/** Flags for the Instance Manager when calculating ideal number of instances per batch */
	enum InstanceManagerFlags
	{
		/** Forces an amount of instances per batch low enough so that vertices * numInst < 65535
			since usually improves performance. In HW instanced techniques, this flag is ignored
		*/
		IM_USE16BIT		= 0x0001,

		/** The num. of instances is adjusted so that as few pixels as possible are wasted
			in the vertex texture */
		IM_VTFBESTFIT	= 0x0002,

		/** Use a limited number of skeleton animations shared among all instances. 
		Update only that limited amount of animations in the vertex texture.*/
		IM_VTFBONEMATRIXLOOKUP = 0x0004,

		IM_USEBONEDUALQUATERNIONS = 0x0008,

		/** Use one weight per vertex when recommended (i.e. VTF). */
		IM_USEONEWEIGHT = 0x0010,

		/** All techniques are forced to one weight per vertex. */
		IM_FORCEONEWEIGHT = 0x0020,

		IM_USEALL		= IM_USE16BIT|IM_VTFBESTFIT|IM_USEONEWEIGHT
	};
    
	
	/** A hashed vector.
	*/
	template <typename T>
	class HashedVector
	{
	public:
		typedef std::vector<T, STLAllocator<T, GeneralAllocPolicy> > VectorImpl;
	protected:
		VectorImpl mList;
		mutable uint32 mListHash;
		mutable bool mListHashDirty;

		void addToHash(const T& newPtr) const
		{
			mListHash = FastHash((const char*)&newPtr, sizeof(T), mListHash);
		}
		void recalcHash() const
		{
			mListHash = 0;
			for (const_iterator i = mList.begin(); i != mList.end(); ++i)
				addToHash(*i);
			mListHashDirty = false;
			
		}

	public:
		typedef typename VectorImpl::value_type value_type;
		typedef typename VectorImpl::pointer pointer;
		typedef typename VectorImpl::reference reference;
		typedef typename VectorImpl::const_reference const_reference;
		typedef typename VectorImpl::size_type size_type;
		typedef typename VectorImpl::difference_type difference_type;
		typedef typename VectorImpl::iterator iterator;
		typedef typename VectorImpl::const_iterator const_iterator;
		typedef typename VectorImpl::reverse_iterator reverse_iterator;
		typedef typename VectorImpl::const_reverse_iterator const_reverse_iterator;

		void dirtyHash()
		{
			mListHashDirty = true;
		}
		bool isHashDirty() const
		{
			return mListHashDirty;
		}

		iterator begin() 
		{ 
			// we have to assume that hash needs recalculating on non-const
			dirtyHash();
			return mList.begin(); 
		}
		iterator end() { return mList.end(); }
		const_iterator begin() const { return mList.begin(); }
		const_iterator end() const { return mList.end(); }
		reverse_iterator rbegin() 
		{ 
			// we have to assume that hash needs recalculating on non-const
			dirtyHash();
			return mList.rbegin(); 
		}
		reverse_iterator rend() { return mList.rend(); }
		const_reverse_iterator rbegin() const { return mList.rbegin(); }
		const_reverse_iterator rend() const { return mList.rend(); }
		size_type size() const { return mList.size(); }
		size_type max_size() const { return mList.max_size(); }
		size_type capacity() const { return mList.capacity(); }
		bool empty() const { return mList.empty(); }
		reference operator[](size_type n) 
		{ 
			// we have to assume that hash needs recalculating on non-const
			dirtyHash();
			return mList[n]; 
		}
		const_reference operator[](size_type n) const { return mList[n]; }
		reference at(size_type n) 
		{ 
			// we have to assume that hash needs recalculating on non-const
			dirtyHash();
			return mList.const_iterator(n); 
		}
		const_reference at(size_type n) const { return mList.at(n); }
		HashedVector() : mListHash(0), mListHashDirty(false) {}
		HashedVector(size_type n) : mList(n), mListHash(0), mListHashDirty(n > 0) {}
		HashedVector(size_type n, const T& t) : mList(n, t), mListHash(0), mListHashDirty(n > 0) {}
		HashedVector(const HashedVector<T>& rhs) 
			: mList(rhs.mList), mListHash(rhs.mListHash), mListHashDirty(rhs.mListHashDirty) {}

		template <class InputIterator>
		HashedVector(InputIterator a, InputIterator b)
			: mList(a, b), mListHashDirty(false)
		{
			dirtyHash();
		}

		~HashedVector() {}
		HashedVector<T>& operator=(const HashedVector<T>& rhs)
		{
			mList = rhs.mList;
			mListHash = rhs.mListHash;
			mListHashDirty = rhs.mListHashDirty;
			return *this;
		}

		void reserve(size_t t) { mList.reserve(t); }
		reference front() 
		{ 
			// we have to assume that hash needs recalculating on non-const
			dirtyHash();
			return mList.front(); 
		}
		const_reference front() const { return mList.front(); }
		reference back()  
		{ 
			// we have to assume that hash needs recalculating on non-const
			dirtyHash();
			return mList.back(); 
		}
		const_reference back() const { return mList.back(); }
		void push_back(const T& t)
		{ 
			mList.push_back(t);
			// Quick progressive hash add
			if (!isHashDirty())
				addToHash(t);
		}
		void pop_back()
		{
			mList.pop_back();
			dirtyHash();
		}
		void swap(HashedVector<T>& rhs)
		{
			mList.swap(rhs.mList);
			dirtyHash();
		}
		iterator insert(iterator pos, const T& t)
		{
			bool recalc = (pos != end());
			iterator ret = mList.insert(pos, t);
			if (recalc)
				dirtyHash();
			else
				addToHash(t);
			return ret;
		}

		template <class InputIterator>
		void insert(iterator pos,
			InputIterator f, InputIterator l)
		{
			mList.insert(pos, f, l);
			dirtyHash();
		}

		void insert(iterator pos, size_type n, const T& x)
		{
			mList.insert(pos, n, x);
			dirtyHash();
		}

		iterator erase(iterator pos)
		{
			iterator ret = mList.erase(pos);
			dirtyHash();
			return ret;
		}
		iterator erase(iterator first, iterator last)
		{
			iterator ret = mList.erase(first, last);
			dirtyHash();
			return ret;
		}
		void clear()
		{
			mList.clear();
			mListHash = 0;
			mListHashDirty = false;
		}

		void resize(size_type n, const T& t = T())
		{
			bool recalc = false;
			if (n != size())
				recalc = true;

			mList.resize(n, t);
			if (recalc)
				dirtyHash();
		}

		bool operator==(const HashedVector<T>& b)
		{ return mListHash == b.mListHash; }

		bool operator<(const HashedVector<T>& b)
		{ return mListHash < b.mListHash; }


		/// Get the hash value
		uint32 getHash() const 
		{ 
			if (isHashDirty())
				recalcHash();

			return mListHash; 
		}
	public:



	};

	class Light;
	typedef HashedVector<Light*> LightList;



    typedef map<String, bool>::type UnaryOptionList;
    typedef map<String, String>::type BinaryOptionList;

	/// Name / value parameter pair (first = name, second = value)
	typedef map<String, String>::type NameValuePairList;

    /// Alias / Texture name pair (first = alias, second = texture name)
    typedef map<String, String>::type AliasTextureNamePairList;

        template< typename T > struct TRect
        {
          T left, top, right, bottom;
          TRect() : left(0), top(0), right(0), bottom(0) {}
          TRect( T const & l, T const & t, T const & r, T const & b )
            : left( l ), top( t ), right( r ), bottom( b )
          {
          }
          TRect( TRect const & o )
            : left( o.left ), top( o.top ), right( o.right ), bottom( o.bottom )
          {
          }
          TRect & operator=( TRect const & o )
          {
            left = o.left;
            top = o.top;
            right = o.right;
            bottom = o.bottom;
            return *this;
          }
          T width() const
          {
            return right - left;
          }
          T height() const
          {
            return bottom - top;
          }
		  bool isNull() const
		  {
			  return width() == 0 || height() == 0;
		  }
		  void setNull()
		  {
			  left = right = top = bottom = 0;
		  }
		  TRect & merge(const TRect& rhs)
		  {
			  if (isNull())
			  {
				  *this = rhs;
			  }
			  else if (!rhs.isNull())
			  {
				  left = std::min(left, rhs.left);
				  right = std::max(right, rhs.right);
				  top = std::min(top, rhs.top);
				  bottom = std::max(bottom, rhs.bottom);
			  }

			  return *this;

		  }
		  TRect intersect(const TRect& rhs) const
		  {
			  TRect ret;
			  if (isNull() || rhs.isNull())
			  {
				  // empty
				  return ret;
			  }
			  else
			  {
				  ret.left = std::max(left, rhs.left);
				  ret.right = std::min(right, rhs.right);
				  ret.top = std::max(top, rhs.top);
				  ret.bottom = std::min(bottom, rhs.bottom);
			  }

			  if (ret.left > ret.right || ret.top > ret.bottom)
			  {
				  // no intersection, return empty
				  ret.left = ret.top = ret.right = ret.bottom = 0;
			  }

			  return ret;

		  }

        };
		template<typename T>
		std::ostream& operator<<(std::ostream& o, const TRect<T>& r)
		{
			o << "TRect<>(l:" << r.left << ", t:" << r.top << ", r:" << r.right << ", b:" << r.bottom << ")";
			return o;
		}

        /** Structure used to define a rectangle in a 2-D floating point space.
        */
        typedef TRect<float> FloatRect;

		/** Structure used to define a rectangle in a 2-D floating point space, 
			subject to double / single floating point settings.
		*/
		typedef TRect<Real> RealRect;

        /** Structure used to define a rectangle in a 2-D integer space.
        */
        typedef TRect< long > Rect;

        /** Structure used to define a box in a 3-D integer space.
         	Note that the left, top, and front edges are included but the right, 
         	bottom and back ones are not.
         */
        struct Box
        {
            size_t left, top, right, bottom, front, back;
			/// Parameterless constructor for setting the members manually
            Box()
				: left(0), top(0), right(1), bottom(1), front(0), back(1)
            {
            }
            /** Define a box from left, top, right and bottom coordinates
            	This box will have depth one (front=0 and back=1).
            	@param	l	x value of left edge
            	@param	t	y value of top edge
            	@param	r	x value of right edge
            	@param	b	y value of bottom edge
            	@note Note that the left, top, and front edges are included 
 		           	but the right, bottom and back ones are not.
            */
            Box( size_t l, size_t t, size_t r, size_t b ):
                left(l),
                top(t),   
                right(r),
                bottom(b),
                front(0),
                back(1)
            {
          		assert(right >= left && bottom >= top && back >= front);
            }
            /** Define a box from left, top, front, right, bottom and back
            	coordinates.
            	@param	l	x value of left edge
            	@param	t	y value of top edge
            	@param  ff  z value of front edge
            	@param	r	x value of right edge
            	@param	b	y value of bottom edge
            	@param  bb  z value of back edge
            	@note Note that the left, top, and front edges are included 
 		           	but the right, bottom and back ones are not.
            */
            Box( size_t l, size_t t, size_t ff, size_t r, size_t b, size_t bb ):
                left(l),
                top(t),   
                right(r),
                bottom(b),
                front(ff),
                back(bb)
            {
          		assert(right >= left && bottom >= top && back >= front);
            }
            
            /// Return true if the other box is a part of this one
            bool contains(const Box &def) const
            {
            	return (def.left >= left && def.top >= top && def.front >= front &&
					def.right <= right && def.bottom <= bottom && def.back <= back);
            }
            
            /// Get the width of this box
            size_t getWidth() const { return right-left; }
            /// Get the height of this box
            size_t getHeight() const { return bottom-top; }
            /// Get the depth of this box
            size_t getDepth() const { return back-front; }
        };

    
	
	/** Locate command-line options of the unary form '-blah' and of the
        binary form '-blah foo', passing back the index of the next non-option.
    @param numargs, argv The standard parameters passed to the main method
    @param unaryOptList Map of unary options (i.e. those that do not require a parameter).
        Should be pre-populated with, for example '-e' in the key and false in the 
        value. Options which are found will be set to true on return.
    @param binOptList Map of binary options (i.e. those that require a parameter
        e.g. '-e afile.txt').
        Should be pre-populated with, for example '-e' and the default setting. 
        Options which are found will have the value updated.
    */
    int _OgreExport findCommandLineOpts(int numargs, char** argv, UnaryOptionList& unaryOptList, 
        BinaryOptionList& binOptList);

	/// Generic result of clipping
	enum ClipResult
	{
		/// Nothing was clipped
		CLIPPED_NONE = 0,
		/// Partially clipped
		CLIPPED_SOME = 1, 
		/// Everything was clipped away
		CLIPPED_ALL = 2
	};

	/// Render window creation parameters.
	struct RenderWindowDescription
	{
		String				name;
		unsigned int		width;
		unsigned int		height;
		bool				useFullScreen;
		NameValuePairList	miscParams;
	};

	/// Render window creation parameters container.
	typedef vector<RenderWindowDescription>::type RenderWindowDescriptionList;

	/// Render window container.
	typedef vector<RenderWindow*>::type RenderWindowList;

	/// Utility class to generate a sequentially numbered series of names
	class _OgreExport NameGenerator
	{
	protected:
		String mPrefix;
		unsigned long long int mNext;
		OGRE_AUTO_MUTEX
	public:
		NameGenerator(const NameGenerator& rhs)
			: mPrefix(rhs.mPrefix), mNext(rhs.mNext) {}
		
		NameGenerator(const String& prefix) : mPrefix(prefix), mNext(1) {}

		/// Generate a new name
		String generate()
		{
			OGRE_LOCK_AUTO_MUTEX
			std::ostringstream s;
			s << mPrefix << mNext++;
			return s.str();
		}

		/// Reset the internal counter
		void reset()
		{
			OGRE_LOCK_AUTO_MUTEX
			mNext = 1ULL;
		}

		/// Manually set the internal counter (use caution)
		void setNext(unsigned long long int val)
		{
			OGRE_LOCK_AUTO_MUTEX
			mNext = val;
		}

		/// Get the internal counter
		unsigned long long int getNext() const
		{
			// lock even on get because 64-bit may not be atomic read
			OGRE_LOCK_AUTO_MUTEX
			return mNext;
		}




	};

	/** Template class describing a simple pool of items.
	*/
	template <typename T>
	class Pool
	{
	protected:
		typedef typename list<T>::type ItemList;
		ItemList mItems;
		OGRE_AUTO_MUTEX
	public:
		Pool() {} 
		virtual ~Pool() {}

		/** Get the next item from the pool.
		@return pair indicating whether there was a free item, and the item if so
		*/
		virtual std::pair<bool, T> removeItem()
		{
			OGRE_LOCK_AUTO_MUTEX
			std::pair<bool, T> ret;
			if (mItems.empty())
			{
				ret.first = false;
			}
			else
			{
				ret.first = true;
				ret.second = mItems.front();
				mItems.pop_front();
			}
			return ret;
		}

		/** Add a new item to the pool. 
		*/
		virtual void addItem(const T& i)
		{
			OGRE_LOCK_AUTO_MUTEX
			mItems.push_front(i);
		}
		/// Clear the pool
		virtual void clear()
		{
			OGRE_LOCK_AUTO_MUTEX
			mItems.clear();
		}



	};
	/** @} */
	/** @} */
}

#endif