Update DgmOctree syntax (#100)

* Improve the syntax of ScalarFieldTools::computeCellGaussianFilter

* Syntax fix

* Restore the definition of DgmOctree::MultiThreadingWrapper in the header to avoid issues when linking CCCoreLib statically on Windows

Author: dgirardeau

include/DgmOctree.h

@@ -24,7 +24,6 @@ namespace CCCoreLib
 	class GenericIndexedCloudPersist;
 	class GenericProgressCallback;
 	class NormalizedProgress;
-	struct MultiThreadingWrapper;
 
 	//! The octree structure used throughout the library
 	/** Implements the GenericOctree interface.
@@ -349,7 +348,7 @@ namespace CCCoreLib
 			- (NormalizedProgress*) optional (normalized) progress callback
 			- return success
 		**/
-		using octreeCellFunc = bool (*)(const octreeCell &, void **, NormalizedProgress *);
+		using octreeCellFunc = bool (*)(const octreeCell&, void**, NormalizedProgress*);
 
 		/******************************/
 		/**          METHODS         **/
@@ -361,7 +360,7 @@ namespace CCCoreLib
 		explicit DgmOctree(GenericIndexedCloudPersist* cloud);
 
 		//! DgmOctree destructor
-		~DgmOctree() override;
+		~DgmOctree() override = default;
 
 		//! Clears the octree
 		virtual void clear();
@@ -1114,7 +1113,7 @@ namespace CCCoreLib
 
 		//! Computes statistics about cells for a given level of subdivision
 		/** This method requires some computation, therefore it shouldn't be
-		called too often.
+			called too often.
 			\param level the level of subdivision
 		**/
 		void computeCellsStatistics(unsigned char level);
@@ -1128,19 +1127,19 @@ namespace CCCoreLib
 			\param level the level of subdivision
 		**/
 		void getNeighborCellsAround(const Tuple3i& cellPos,
-			cellIndexesContainer &neighborCellsIndexes,
-			int neighbourhoodLength,
-			unsigned char level) const;
+									cellIndexesContainer &neighborCellsIndexes,
+									int neighbourhoodLength,
+									unsigned char level) const;
 
 		//! Gets point in the neighbourhing cells of a specific cell
 		/** \warning May throw a std::bad_alloc exception if memory is insufficient.
 			\param nNSS NN search parameters (from which are used: cellPos, pointsInNeighbourCells and level)
 			\param neighbourhoodLength the new distance (in terms of cells) at which to look for neighbour cells
 			\param getOnlyPointsWithValidScalar whether to ignore points having an invalid associated scalar value
 		**/
-		void getPointsInNeighbourCellsAround(NearestNeighboursSearchStruct &nNSS,
-			int neighbourhoodLength,
-			bool getOnlyPointsWithValidScalar = false) const;
+		void getPointsInNeighbourCellsAround(	NearestNeighboursSearchStruct &nNSS,
+												int neighbourhoodLength,
+												bool getOnlyPointsWithValidScalar = false) const;
 
 		//! Returns the index of a given cell represented by its code
 		/** Same algorithm as the other "getCellIndex" method, but in an optimized form.
@@ -1215,8 +1214,39 @@ namespace CCCoreLib
 		//! Std. dev. of cell population per level of subdivision
 		double m_stdDevCellPopulation[MAX_OCTREE_LEVEL + 1];
 
-		//! Multithreading wrapper
-		MultiThreadingWrapper* m_MT_wrapper;
+		//! Octree cell description helper struct
+		struct octreeCellDesc
+		{
+			DgmOctree::CellCode truncatedCode;
+			unsigned i1, i2;
+			unsigned char level;
+		};
+
+		//! Structure containing objects needed to run octree operations in parallel
+		struct MultiThreadingWrapper
+		{
+			octreeCellFunc cellFunc = nullptr;
+			bool cellFuncSuccess = true;
+			DgmOctree* octree = nullptr;
+			GenericProgressCallback* progressCb = nullptr;
+			NormalizedProgress* normProgressCb = nullptr;
+			void** userParams = nullptr;
+
+			void reset()
+			{
+				cellFunc = nullptr;
+				cellFuncSuccess = true;
+				octree = nullptr;
+				normProgressCb = nullptr;
+				progressCb = nullptr;
+				userParams = nullptr;
+			}
+
+			void launchOctreeCellFunc(const octreeCellDesc& desc);
+		};
+
+		//! Multi-threading wrapper structure
+		MultiThreadingWrapper m_MT_wrapper;
 
 		//! Whether the octree build is in progress
 		std::atomic<bool> m_buildInProgress;

include/ScalarFieldTools.h

@@ -71,16 +71,15 @@ namespace CCCoreLib
 												DgmOctree* theOctree = nullptr);
 
 		//! Computes a spatial gaussian filter on a scalar field associated to a point cloud
-		/** The "amplitutde" of the gaussian filter must be precised (sigma).
-			As 99% of the gaussian distribution is between -3*sigma and +3*sigma
-			around the mean value, this filter will only look for neighbouring
-			points (around each point) in a sphere of radius 3*sigma.
-			It also permits to use the filter as a bilateral filter. Where the wights are computed also considering the
-			distance of the neighbor's scalar value from the current point scalar value. (weighted with gaussian as distances are)
-			Warning: this method assumes the input scalar field is different from output.
+		/** The "amplitutde" of the Gaussian filter must be specified (sigma).
+			As 99% of the Gaussian distribution is between -3*sigma and +3*sigma around the mean value,
+			this filter will only look for neighbors within a sphere of radius 3*sigma.
+			One can also use the filter as a Bilateral filter. In this case the weights are computed considering the
+			difference of the neighbors SF values with the current point SF value (also following a Gaussian distribution).
+			Warning: this method assumes the input scalar field is different from the output one.
 			\param sigma filter variance
 			\param theCloud a point cloud (associated to scalar values)
-			\param sigmaSF the sigma for the bilateral filter. when different than -1 turns the gaussian filter into a bilateral filter
+			\param sigmaSF if strictly positive, the variance for the Bilateral filter
 			\param progressCb the client application can get some notification of the process progress through this callback mechanism (see GenericProgressCallback)
 			\param theOctree the octree, if it has already been computed
 			\return success