GemaCoreLib
The GeMA Core library
Loading...
Searching...
No Matches
GmDualVector< T > Class Template Reference

A vector tailored for storing mesh related data for the common scenario where elements are added to the mesh but the number of added elements is much smaller than the original mesh size. Although conceptually a single vector, internally the storage is split in two parts. A fixed one with the original mesh information and a second one that grows as new elements are added to the mesh. Since the second part is small, standard vector growing practices are not too hurtfull. More...

#include <gmDualVector.h>

Inheritance diagram for GmDualVector< T >:

Classes

class  iterator
 An iterator class to enable usage of the vector with standard algorithms such as std::sort. More...
 

Public Types

typedef T DataType
 The stored data type.
 

Public Member Functions

 GmDualVector ()
 Default constructor.
 
 GmDualVector (size_t nvalues)
 The constructor initializing the "fixed" part size. The vector contents is NOT initialized.
 
 ~GmDualVector ()
 Destructor.
 
bool isDual () const
 Returns true, since this IS a Dual vector.
 
size_t size () const
 Returns the vector size.
 
const T * iptr (size_t index) const
 Returns a pointer to the index data inside the vector. By the very nature of this class, accessing following items is only safe if you know that they belong to the same internal vector. This is usually guaranteed by knowing that all data bellonging to a cell, for example, is always together in the same vector, but depends on the class usage.
 
T * iptr (size_t index)
 Non-const overload for iptr()
 
const T & operator[] (size_t index) const
 Standard indexing operator.
 
T & operator[] (size_t index)
 Non-const overload for the standard indexing operator.
 
iterator begin ()
 
iterator end ()
 
bool addValues (size_t numAddedValues, bool tight=false)
 Adds numAddedValues to the set, without any initialization. Returns true on success, false on error. On errors previous values are kept. If the vector is empty, the allocated size will be adjusted to exactly the given size. Otherwise, a growing factor of half of the previous size will be used, unless that is insufficient. If tight is set to true, the allocated size will always match the needed size (unless already allocated).
 
void restoreSize (size_t oldNumValues)
 Restores the size of the set to the previous size before the last call to addValues(). Can be called with the current size if addValues was not called.
 
void clear ()
 Clears all stored data returning the object to a default constructed state.
 
void zero (size_t index=0)
 Clears the vector area by setting to zero all entries starting at the given index.
 
size_t usedMemory () const
 Returns an estimative of the memory used by the data set in bytes.
 
int numDumpBuffers () const
 Returns the number of internal buffers used by this implementation.
 
char * dumpBuffer (int i) const
 Returns the i'th internal buffer, i from 0 to numDumpBuffers()-1.
 
size_t dumpBufferSize (int i) const
 Returns the size in bytes of the i'th internal buffer returned by dumpBuffer(i)
 

Protected Attributes

T * _fixedData
 The fixed part of the dual vector with _fixedSize entries.
 
size_t _fixedSize
 The number of entries in _fixedData.
 
GmPODVector< T > _growData
 The variable part.
 

Private Member Functions

 Q_DISABLE_COPY (GmDualVector)
 

Detailed Description

template<class T>
class GmDualVector< T >

A vector tailored for storing mesh related data for the common scenario where elements are added to the mesh but the number of added elements is much smaller than the original mesh size. Although conceptually a single vector, internally the storage is split in two parts. A fixed one with the original mesh information and a second one that grows as new elements are added to the mesh. Since the second part is small, standard vector growing practices are not too hurtfull.

This class provides a basic vector interface tailored for both the ValueSet needs and for usage on the mesh plugin class. This class does NOT allow for reducing the vector size. See GmDualVectorR instead.

Accepts only POD types.

Member Function Documentation

◆ usedMemory()

template<class T >
size_t GmDualVector< T >::usedMemory ( ) const
inline

Returns an estimative of the memory used by the data set in bytes.

  It returns in general the variable memory that depends on the number of set
  entries but does not include constant overheads used by the class for control

The documentation for this class was generated from the following file: