Torque Game Engine Documentation
TGE Version 1.5.2

TSShapeAlloc Class Reference

#include <tsShapeAlloc.h>

Detailed Description

Alloc structure used in the reading/writing of shapes.

In read mode we assemble contents of 32-bit, 16-bit, and 8-bit buffers into a single destination buffer.

In write mode we dissemble a stream of memory (which may be scattered in physical memory) into 32-bit, 16-bit, 8-bit, Point3F, and Point2F buffers using function calls similar to the read calls.

Read usage: 1. call "setRead" with each incoming memory buffers and clear=true. 2. run through set of operations for allocating and transfering memory to target buffer these are the operations under "DECLARE_ALLOC" that call readOnly in the .cc file. 3. call "doAlloc" to create buffer exactly as large as needed. 4. repeat step 1 & 2 to do the actual transfer of memory, except with clear=false (note: first time through nothing was copied to the shape, we only kept track of the size of the transfer). 5. call getBuffer to get the target (destination buffer)

write usage: 1. call "setWrite" (no parameters). 2. run through set of operations for allocating and transfering memory to internal buffers these are the operations under "DECLARE_ALLOC" that call writeOnly in the .cc file. 3. call getBuffer32 and getBufferSize32 to get 32-bit buffer and size. Similarly for 16-bit, 8-bit (getBuffer16, getBuffer8).

TSShape::assesmbleShape and TSShape::dissembleShape can be used as examples

Public Types

enum {
  ReadMode = 0,
  WriteMode = 1,
  PageSize = 1024
}

Public Member Functions

void setRead (S32 *buff32, S16 *buff16, S8 *buff8, bool clear)

void setWrite ()

void doAlloc ()

void align32 ()

align on dword boundary

S8 * getBuffer ()

S32 getSize ()

void setSkipMode (bool skip)

void checkGuard ()

void setGuard ()

Private Attributes

S32 mMode

read or write

S32 * mMemBuffer32

reading and writing (when reading these are the input; when writing these are the output)

S16 * mMemBuffer16

S8 * mMemBuffer8

S32 mSize32

for writing only...

S32 mSize16

S32 mSize8

S32 mFullSize32

S32 mFullSize16

S32 mFullSize8

S32 mMemGuard32

reading and writing...

S16 mMemGuard16

S8 mMemGuard8

S32 mSaveGuard32

reading

S16 mSaveGuard16

S8 mSaveGuard8

S8 * mDest

reading only...this is the output

S32 mSize

S32 mMult

mult incoming sizes by this (when 0, then mDest doesn't grow --> skip mode)