Merge image related changes from the render branch. This includes the image

tile cache code in imbuf, but it is not hooked up to the render engine.


Imbuf module: some small refactoring and removing a lot of unused or old code
(about 6.5k lines).

* Added a ImFileType struct with callbacks to make adding an file format type,
  or making changes to the API easier.
* Move imbuf init/exit code into IMB_init()/IMB_exit() functions.
* Increased mipmap levels from 10 to 20, you run into this limit already with
  a 2k image.
* Removed hamx, amiga, anim5 format support.
* Removed colormap saving, only simple colormap code now for reading tga.
* Removed gen_dynlibtiff.py, editing this is almost as much work as just
  editing the code directly.
* Functions removed that were only used for sequencer plugin API:
  IMB_anim_nextpic, IMB_clever_double, IMB_antialias, IMB_gamwarp,
  IMB_scalefieldImBuf, IMB_scalefastfieldImBuf, IMB_onethird, IMB_halflace,
  IMB_dit0, IMB_dit2, IMB_cspace
* Write metadata info into OpenEXR images. Can be viewed with the command
  line utility 'exrheader'

For the image tile cache code, see this page:
http://wiki.blender.org/index.php/Dev:2.5/Source/Imaging/ImageTileCache

1 files changed, 0 insertions, 580 deletions

diff --git a/source/blender/imbuf/intern/cmap.c b/source/blender/imbuf/intern/cmap.c
deleted file mode 100644
index 49edc93b507..00000000000
--- a/source/blender/imbuf/intern/cmap.c
+++ /dev/null
@@ -1,580 +0,0 @@
-/**
- * cmap.c
- *
- * $Id$
- *
- * ***** BEGIN GPL LICENSE BLOCK *****
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version 2
- * of the License, or (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software Foundation,
- * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
- *
- * The Original Code is Copyright (C) 2001-2002 by NaN Holding BV.
- * All rights reserved.
- *
- * The Original Code is: all of this file.
- *
- * Contributor(s): none yet.
- *
- * ***** END GPL LICENSE BLOCK *****
- */
-
-#include <ctype.h>
-#include "BLI_blenlib.h"
-
-#include "imbuf.h"
-#include "imbuf_patch.h"
-#include "IMB_imbuf_types.h"
-#include "IMB_imbuf.h"
-
-#include "IMB_cmap.h"
-
-static short *lastcube = 0;
-static uchar *lastcoltab = 0;
-static short lastmaxcol;
-static short lastmincol;
-static short lastcbits;
-short alpha_col0 = FALSE;
-
-extern void IMB_free_cache_limiter();
-
-/*
- * there still is a bug here. If you want to convert an image to a 1 bit colormap you get
- * a black image. All conversion to less than 4 bits is too dark anyway.
- */
-
-void IMB_freeImBufdata(void)
-{
-	if (lastcube) free(lastcube);
-	lastcube= 0;
-	if (lastcoltab) free(lastcoltab);
-	lastcoltab= 0;
-	IMB_free_cache_limiter();
-}
-
-
-int IMB_alpha_to_col0(int value)
-{
-	int old;
-	
-	old = alpha_col0;
-	alpha_col0 = value;
-	return (old);
-}
-
-
-void imb_losecmapbits(struct ImBuf *ibuf, unsigned int *coltab)
-{
-	int i,bits;
-	unsigned int col, and1, and2, *rect;
-
-	if (ibuf == 0) return;
-	if (ibuf->rect == 0) return;
-	if (ibuf->cbits == 0) return;
-	if (ibuf->cbits >= 8) return;
-
-/*
-	bij cbits = 5:
-	and1 = 11100000;
-	bij cbits = 6:
-	and1 = 11000000;
-*/
-
-	bits = ibuf->cbits;
-	and1 = ((1 << (8-bits)) - 1) & 0xff;
-	and1 |= (and1 << 24) + (and1 << 16) + (and1 << 8);
-	and2 = ~and1;
-	and1 <<= bits;
-
-	rect = ibuf->rect;
-	for (i = ibuf->x * ibuf->y ; i > 0; i--) {
-		col = rect[0]; 
-		*rect++ = col - ((col & and1) >> bits);
-	}
-	
-	if (coltab){
-		for (i = 0 ; i < ibuf->maxcol ; i++) {
-			col = coltab[i];
-			coltab[i] = (col - ((col & and1) >> bits)) & and2;
-		}
-	}
-}
-
-
-static void addcmapbits(struct ImBuf *ibuf)
-{
-	int i,bits;
-	int div,mul;
-	uchar * cmap;
-	
-	if (ibuf == 0) return;
-	if (ibuf->cmap == 0) return;
-	if (ibuf->cbits == 0) return;
-	if (ibuf->cbits >= 8) return;
-
-	bits = ibuf->cbits;
-
-	/* bits = 4 -> div = 0xf0
-	 * bits = 5 -> div = 0xf8
-	 */
-	 
-	div = ((1 << bits) - 1) << (8 - bits);
-	mul = 0xffff / div;
-	
-	if (ibuf->cmap){
-		cmap = (uchar *) ibuf->cmap;
-		for (i = 0 ; i < ibuf->maxcol ; i++){
-			cmap[1] = (mul * cmap[1]) >> 8;
-			cmap[2] = (mul * cmap[2]) >> 8;
-			cmap[3] = (mul * cmap[3]) >> 8;
-			cmap += 4;
-		}
-	}
-}
-
-
-static short addplanetocube(short *cube, short *plane, int minx, int miny, int sizep, int addcx, int addcy, int sizec, int col)
-{
-	short done = FALSE;
-	int x, numx, numy, skipc, skipp, temp;
-
-	/* clip first */
-
-	numx = numy = sizep;
-
-	temp = minx + sizep - 1;
-	if (temp > sizec) numx -= temp - sizec;
-
-	temp = miny + sizep - 1;
-	if (temp > sizec) numy -= temp - sizec;
-
-	if (minx < 0){
-		plane -= minx;
-		cube -= minx * addcx;
-		numx += minx;
-	}
-
-	if (miny < 0){
-		plane -= miny * sizep;
-		cube -= miny * addcy;
-		numy += miny;
-	}
-
-	skipc = addcy - (numx * addcx);
-	skipp = sizep - numx;
-
-	for (; numy > 0 ; numy--){
-		for (x = numx ; x > 0; x--) {
-			
-			if (plane[0] < cube[1]) {
-			
-				cube[0] = col;
-				cube[1] = plane[0];
-				done = TRUE;
-			}
-			plane ++;
-			cube += addcx;
-		}
-		plane += skipp;
-		cube += skipc;
-	}
-
-	return (done);
-}
-
-
-
-short *imb_coldeltatab(unsigned char *coltab, short mincol, short maxcol, short cbits)
-{
-	short max, *quadr, *_quadr, *_cube, *cube, *_plane, done, nocol;
-	unsigned int addcb, addcg, addcr, sizep;
-	uchar *_colp, *colp, *col;
-	int i, j, k, addcube;
-	int r, g, b;
-
-	max = (1 << cbits) - 1;
-	nocol = maxcol - mincol;
-	coltab += 4 * mincol;
-	
-	/* reduce colors to the right amount of bits */
-
-	{
-		unsigned int * lctab, and;
-
-		lctab = (unsigned int *) coltab;
-		and = max << (8 - cbits);
-		and = and + (and << 8) + (and << 16) + (and << 24);
-		for (i=nocol-1 ; i >= 0 ; i--) lctab[i] = (lctab[i] & and) >> (8 - cbits);
-	}
-
-	/* is this data the same as previous ? */
-
-	if (lastcube){
-		if (mincol == lastmincol && maxcol == lastmaxcol && cbits == lastcbits){
-			if (lastcoltab){
-				if (memcmp(lastcoltab, coltab, 4 * nocol) == 0) return(lastcube);
-			}
-		}
-	}
-	if (lastcube) free(lastcube);
-	if (lastcoltab) free(lastcoltab);
-
-	lastcube = 0; 
-	lastcoltab = 0;
-	_cube = malloc(2 * (1 << (3 * cbits)) * sizeof(short));
-	_plane = malloc((2 * max + 1) * (2 * max + 1) * sizeof(short));
-	_quadr = malloc((2 * max + 1) * sizeof(short));
-	_colp = malloc(6 * nocol);
-
-	if (_cube == 0 || _plane == 0 || _quadr == 0 || _colp == 0){
-		if (_cube) free(_cube);
-		if (_plane) free(_plane);
-		if (_quadr) free(_quadr);
-		if (_colp) free(_colp);
-		return(0);
-	}
-
-	lastcoltab = malloc(4 * nocol);
-	if (lastcoltab) memcpy(lastcoltab, coltab, 4 * nocol);
-	lastcube = _cube;
-	lastmincol = mincol;
-	lastmaxcol = maxcol;
-	lastcbits = cbits;
-
-	/* cube initialise */
-
-	cube = _cube;
-	for (i = (1 << (3 * cbits)); i > 0 ; i--){
-		cube[0] = 0;
-		cube[1] = 32767;
-		cube += 2;
-	}
-
-	/* mak error look up table */
-
-	{
-		unsigned int delta;
-
-		quadr = _quadr + max + 1;
-		quadr[0] = 0;
-		delta = 3;
-		for (i = 1 ; i <= max ; i++){
-			quadr[i] = quadr[-i] = delta;
-			delta += i + 3;
-		}
-	}
-
-	/* colorplane initialise */
-
-	for (i = 6 * nocol - 1; i >= 0; i--) _colp[i] = 1;
-
-	addcr = 2;
-	addcg = (addcr << cbits);
-	addcb = (addcg << cbits);
-
-	/* fill in first round */
-
-	{
-		unsigned int ofs;
-
-		col = coltab;
-		cube = _cube;
-
-		for (i = 0 ; i < nocol ; i++){
-			ofs = (col[3] * addcr) + (col[2] * addcg) + (col[1] * addcb);
-			/* color been filled in -> then skip */
-			if (cube[ofs + 1]) cube[ofs] = i + mincol;
-			cube[ofs + 1] = 0;
-			col += 4;
-		}
-	}
-
-	for (i = 1; i <= max ; i++){
-		colp = _colp;
-		col = coltab;
-		done = FALSE;
-		sizep = 2*i +1;
-
-		/* plane initialise */
-		{
-			unsigned int delta;
-			short *plane;
-
-			plane = _plane;
-			for (j = -i ; j <= i; j++){
-				delta = quadr[i] + quadr[j];
-				for (k = -i; k <= i; k++){
-					*plane++ = delta + quadr[k];
-				}
-			}
-		}
-
-		for (j = mincol; j < maxcol; j++){
-			b = col[1] - i;
-			g = col[2] - i;
-			r = col[3] - i;
-			
-			addcube= (addcr * r) + (addcg * g) + (addcb * b);
-			/* PRINT4(d, d, d, d, addcube, r, g, b); */
-			/* if(addcube >= 2 * (1 << (3 * cbits))) { */
-			/* 	printf("maxerror: %d %d\n", addcube, 2 * (1 << (3 * cbits))); */
-				/* add_cube= 2 * (1 << (3 * cbits)) -1; */
-			/* } */
-			cube = _cube + addcube;
-
-			if (colp[0]){
-				if (b < 0) colp[0] = 0;
-				else done |= colp[0] = addplanetocube(cube, _plane, r, g, sizep, addcr, addcg, max, j);
-			}
-			if (colp[1]){
-				if (g < 0) colp[1] = 0;
-				else done |= colp[1] = addplanetocube(cube, _plane, r, b, sizep, addcr, addcb, max, j);
-			}
-			if (colp[2]){
-				if (r < 0) colp[2] = 0;
-				else done |= colp[2] = addplanetocube(cube, _plane, b, g, sizep, addcb, addcg, max, j);
-			}
-			if (colp[3]){
-				if ((b + sizep - 1) > max) colp[3] = 0;
-				else done |= colp[3] = addplanetocube(cube + (sizep -1) * addcb, _plane, r, g, sizep, addcr,
-					addcg, max, j);
-			}
-			if (colp[4]){
-				if ((g + sizep - 1) > max) colp[4] = 0;
-				else done |= colp[4] = addplanetocube(cube + (sizep -1) * addcg, _plane, r, b, sizep, addcr,
-					addcb, max, j);
-			}
-			if (colp[5]){
-				if ((r + sizep - 1) > max) colp[5] = 0;
-				else done |= colp[5] = addplanetocube(cube + (sizep -1) * addcr, _plane, b, g, sizep, addcb,
-					addcg, max, j);
-			}
-
-			colp += 6;
-			col += 4;
-		}
-		if (done == 0) break;
-	}
-
-	free(_quadr);
-	free(_plane);
-	free(_colp);
-	return(_cube);
-}
-
-
-static void convcmap(struct ImBuf* ibuf, short *deltab, short cbits)
-{
-	unsigned int *rect;
-	short x,y;
-	unsigned int col;
-	unsigned int bbits,gbits,rbits;
-	unsigned int bmask,gmask,rmask;
-
-	bbits = 24 - 3 * cbits - 1;
-	gbits = 16 - 2 * cbits - 1;
-	rbits = 8 - cbits - 1;
-
-	rmask = ((1 << cbits) - 1) << (8 - cbits);
-	gmask = rmask << 8;
-	bmask = gmask << 8;
-
-	rect =(unsigned int *)ibuf->rect;
-
-	for(y=ibuf->y;y>0;y--){
-		for(x=ibuf->x;x>0;x--){
-			col = *rect;
-			col = ((col & bmask) >> bbits) + ((col & gmask) >> gbits) + ((col & rmask) >> rbits);
-			*rect++ = deltab[col];
-		}
-	}
-}
-
-short IMB_converttocmap(struct ImBuf *ibuf)
-{
-	unsigned int *coltab;
-	short *deltab=0, cbits;
-	int i;
-	int mincol, mask = 0;
-	struct ImBuf * abuf = 0;
-	unsigned int * rect, * arect;
-	
-	cbits = 5;
-	if (ibuf->cmap == 0) return(0);
-	
-	if ((ibuf->cbits > 0) && (ibuf->cbits <8)) cbits = ibuf->cbits;
-
-	coltab = calloc(ibuf->maxcol, sizeof(unsigned int));
-	if (coltab == 0) return(0);
-	memcpy(coltab, ibuf->cmap, ibuf->maxcol * sizeof(unsigned int));
-	
-	mincol = ibuf->mincol;	
-	if (alpha_col0) {
-		if (mincol == 0) mincol = 1;
-		abuf = IMB_dupImBuf(ibuf);
-	}
-	
-	imb_losecmapbits(ibuf, coltab);
-	deltab = imb_coldeltatab((uchar *) coltab, mincol ,ibuf->maxcol, cbits);
-	
-	if (deltab == 0) {
-		free(coltab);
-		if (abuf) IMB_freeImBuf(abuf);
-		return(0);
-	}
-
-
-	IMB_dit0(ibuf,1,cbits);
-	IMB_dit0(ibuf,2,cbits);
-	IMB_dit0(ibuf,3,cbits);
-	convcmap(ibuf, deltab, cbits);
-	
-	if (abuf) {
-		/* convert alpha to color 0 */
-		rect = ibuf->rect;
-		arect = abuf->rect;
-		
-		if (alpha_col0 == 1) mask = 0xff000000; /* alpha ==  0 -> 0 */
-		if (alpha_col0 == 2) mask = 0x80000000; /* alpha < 128 -> 0 */
-		
-		for (i = ibuf->x * ibuf->y; i > 0; i--) {
-			if ((*arect++ & mask) == 0) rect[0] = 0;
-			rect++;
-		}
-		
-		IMB_freeImBuf(abuf);
-	}
-
-	free(coltab);
-	
-	return (TRUE);
-}
-
-
-void imb_makecolarray(struct ImBuf *ibuf, unsigned char *mem, short nocols)
-{
-	short i,bits = 0;
-	uchar *cmap;
-
-	/* what's the theory behind this? */
-	
-	nocols = ibuf->maxcol;
-
-	if (ibuf->cmap){
-		cmap = (uchar *) ibuf->cmap;
-		for (i = 0; i < nocols; i++){
-			cmap[3] = mem[0];
-			cmap[2] = mem[1];
-			cmap[1] = mem[2];
-			cmap[0] = 0;
-			
-			bits |= mem[0] | mem[1] | mem[2];
-			mem += 3;
-			cmap += 4;
-		}
-
-		/* patch voor AdPro II */
-		if (IS_ham(ibuf)){
-			i = ibuf->depth - 2;
-			bits = ((1 << i) - 1) << (8 - i);
-			for (i=0 ; i<nocols ; i++) ibuf->cmap[i] &= (bits << 24) + (bits << 16) + (bits << 8) + bits;
-		}
-
-		if ((bits & 0x1f) == 0){
-			ibuf->cbits = 3;
-		} else if ((bits & 0x0f) == 0){
-			ibuf->cbits = 4;
-		} else if ((bits & 0x07) == 0){
-			ibuf->cbits = 5;
-		} else if ((bits & 0x03) == 0){
-			ibuf->cbits = 6;
-		} else ibuf->cbits = 8;
-
-		addcmapbits(ibuf);
-
-		if (IS_hbrite(ibuf)){
-			for (i=31;i>=0;i--){
-				ibuf->cmap[i+32] =  (ibuf->cmap[i]  & 0xfefefefe) >> 1;
-			}
-		}
-		
-		if (IS_amiga(ibuf)){
-			cmap = (uchar * ) (ibuf->cmap + 1);
-			for (i = 1; i < nocols; i++){
-				cmap[0] = 0xff;
-				cmap += 4;
-			}
-		}
-	}
-}
-
-/* temporal... rects now are rgba, cmaps are abgr */
-#define SWITCH_INT(a)	{char s_i, *p_i; p_i= (char *)&(a); s_i= p_i[0]; p_i[0]= p_i[3]; p_i[3]= s_i; s_i= p_i[1]; p_i[1]= p_i[2]; p_i[2]= s_i; }
-
-void IMB_applycmap(struct ImBuf *ibuf)
-{
-	unsigned int *rect, *cmap;
-	int x, y, i, col, code;
-	int *mask = 0;
-	
-	if (ibuf == 0) return;
-	if (ibuf->rect == 0 || ibuf->cmap == 0) return;
-	
-	rect = ibuf->rect;
-	cmap = ibuf->cmap;
-
-	if (IS_ham(ibuf)){
-		
-		/* generate mask of max (8 + 2) bits */
-		mask = malloc(1024 * 2 * sizeof(int));
-
-		x = 1 << (ibuf->depth - 2);
-		y = 65535 / (x - 1);
-		
-		for (i = 0; i < x; i++){
-			mask[i] = 0;
-			mask[i + x]	= 0x00ffff;
-			mask[i + x + x] = 0xffff00;
-			mask[i + x + x + x] = 0xff00ff;
-
-			col = (y * i) >> 8;
-			
-			mask[i + 1024] = 0xff000000 | ibuf->cmap[i];
-			mask[i + x + 1024] = 0xff000000 | col << 16;
-			mask[i + x + x + 1024] = 0xff000000 | col;
-			mask[i + x + x + x + 1024] = 0xff000000 | col << 8;
-		}
-		
-		/* only color 0 transparant */
-		mask[0+1024] =ibuf->cmap[0];
-		
-		for (y = ibuf->y ; y>0 ; y--){
-			col = cmap[0];
-			for (x=ibuf->x ; x>0 ; x--){
-				code = *rect;
-				*rect++ = col = (col & mask[code]) | mask[code + 1024];
-			}
-		}
-		free(mask);
-	} else {
-	
-		for(i = ibuf->x * ibuf->y; i>0; i--){
-			col = *rect;
-			if (col >= 0 && col < ibuf->maxcol) *rect = cmap[col];
-			rect++;
-
-			/* *(rect++) = cmap[*rect]; */
-		}
-	}
-}
-