/* 8 bit decoding routines */ #include #include #include "decoders.h" static void dispatchDecoder(unsigned char **pFrame, unsigned char codeType, unsigned char **pData, int *pDataRemain, int *curXb, int *curYb); void decodeFrame8(unsigned char *pFrame, unsigned char *pMap, int mapRemain, unsigned char *pData, int dataRemain) { int i, j; int xb, yb; xb = g_width >> 3; yb = g_height >> 3; for (j=0; j= ((unsigned char *)g_vBackBuf1) + g_width*g_height) fprintf(stderr, "danger! pointing out of bounds above after dispatch decoder: %d, %d (1) [%x]\n", i, j, (*pMap) & 0xf); dispatchDecoder(&pFrame, (*pMap) >> 4, &pData, &dataRemain, &i, &j); if (pFrame < (unsigned char *)g_vBackBuf1) fprintf(stderr, "danger! pointing out of bounds below after dispatch decoder: %d, %d (2) [%x]\n", i, j, (*pMap) >> 4); else if (pFrame >= ((unsigned char *)g_vBackBuf1) + g_width*g_height) fprintf(stderr, "danger! pointing out of bounds above after dispatch decoder: %d, %d (2) [%x]\n", i, j, (*pMap) >> 4); ++pMap; --mapRemain; } pFrame += 7*g_width; } } static void relClose(int i, int *x, int *y) { int ma, mi; ma = i >> 4; mi = i & 0xf; *x = mi - 8; *y = ma - 8; } static void relFar(int i, int sign, int *x, int *y) { if (i < 56) { *x = sign * (8 + (i % 7)); *y = sign * (i / 7); } else { *x = sign * (-14 + (i - 56) % 29); *y = sign * (8 + (i - 56) / 29); } } /* copies an 8x8 block from pSrc to pDest. pDest and pSrc are both g_width bytes wide */ static void copyFrame(unsigned char *pDest, unsigned char *pSrc) { int i; for (i=0; i<8; i++) { memcpy(pDest, pSrc, 8); pDest += g_width; pSrc += g_width; } } // Fill in the next eight bytes with p[0], p[1], p[2], or p[3], // depending on the corresponding two-bit value in pat0 and pat1 static void patternRow4Pixels(unsigned char *pFrame, unsigned char pat0, unsigned char pat1, unsigned char *p) { unsigned short mask=0x0003; unsigned short shift=0; unsigned short pattern = (pat1 << 8) | pat0; while (mask != 0) { *pFrame++ = p[(mask & pattern) >> shift]; mask <<= 2; shift += 2; } } // Fill in the next four 2x2 pixel blocks with p[0], p[1], p[2], or p[3], // depending on the corresponding two-bit value in pat0. static void patternRow4Pixels2(unsigned char *pFrame, unsigned char pat0, unsigned char *p) { unsigned char mask=0x03; unsigned char shift=0; unsigned char pel; while (mask != 0) { pel = p[(mask & pat0) >> shift]; pFrame[0] = pel; pFrame[1] = pel; pFrame[g_width + 0] = pel; pFrame[g_width + 1] = pel; pFrame += 2; mask <<= 2; shift += 2; } } // Fill in the next four 2x1 pixel blocks with p[0], p[1], p[2], or p[3], // depending on the corresponding two-bit value in pat. static void patternRow4Pixels2x1(unsigned char *pFrame, unsigned char pat, unsigned char *p) { unsigned char mask=0x03; unsigned char shift=0; unsigned char pel; while (mask != 0) { pel = p[(mask & pat) >> shift]; pFrame[0] = pel; pFrame[1] = pel; pFrame += 2; mask <<= 2; shift += 2; } } // Fill in the next 4x4 pixel block with p[0], p[1], p[2], or p[3], // depending on the corresponding two-bit value in pat0, pat1, pat2, and pat3. static void patternQuadrant4Pixels(unsigned char *pFrame, unsigned char pat0, unsigned char pat1, unsigned char pat2, unsigned char pat3, unsigned char *p) { unsigned long mask = 0x00000003UL; int shift=0; int i; unsigned long pat = (pat3 << 24) | (pat2 << 16) | (pat1 << 8) | pat0; for (i=0; i<16; i++) { pFrame[i&3] = p[(pat & mask) >> shift]; if ((i&3) == 3) pFrame += g_width; mask <<= 2; shift += 2; } } // fills the next 8 pixels with either p[0] or p[1], depending on pattern static void patternRow2Pixels(unsigned char *pFrame, unsigned char pat, unsigned char *p) { unsigned char mask=0x01; while (mask != 0) { *pFrame++ = p[(mask & pat) ? 1 : 0]; mask <<= 1; } } // fills the next four 2 x 2 pixel boxes with either p[0] or p[1], depending on pattern static void patternRow2Pixels2(unsigned char *pFrame, unsigned char pat, unsigned char *p) { unsigned char pel; unsigned char mask=0x1; while (mask != 0x10) { pel = p[(mask & pat) ? 1 : 0]; pFrame[0] = pel; // upper-left pFrame[1] = pel; // upper-right pFrame[g_width + 0] = pel; // lower-left pFrame[g_width + 1] = pel; // lower-right pFrame += 2; mask <<= 1; } } // fills pixels in the next 4 x 4 pixel boxes with either p[0] or p[1], depending on pat0 and pat1. static void patternQuadrant2Pixels(unsigned char *pFrame, unsigned char pat0, unsigned char pat1, unsigned char *p) { unsigned char pel; unsigned short mask = 0x0001; int i, j; unsigned short pat = (pat1 << 8) | pat0; for (i=0; i<4; i++) { for (j=0; j<4; j++) { pel = p[(pat & mask) ? 1 : 0]; pFrame[j + i * g_width] = pel; mask <<= 1; } } } static void dispatchDecoder(unsigned char **pFrame, unsigned char codeType, unsigned char **pData, int *pDataRemain, int *curXb, int *curYb) { unsigned char p[4]; unsigned char pat[16]; int i, j, k; int x, y; /* Data is processed in 8x8 pixel blocks. There are 16 ways to encode each block. */ switch(codeType) { case 0x0: /* block is copied from block in current frame */ copyFrame(*pFrame, *pFrame + ((unsigned char *)g_vBackBuf2 - (unsigned char *)g_vBackBuf1)); case 0x1: /* block is unchanged from two frames ago */ *pFrame += 8; break; case 0x2: /* Block is copied from nearby (below and/or to the right) within the new frame. The offset within the buffer from which to grab the patch of 8 pixels is given by grabbing a byte B from the data stream, which is broken into a positive x and y offset according to the following mapping: if B < 56: x = 8 + (B % 7) y = B / 7 else x = -14 + ((B - 56) % 29) y = 8 + ((B - 56) / 29) */ relFar(*(*pData)++, 1, &x, &y); copyFrame(*pFrame, *pFrame + x + y*g_width); *pFrame += 8; --*pDataRemain; break; case 0x3: /* Block is copied from nearby (above and/or to the left) within the new frame. if B < 56: x = -(8 + (B % 7)) y = -(B / 7) else x = -(-14 + ((B - 56) % 29)) y = -( 8 + ((B - 56) / 29)) */ relFar(*(*pData)++, -1, &x, &y); copyFrame(*pFrame, *pFrame + x + y*g_width); *pFrame += 8; --*pDataRemain; break; case 0x4: /* Similar to 0x2 and 0x3, except this method copies from the "current" frame, rather than the "new" frame, and instead of the lopsided mapping they use, this one uses one which is symmetric and centered around the top-left corner of the block. This uses only 1 byte still, though, so the range is decreased, since we have to encode all directions in a single byte. The byte we pull from the data stream, I'll call B. Call the highest 4 bits of B BH and the lowest 4 bytes BL. Then the offset from which to copy the data is: x = -8 + BL y = -8 + BH */ relClose(*(*pData)++, &x, &y); copyFrame(*pFrame, *pFrame + ((unsigned char *)g_vBackBuf2 - (unsigned char *)g_vBackBuf1) + x + y*g_width); *pFrame += 8; --*pDataRemain; break; case 0x5: /* Similar to 0x4, but instead of one byte for the offset, this uses two bytes to encode a larger range, the first being the x offset as a signed 8-bit value, and the second being the y offset as a signed 8-bit value. */ x = (signed char)*(*pData)++; y = (signed char)*(*pData)++; copyFrame(*pFrame, *pFrame + ((unsigned char *)g_vBackBuf2 - (unsigned char *)g_vBackBuf1) + x + y*g_width); *pFrame += 8; *pDataRemain -= 2; break; case 0x6: /* I can't figure out how any file containing a block of this type could still be playable, since it appears that it would leave the internal bookkeeping in an inconsistent state in the BG player code. Ahh, well. Perhaps it was a bug in the BG player code that just didn't happen to be exposed by any of the included movies. Anyway, this skips the next two blocks, doing nothing to them. Note that if you've reached the end of a row, this means going on to the next row. */ for (i=0; i<2; i++) { *pFrame += 16; if (++*curXb == (g_width >> 3)) { *pFrame += 7*g_width; *curXb = 0; if (++*curYb == (g_height >> 3)) return; } } break; case 0x7: /* Ok, here's where it starts to get really...interesting. This is, incidentally, the part where they started using self-modifying code. So, most of the following encodings are "patterned" blocks, where we are given a number of pixel values and then bitmapped values to specify which pixel values belong to which squares. For this encoding, we are given the following in the data stream: P0 P1 These are pixel values (i.e. 8-bit indices into the palette). If P0 <= P1, we then get 8 more bytes from the data stream, one for each row in the block: B0 B1 B2 B3 B4 B5 B6 B7 For each row, the leftmost pixel is represented by the low-order bit, and the rightmost by the high-order bit. Use your imagination in between. If a bit is set, the pixel value is P1 and if it is unset, the pixel value is P0. So, for example, if we had: 11 22 fe 83 83 83 83 83 83 fe This would represent the following layout: 11 22 22 22 22 22 22 22 ; fe == 11111110 22 22 11 11 11 11 11 22 ; 83 == 10000011 22 22 11 11 11 11 11 22 ; 83 == 10000011 22 22 11 11 11 11 11 22 ; 83 == 10000011 22 22 11 11 11 11 11 22 ; 83 == 10000011 22 22 11 11 11 11 11 22 ; 83 == 10000011 22 22 11 11 11 11 11 22 ; 83 == 10000011 11 22 22 22 22 22 22 22 ; fe == 11111110 If, on the other hand, P0 > P1, we get two more bytes from the data stream: B0 B1 Each of these bytes contains two 4-bit patterns. These patterns work like the patterns above with 8 bytes, except each bit represents a 2x2 pixel region. B0 contains the pattern for the top two rows and B1 contains the pattern for the bottom two rows. Note that the low-order nibble of each byte contains the pattern for the upper of the two rows that that byte controls. So if we had: 22 11 7e 83 The output would be: 11 11 22 22 22 22 22 22 ; e == 1 1 1 0 11 11 22 22 22 22 22 22 ; 22 22 22 22 22 22 11 11 ; 7 == 0 1 1 1 22 22 22 22 22 22 11 11 ; 11 11 11 11 11 11 22 22 ; 3 == 1 0 0 0 11 11 11 11 11 11 22 22 ; 22 22 22 22 11 11 11 11 ; 8 == 0 0 1 1 22 22 22 22 11 11 11 11 ; */ p[0] = *(*pData)++; p[1] = *(*pData)++; if (p[0] <= p[1]) { for (i=0; i<8; i++) { patternRow2Pixels(*pFrame, *(*pData)++, p); *pFrame += g_width; } } else { for (i=0; i<2; i++) { patternRow2Pixels2(*pFrame, *(*pData) & 0xf, p); *pFrame += 2*g_width; patternRow2Pixels2(*pFrame, *(*pData)++ >> 4, p); *pFrame += 2*g_width; } } *pFrame -= (8*g_width - 8); break; case 0x8: /* Ok, this one is basically like encoding 0x7, only more complicated. Again, we start out by getting two bytes on the data stream: P0 P1 if P0 <= P1 then we get the following from the data stream: B0 B1 P2 P3 B2 B3 P4 P5 B4 B5 P6 P7 B6 B7 P0 P1 and B0 B1 are used for the top-left corner, P2 P3 B2 B3 for the bottom-left corner, P4 P5 B4 B5 for the top-right, P6 P7 B6 B7 for the bottom-right. (So, each codes for a 4x4 pixel array.) Since we have 16 bits in B0 B1, there is one bit for each pixel in the array. The convention for the bit-mapping is, again, left to right and top to bottom. So, basically, the top-left quarter of the block is an arbitrary pattern with 2 pixels, the bottom-left a different arbitrary pattern with 2 different pixels, and so on. For example if the next 16 bytes were: 00 22 f9 9f 44 55 aa 55 11 33 cc 33 66 77 01 ef We'd draw: 22 22 22 22 | 11 11 33 33 ; f = 1111, c = 1100 22 00 00 22 | 11 11 33 33 ; 9 = 1001, c = 1100 22 00 00 22 | 33 33 11 11 ; 9 = 1001, 3 = 0011 22 22 22 22 | 33 33 11 11 ; f = 1111, 3 = 0011 ------------+------------ 44 55 44 55 | 66 66 66 66 ; a = 1010, 0 = 0000 44 55 44 55 | 77 66 66 66 ; a = 1010, 1 = 0001 55 44 55 44 | 66 77 77 77 ; 5 = 0101, e = 1110 55 44 55 44 | 77 77 77 77 ; 5 = 0101, f = 1111 I've added a dividing line in the above to clearly delineate the quadrants. Now, if P0 > P1 then we get 10 more bytes from the data stream: B0 B1 B2 B3 P2 P3 B4 B5 B6 B7 Now, if P2 <= P3, then the first six bytes [P0 P1 B0 B1 B2 B3] represent the left half of the block and the latter six bytes [P2 P3 B4 B5 B6 B7] represent the right half. For example: 22 00 01 37 f7 31 11 66 8c e6 73 31 yeilds: 22 22 22 22 | 11 11 11 66 ; 0: 0000 | 8: 1000 00 22 22 22 | 11 11 66 66 ; 1: 0001 | C: 1100 00 00 22 22 | 11 66 66 66 ; 3: 0011 | e: 1110 00 00 00 22 | 11 66 11 66 ; 7: 0111 | 6: 0101 00 00 00 00 | 66 66 66 11 ; f: 1111 | 7: 0111 00 00 00 22 | 66 66 11 11 ; 7: 0111 | 3: 0011 00 00 22 22 | 66 66 11 11 ; 3: 0011 | 3: 0011 00 22 22 22 | 66 11 11 11 ; 1: 0001 | 1: 0001 On the other hand, if P0 > P1 and P2 > P3, then [P0 P1 B0 B1 B2 B3] represent the top half of the block and [P2 P3 B4 B5 B6 B7] represent the bottom half. For example: 22 00 cc 66 33 19 66 11 18 24 42 81 yeilds: 22 22 00 00 22 22 00 00 ; cc: 11001100 22 00 00 22 22 00 00 22 ; 66: 01100110 00 00 22 22 00 00 22 22 ; 33: 00110011 00 22 22 00 00 22 22 22 ; 19: 00011001 ----------------------- 66 66 66 11 11 66 66 66 ; 18: 00011000 66 66 11 66 66 11 66 66 ; 24: 00100100 66 11 66 66 66 66 11 66 ; 42: 01000010 11 66 66 66 66 66 66 11 ; 81: 10000001 */ if ( (*pData)[0] <= (*pData)[1]) { // four quadrant case for (i=0; i<4; i++) { p[0] = *(*pData)++; p[1] = *(*pData)++; pat[0] = *(*pData)++; pat[1] = *(*pData)++; patternQuadrant2Pixels(*pFrame, pat[0], pat[1], p); // alternate between moving down and moving up and right if (i & 1) *pFrame += 4 - 4*g_width; // up and right else *pFrame += 4*g_width; // down } } else if ( (*pData)[6] <= (*pData)[7]) { // split horizontal for (i=0; i<4; i++) { if ((i & 1) == 0) { p[0] = *(*pData)++; p[1] = *(*pData)++; } pat[0] = *(*pData)++; pat[1] = *(*pData)++; patternQuadrant2Pixels(*pFrame, pat[0], pat[1], p); if (i & 1) *pFrame -= (4*g_width - 4); else *pFrame += 4*g_width; } } else { // split vertical for (i=0; i<8; i++) { if ((i & 3) == 0) { p[0] = *(*pData)++; p[1] = *(*pData)++; } patternRow2Pixels(*pFrame, *(*pData)++, p); *pFrame += g_width; } *pFrame -= (8*g_width - 8); } break; case 0x9: /* Similar to the previous 2 encodings, only more complicated. And it will get worse before it gets better. No longer are we dealing with patterns over two pixel values. Now we are dealing with patterns over 4 pixel values with 2 bits assigned to each pixel (or block of pixels). So, first on the data stream are our 4 pixel values: P0 P1 P2 P3 Now, if P0 <= P1 AND P2 <= P3, we get 16 bytes of pattern, each 2 bits representing a 1x1 pixel (00=P0, 01=P1, 10=P2, 11=P3). The ordering is again left to right and top to bottom. The most significant bits represent the left side at the top, and so on. If P0 <= P1 AND P2 > P3, we get 4 bytes of pattern, each 2 bits representing a 2x2 pixel. Ordering is left to right and top to bottom. if P0 > P1 AND P2 <= P3, we get 8 bytes of pattern, each 2 bits representing a 2x1 pixel (i.e. 2 pixels wide, and 1 high). if P0 > P1 AND P2 > P3, we get 8 bytes of pattern, each 2 bits representing a 1x2 pixel (i.e. 1 pixel wide, and 2 high). */ if ( (*pData)[0] <= (*pData)[1]) { if ( (*pData)[2] <= (*pData)[3]) { p[0] = *(*pData)++; p[1] = *(*pData)++; p[2] = *(*pData)++; p[3] = *(*pData)++; for (i=0; i<8; i++) { pat[0] = *(*pData)++; pat[1] = *(*pData)++; patternRow4Pixels(*pFrame, pat[0], pat[1], p); *pFrame += g_width; } *pFrame -= (8*g_width - 8); } else { p[0] = *(*pData)++; p[1] = *(*pData)++; p[2] = *(*pData)++; p[3] = *(*pData)++; patternRow4Pixels2(*pFrame, *(*pData)++, p); *pFrame += 2*g_width; patternRow4Pixels2(*pFrame, *(*pData)++, p); *pFrame += 2*g_width; patternRow4Pixels2(*pFrame, *(*pData)++, p); *pFrame += 2*g_width; patternRow4Pixels2(*pFrame, *(*pData)++, p); *pFrame -= (6*g_width - 8); } } else { if ( (*pData)[2] <= (*pData)[3]) { // draw 2x1 strips p[0] = *(*pData)++; p[1] = *(*pData)++; p[2] = *(*pData)++; p[3] = *(*pData)++; for (i=0; i<8; i++) { pat[0] = *(*pData)++; patternRow4Pixels2x1(*pFrame, pat[0], p); *pFrame += g_width; } *pFrame -= (8*g_width - 8); } else { // draw 1x2 strips p[0] = *(*pData)++; p[1] = *(*pData)++; p[2] = *(*pData)++; p[3] = *(*pData)++; for (i=0; i<4; i++) { pat[0] = *(*pData)++; pat[1] = *(*pData)++; patternRow4Pixels(*pFrame, pat[0], pat[1], p); *pFrame += g_width; patternRow4Pixels(*pFrame, pat[0], pat[1], p); *pFrame += g_width; } *pFrame -= (8*g_width - 8); } } break; case 0xa: /* Similar to the previous, only a little more complicated. We are still dealing with patterns over 4 pixel values with 2 bits assigned to each pixel (or block of pixels). So, first on the data stream are our 4 pixel values: P0 P1 P2 P3 Now, if P0 <= P1, the block is divided into 4 quadrants, ordered (as with opcode 0x8) TL, BL, TR, BR. In this case the next data in the data stream should be: B0 B1 B2 B3 P4 P5 P6 P7 B4 B5 B6 B7 P8 P9 P10 P11 B8 B9 B10 B11 P12 P13 P14 P15 B12 B13 B14 B15 Each 2 bits represent a 1x1 pixel (00=P0, 01=P1, 10=P2, 11=P3). The ordering is again left to right and top to bottom. The most significant bits represent the right side at the top, and so on. If P0 > P1 then the next data on the data stream is: B0 B1 B2 B3 B4 B5 B6 B7 P4 P5 P6 P7 B8 B9 B10 B11 B12 B13 B14 B15 Now, in this case, if P4 <= P5, [P0 P1 P2 P3 B0 B1 B2 B3 B4 B5 B6 B7] represent the left half of the block and the other bytes represent the right half. If P4 > P5, then [P0 P1 P2 P3 B0 B1 B2 B3 B4 B5 B6 B7] represent the top half of the block and the other bytes represent the bottom half. */ if ( (*pData)[0] <= (*pData)[1]) { for (i=0; i<4; i++) { p[0] = *(*pData)++; p[1] = *(*pData)++; p[2] = *(*pData)++; p[3] = *(*pData)++; pat[0] = *(*pData)++; pat[1] = *(*pData)++; pat[2] = *(*pData)++; pat[3] = *(*pData)++; patternQuadrant4Pixels(*pFrame, pat[0], pat[1], pat[2], pat[3], p); if (i & 1) *pFrame -= (4*g_width - 4); else *pFrame += 4*g_width; } } else { if ( (*pData)[12] <= (*pData)[13]) { // split vertical for (i=0; i<4; i++) { if ((i&1) == 0) { p[0] = *(*pData)++; p[1] = *(*pData)++; p[2] = *(*pData)++; p[3] = *(*pData)++; } pat[0] = *(*pData)++; pat[1] = *(*pData)++; pat[2] = *(*pData)++; pat[3] = *(*pData)++; patternQuadrant4Pixels(*pFrame, pat[0], pat[1], pat[2], pat[3], p); if (i & 1) *pFrame -= (4*g_width - 4); else *pFrame += 4*g_width; } } else { // split horizontal for (i=0; i<8; i++) { if ((i&3) == 0) { p[0] = *(*pData)++; p[1] = *(*pData)++; p[2] = *(*pData)++; p[3] = *(*pData)++; } pat[0] = *(*pData)++; pat[1] = *(*pData)++; patternRow4Pixels(*pFrame, pat[0], pat[1], p); *pFrame += g_width; } *pFrame -= (8*g_width - 8); } } break; case 0xb: /* In this encoding we get raw pixel data in the data stream -- 64 bytes of pixel data. 1 byte for each pixel, and in the standard order (l->r, t->b). */ for (i=0; i<8; i++) { memcpy(*pFrame, *pData, 8); *pFrame += g_width; *pData += 8; *pDataRemain -= 8; } *pFrame -= (8*g_width - 8); break; case 0xc: /* In this encoding we get raw pixel data in the data stream -- 16 bytes of pixel data. 1 byte for each block of 2x2 pixels, and in the standard order (l->r, t->b). */ for (i=0; i<4; i++) { for (j=0; j<2; j++) { for (k=0; k<4; k++) { (*pFrame)[2*k] = (*pData)[k]; (*pFrame)[2*k+1] = (*pData)[k]; } *pFrame += g_width; } *pData += 4; *pDataRemain -= 4; } *pFrame -= (8*g_width - 8); break; case 0xd: /* In this encoding we get raw pixel data in the data stream -- 4 bytes of pixel data. 1 byte for each block of 4x4 pixels, and in the standard order (l->r, t->b). */ for (i=0; i<2; i++) { for (j=0; j<4; j++) { for (k=0; k<4; k++) { (*pFrame)[k*g_width+j] = (*pData)[0]; (*pFrame)[k*g_width+j+4] = (*pData)[1]; } } *pFrame += 4*g_width; *pData += 2; *pDataRemain -= 2; } *pFrame -= (8*g_width - 8); break; case 0xe: /* This encoding represents a solid 8x8 frame. We get 1 byte of pixel data from the data stream. */ for (i=0; i<8; i++) { memset(*pFrame, **pData, 8); *pFrame += g_width; } ++*pData; --*pDataRemain; *pFrame -= (8*g_width - 8); break; case 0xf: /* This encoding represents a "dithered" frame, which is checkerboarded with alternate pixels of two colors. We get 2 bytes of pixel data from the data stream, and these bytes are alternated: P0 P1 P0 P1 P0 P1 P0 P1 P1 P0 P1 P0 P1 P0 P1 P0 ... P0 P1 P0 P1 P0 P1 P0 P1 P1 P0 P1 P0 P1 P0 P1 P0 */ for (i=0; i<8; i++) { for (j=0; j<8; j++) { (*pFrame)[j] = (*pData)[(i+j)&1]; } *pFrame += g_width; } *pData += 2; *pDataRemain -= 2; *pFrame -= (8*g_width - 8); break; default: break; } }