guijan
/
legacy-imlib2
forked from old/legacy-imlib2
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

x11_grab.c: Correct 16 and 15 bit depth grabbing 

- Handling of odd sizes and offsets was broken
  Might be a bit slower than before but should now be correct.

- Color mapping changed
  Before we had
  16 bit: RRRRRGGGGGGBBBBB -> aaaaaaaaRRRRR000GGGGGG00BBBBB000
  15 bit: xRRRRRGGGGGBBBBB -> aaaaaaaaRRRRR000GGGGG000BBBBB000
  Now we do
  16 bit: RRRRRGGGGGGBBBBB -> aaaaaaaaRRRRRRRRGGGGGGGGBBBBBBBB
  15 bit: xRRRRRGGGGGBBBBB -> aaaaaaaaRRRRRRRRGGGGGGGGBBBBBBBB
  Probably somewhat slower than before because this mapping requires
  multiply/divide instead of just shifts, but is more correct, IMO.
This commit is contained in:
Kim Woelders 2022-02-28 17:54:11 +01:00
parent b0f3bfc907
commit d545dea9c6
1 changed files with 48 additions and 104 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

152
src/lib/x11_grab.c
View File

@ -29,8 +29,10 @@ __imlib_GrabXImageToRGBA(DATA32 * data,
{
int x, y, inx, iny;
const DATA32 *src;
const DATA16 *s16;
DATA32 *ptr;
int pixel;
DATA16 p16;
int bgr = 0;
if (!data)
@ -160,73 +162,44 @@ __imlib_GrabXImageToRGBA(DATA32 * data,
#undef RMSK
#undef GMSK
#undef BMSK
#undef R1SH
#undef G1SH
#undef B1SH
#undef R2SH
#undef G2SH
#undef B2SH
#undef P1
#undef P2
#define MP(x, y) ((XGetPixel(mxim, (x), (y))) ? 0xff000000 : 0)
#define RMSK 0xf80000
#define GMSK 0x00fc00
#define BMSK 0x0000f8
#define R1SH(p) ((p) << 8)
#define G1SH(p) ((p) << 5)
#define B1SH(p) ((p) << 3)
#define R2SH(p) ((p) >> 8)
#define G2SH(p) ((p) >> 11)
#define B2SH(p) ((p) >> 13)
#define P1(p) (R1SH(p) & RMSK) | (G1SH(p) & GMSK) | (B1SH(p) & BMSK)
#define P2(p) (R2SH(p) & RMSK) | (G2SH(p) & GMSK) | (B2SH(p) & BMSK)
#undef RSH
#undef GSH
#undef BSH
#define MP(x, y) ((XGetPixel(mxim, (x), (y))) ? 0xff: 0)
#define RMSK 0x1f
#define GMSK 0x3f
#define BMSK 0x1f
#define RSH(p) ((p) >> 11)
#define GSH(p) ((p) >> 5)
#define BSH(p) ((p) >> 0)
#define RVAL(p) (255 * (RSH(p) & RMSK)) / RMSK
#define GVAL(p) (255 * (GSH(p) & GMSK)) / GMSK
#define BVAL(p) (255 * (BSH(p) & BMSK)) / BMSK
if (mxim)
{
for (y = 0; y < h_src; y++)
{
src = (DATA32 *) (xim->data + (xim->bytes_per_line * y));
s16 = (DATA16 *) (xim->data + (xim->bytes_per_line * y));
ptr = data + ((y + iny) * w_dst) + inx;
for (x = 0; x < (w_src - 1); x += 2)
for (x = 0; x < w_src; x++)
{
#ifdef WORDS_BIGENDIAN
*ptr++ = MP(x + 1, y) | P2(*src);
*ptr++ = MP(x, y) | P1(*src);
#else
*ptr++ = MP(x, y) | P1(*src);
*ptr++ = MP(x + 1, y) | P2(*src);
#endif
src++;
}
if (x == (w_src - 1))
{
pixel = XGetPixel(xim, x, y);
*ptr++ = MP(x, y) | P1(pixel);
p16 = *s16++;
*ptr++ =
PIXEL_ARGB(MP(x, y), RVAL(p16), GVAL(p16), BVAL(p16));
}
}
}
#undef MP
#define MP(x, y) (0xff000000)
else
{
for (y = 0; y < h_src; y++)
{
src = (DATA32 *) (xim->data + (xim->bytes_per_line * y));
s16 = (DATA16 *) (xim->data + (xim->bytes_per_line * y));
ptr = data + ((y + iny) * w_dst) + inx;
for (x = 0; x < (w_src - 1); x += 2)
for (x = 0; x < w_src; x++)
{
#ifdef WORDS_BIGENDIAN
*ptr++ = MP(x + 1, y) | P2(*src);
*ptr++ = MP(x, y) | P1(*src);
#else
*ptr++ = MP(x, y) | P1(*src);
*ptr++ = MP(x + 1, y) | P2(*src);
#endif
src++;
}
if (x == (w_src - 1))
{
pixel = XGetPixel(xim, x, y);
*ptr++ = MP(x, y) | P1(pixel);
p16 = *s16++;
*ptr++ =
PIXEL_ARGB(0xff, RVAL(p16), GVAL(p16), BVAL(p16));
}
}
}
@ -236,73 +209,44 @@ __imlib_GrabXImageToRGBA(DATA32 * data,
#undef RMSK
#undef GMSK
#undef BMSK
#undef R1SH
#undef G1SH
#undef B1SH
#undef R2SH
#undef G2SH
#undef B2SH
#undef P1
#undef P2
#define MP(x, y) ((XGetPixel(mxim, (x), (y))) ? 0xff000000 : 0)
#define RMSK 0xf80000
#define GMSK 0x00f800
#define BMSK 0x0000f8
#define R1SH(p) ((p) << 9)
#define G1SH(p) ((p) << 6)
#define B1SH(p) ((p) << 3)
#define R2SH(p) ((p) >> 7)
#define G2SH(p) ((p) >> 10)
#define B2SH(p) ((p) >> 13)
#define P1(p) (R1SH(p) & RMSK) | (G1SH(p) & GMSK) | (B1SH(p) & BMSK)
#define P2(p) (R2SH(p) & RMSK) | (G2SH(p) & GMSK) | (B2SH(p) & BMSK)
#undef RSH
#undef GSH
#undef BSH
#define MP(x, y) ((XGetPixel(mxim, (x), (y))) ? 0xff: 0)
#define RMSK 0x1f
#define GMSK 0x1f
#define BMSK 0x1f
#define RSH(p) ((p) >> 10)
#define GSH(p) ((p) >> 5)
#define BSH(p) ((p) >> 0)
#define RVAL(p) (255 * (RSH(p) & RMSK)) / RMSK
#define GVAL(p) (255 * (GSH(p) & GMSK)) / GMSK
#define BVAL(p) (255 * (BSH(p) & BMSK)) / BMSK
if (mxim)
{
for (y = 0; y < h_src; y++)
{
src = (DATA32 *) (xim->data + (xim->bytes_per_line * y));
s16 = (DATA16 *) (xim->data + (xim->bytes_per_line * y));
ptr = data + ((y + iny) * w_dst) + inx;
for (x = 0; x < (w_src - 1); x += 2)
for (x = 0; x < w_src; x++)
{
#ifdef WORDS_BIGENDIAN
*ptr++ = MP(x + 1, y) | P2(*src);
*ptr++ = MP(x, y) | P1(*src);
#else
*ptr++ = MP(x, y) | P1(*src);
*ptr++ = MP(x + 1, y) | P2(*src);
#endif
src++;
}
if (x == (w_src - 1))
{
pixel = XGetPixel(xim, x, y);
*ptr++ = MP(x, y) | P1(pixel);
p16 = *s16++;
*ptr++ =
PIXEL_ARGB(MP(x, y), RVAL(p16), GVAL(p16), BVAL(p16));
}
}
}
#undef MP
#define MP(x, y) (0xff000000)
else
{
for (y = 0; y < h_src; y++)
{
src = (DATA32 *) (xim->data + (xim->bytes_per_line * y));
s16 = (DATA16 *) (xim->data + (xim->bytes_per_line * y));
ptr = data + ((y + iny) * w_dst) + inx;
for (x = 0; x < (w_src - 1); x += 2)
for (x = 0; x < w_src; x++)
{
#ifdef WORDS_BIGENDIAN
*ptr++ = MP(x + 1, y) | P2(*src);
*ptr++ = MP(x, y) | P1(*src);
#else
*ptr++ = MP(x, y) | P1(*src);
*ptr++ = MP(x + 1, y) | P2(*src);
#endif
src++;
}
if (x == (w_src - 1))
{
pixel = XGetPixel(xim, x, y);
*ptr++ = MP(x, y) | P1(pixel);
p16 = *s16++;
*ptr++ =
PIXEL_ARGB(0xff, RVAL(p16), GVAL(p16), BVAL(p16));
}
}
}