mirrors/gstreamer
mirrors/gstreamer
1
1
Fork 0
mirror of https://gitlab.freedesktop.org/gstreamer/gstreamer.git synced 2024-11-27 12:11:13 +00:00
Code Issues 1 Projects Releases Wiki Activity

assrender: refactor blitting, avoid writing past end of buffer

Browse source 
Previous blitting code could potentially write past the
end of the buffer if the x or y position was odd, and for
the same underlying reason, didn't get the chroma registration
correct in the odd position case.

https://bugzilla.gnome.org/show_bug.cgi?id=647830
This commit is contained in:
David Schleef 2011-04-14 20:46:52 -07:00 committed by Tim-Philipp Müller
parent b18bf6a018
commit 51e6acad1f
1 changed files with 95 additions and 66 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

161
ext/assrender/gstassrender.c
View file

@ -618,101 +618,130 @@ blit_i420 (GstAssRender * render, ASS_Image * ass_image, GstBuffer * buffer)
buffer->data + y_offset + ass_image->dst_y * y_stride +
ass_image->dst_x;
dst_u =
buffer->data + u_offset + ((ass_image->dst_y + 1) / 2) * u_stride +
(ass_image->dst_x + 1) / 2;
buffer->data + u_offset + (ass_image->dst_y / 2) * u_stride +
ass_image->dst_x / 2;
dst_v =
buffer->data + v_offset + ((ass_image->dst_y + 1) / 2) * v_stride +
(ass_image->dst_x + 1) / 2;
buffer->data + v_offset + (ass_image->dst_y / 2) * v_stride +
ass_image->dst_x / 2;
for (y = 0; y < h - 1; y += 2) {
for (x = 0; x < w - 1; x += 2) {
k = src[0] * alpha / 255;
k2 = k;
dst_y[0] = (k * Y + (255 - k) * dst_y[0]) / 255;
for (y = 0; y < h; y++) {
dst_y = buffer->data + y_offset + (ass_image->dst_y + y) * y_stride +
ass_image->dst_x;
for (x = 0; x < w; x++) {
k = src[y * ass_image->w + x] * alpha / 255;
dst_y[x] = (k * Y + (255 - k) * dst_y[x]) / 255;
}
}
k = src[1] * alpha / 255;
k2 += k;
dst_y[1] = (k * Y + (255 - k) * dst_y[1]) / 255;
src += src_stride;
dst_y += y_stride;
k = src[0] * alpha / 255;
k2 += k;
dst_y[0] = (k * Y + (255 - k) * dst_y[0]) / 255;
k = src[1] * alpha / 255;
k2 += k;
dst_y[1] = (k * Y + (255 - k) * dst_y[1]) / 255;
k2 /= 4;
y = 0;
if (ass_image->dst_y & 1) {
dst_u =
buffer->data + u_offset + (ass_image->dst_y / 2) * u_stride +
ass_image->dst_x / 2;
dst_v =
buffer->data + v_offset + (ass_image->dst_y / 2) * v_stride +
ass_image->dst_x / 2;
x = 0;
if (ass_image->dst_x & 1) {
k2 = src[y * ass_image->w + x] * alpha / 255;
k2 = (k2 + 2) >> 2;
dst_u[0] = (k2 * U + (255 - k2) * dst_u[0]) / 255;
dst_v[0] = (k2 * V + (255 - k2) * dst_v[0]) / 255;
x++;
dst_u++;
dst_v++;
}
for (; x < w - 1; x += 2) {
k2 = src[y * ass_image->w + x] * alpha / 255;
k2 += src[y * ass_image->w + x + 1] * alpha / 255;
k2 = (k2 + 2) >> 2;
dst_u[0] = (k2 * U + (255 - k2) * dst_u[0]) / 255;
dst_v[0] = (k2 * V + (255 - k2) * dst_v[0]) / 255;
dst_u++;
dst_v++;
src += -src_stride + 2;
dst_y += -y_stride + 2;
}
if (x < w) {
k = src[0] * alpha / 255;
k2 = k;
dst_y[0] = (k * Y + (255 - k) * dst_y[0]) / 255;
k2 = src[y * ass_image->w + x] * alpha / 255;
k2 = (k2 + 2) >> 2;
dst_u[0] = (k2 * U + (255 - k2) * dst_u[0]) / 255;
dst_v[0] = (k2 * V + (255 - k2) * dst_v[0]) / 255;
}
}
src += src_stride;
dst_y += y_stride;
k = src[0] * alpha / 255;
k2 += k;
dst_y[0] = (k * Y + (255 - k) * dst_y[0]) / 255;
k2 /= 2;
for (; y < h - 1; y += 2) {
dst_u =
buffer->data + u_offset + ((ass_image->dst_y + y) / 2) * u_stride +
ass_image->dst_x / 2;
dst_v =
buffer->data + v_offset + ((ass_image->dst_y + y) / 2) * v_stride +
ass_image->dst_x / 2;
x = 0;
if (ass_image->dst_x & 1) {
k2 = src[y * ass_image->w + x] * alpha / 255;
k2 += src[(y + 1) * ass_image->w + x] * alpha / 255;
k2 = (k2 + 2) >> 2;
dst_u[0] = (k2 * U + (255 - k2) * dst_u[0]) / 255;
dst_v[0] = (k2 * V + (255 - k2) * dst_v[0]) / 255;
x++;
dst_u++;
dst_v++;
}
for (; x < w - 1; x += 2) {
k2 = src[y * ass_image->w + x] * alpha / 255;
k2 += src[y * ass_image->w + x + 1] * alpha / 255;
k2 += src[(y + 1) * ass_image->w + x] * alpha / 255;
k2 += src[(y + 1) * ass_image->w + x + 1] * alpha / 255;
k2 = (k2 + 2) >> 2;
dst_u[0] = (k2 * U + (255 - k2) * dst_u[0]) / 255;
dst_v[0] = (k2 * V + (255 - k2) * dst_v[0]) / 255;
dst_u++;
dst_v++;
src += -src_stride + 1;
dst_y += -y_stride + 1;
}
src += src_stride + (src_stride - w);
dst_y += y_stride + (y_stride - w);
dst_u += u_stride - w2;
dst_v += v_stride - w2;
if (x < w) {
k2 = src[y * ass_image->w + x] * alpha / 255;
k2 += src[(y + 1) * ass_image->w + x] * alpha / 255;
k2 = (k2 + 2) >> 2;
dst_u[0] = (k2 * U + (255 - k2) * dst_u[0]) / 255;
dst_v[0] = (k2 * V + (255 - k2) * dst_v[0]) / 255;
}
}
if (y < h) {
for (x = 0; x < w - 1; x += 2) {
k = src[0] * alpha / 255;
k2 = k;
dst_y[0] = (k * Y + (255 - k) * dst_y[0]) / 255;
k = src[1] * alpha / 255;
k2 += k;
dst_y[1] = (k * Y + (255 - k) * dst_y[1]) / 255;
k2 /= 2;
dst_u =
buffer->data + u_offset + (ass_image->dst_y / 2) * u_stride +
ass_image->dst_x / 2;
dst_v =
buffer->data + v_offset + (ass_image->dst_y / 2) * v_stride +
ass_image->dst_x / 2;
x = 0;
if (ass_image->dst_x & 1) {
k2 = src[y * ass_image->w + x] * alpha / 255;
k2 = (k2 + 2) >> 2;
dst_u[0] = (k2 * U + (255 - k2) * dst_u[0]) / 255;
dst_v[0] = (k2 * V + (255 - k2) * dst_v[0]) / 255;
x++;
dst_u++;
dst_v++;
}
for (; x < w - 1; x += 2) {
k2 = src[y * ass_image->w + x] * alpha / 255;
k2 += src[y * ass_image->w + x + 1] * alpha / 255;
k2 = (k2 + 2) >> 2;
dst_u[0] = (k2 * U + (255 - k2) * dst_u[0]) / 255;
dst_v[0] = (k2 * V + (255 - k2) * dst_v[0]) / 255;
dst_u++;
dst_v++;
src += 2;
dst_y += 2;
}
if (x < w) {
k = src[0] * alpha / 255;
k2 = k;
dst_y[0] = (k * Y + (255 - k) * dst_y[0]) / 255;
k2 = src[y * ass_image->w + x] * alpha / 255;
k2 = (k2 + 2) >> 2;
dst_u[0] = (k2 * U + (255 - k2) * dst_u[0]) / 255;
dst_v[0] = (k2 * V + (255 - k2) * dst_v[0]) / 255;
}
}
next:
counter++;
ass_image = ass_image->next;