forked from enlightenment/efl
parent
6282cc0a6d
commit
8cca4234c4
|
@ -242,6 +242,7 @@ scale_rgba_in_to_out_clip_sample_internal(RGBA_Image *src, RGBA_Image *dst,
|
||||||
func = evas_common_gfx_func_composite_pixel_mask_span_get(src, dst, dst_clip_w, dc->render_op);
|
func = evas_common_gfx_func_composite_pixel_mask_span_get(src, dst, dst_clip_w, dc->render_op);
|
||||||
maskobj = dc->mask.mask;
|
maskobj = dc->mask.mask;
|
||||||
mask = maskobj->mask.mask;
|
mask = maskobj->mask.mask;
|
||||||
|
/*
|
||||||
if (1 || dst_region_w > src_region_w || dst_region_h > src_region_h){
|
if (1 || dst_region_w > src_region_w || dst_region_h > src_region_h){
|
||||||
printf("Mask w/h: %d/%d\n",maskobj->cache_entry.w,
|
printf("Mask w/h: %d/%d\n",maskobj->cache_entry.w,
|
||||||
maskobj->cache_entry.h);
|
maskobj->cache_entry.h);
|
||||||
|
@ -249,6 +250,7 @@ scale_rgba_in_to_out_clip_sample_internal(RGBA_Image *src, RGBA_Image *dst,
|
||||||
dst_region_w,src_region_w,
|
dst_region_w,src_region_w,
|
||||||
dst_region_h,src_region_h);
|
dst_region_h,src_region_h);
|
||||||
}
|
}
|
||||||
|
*/
|
||||||
}
|
}
|
||||||
else if (dc->mul.use)
|
else if (dc->mul.use)
|
||||||
func = evas_common_gfx_func_composite_pixel_color_span_get(src, dc->mul.col, dst, dst_clip_w, dc->render_op);
|
func = evas_common_gfx_func_composite_pixel_color_span_get(src, dc->mul.col, dst, dst_clip_w, dc->render_op);
|
||||||
|
@ -286,6 +288,13 @@ scale_rgba_in_to_out_clip_sample_internal(RGBA_Image *src, RGBA_Image *dst,
|
||||||
/* * blend here [clip_w *] ptr -> dst_ptr * */
|
/* * blend here [clip_w *] ptr -> dst_ptr * */
|
||||||
if (mask)
|
if (mask)
|
||||||
{
|
{
|
||||||
|
// nash: problem here. normally dst_clip_x == dc->mask.x
|
||||||
|
// but then... at some point they cease to be equal
|
||||||
|
// and thus you add a negative value to mask here
|
||||||
|
// in fact... u simply don't handle the mask being
|
||||||
|
// disjoint from the object. now maybe the test in
|
||||||
|
// expedite has a bug where it moves the mask img
|
||||||
|
// wrongly - but... i can see this code is fragile
|
||||||
mask += dst_clip_x - dc->mask.x;
|
mask += dst_clip_x - dc->mask.x;
|
||||||
mask += (dst_clip_y - dc->mask.y) * maskobj->cache_entry.w;
|
mask += (dst_clip_y - dc->mask.y) * maskobj->cache_entry.w;
|
||||||
}
|
}
|
||||||
|
|
Loading…
Reference in New Issue