Commit 09584edc authored by Jan Schmidt's avatar Jan Schmidt

glimagesink: Correct PAR in output caps when transforming

When doing a 3D/multiview transformation and rescaling to
match the output window size, the resulting PAR may
not match the input any more and needs recalculating,
or else the GstSample reported to client-draw has the
wrong PAR.
parent 977a2d04
......@@ -1418,13 +1418,35 @@ update_output_format (GstGLImageSink * glimage_sink)
ret = configure_display_from_info (glimage_sink, out_info);
if (glimage_sink->convert_views) {
gint new_w, new_h;
gint par_n, par_d;
/* Match actual output window size for pixel-aligned output,
* even though we can't necessarily match the starting left/right
* view parity properly */
glimage_sink->out_info.width = MAX (1, glimage_sink->display_rect.w);
glimage_sink->out_info.height = MAX (1, glimage_sink->display_rect.h);
GST_LOG_OBJECT (glimage_sink, "Set 3D output scale to %d,%d",
glimage_sink->display_rect.w, glimage_sink->display_rect.h);
* view parity properly for line-by-line modes, because that
* depends on the window being placed correctly.
* FIXME: Should this rescaling be configurable? */
new_w = MAX (1, glimage_sink->display_rect.w);
new_h = MAX (1, glimage_sink->display_rect.h);
if (new_w != out_info->width || new_h != out_info->height) {
/* Recalculate PAR if rescaling */
gint from_dar_n, from_dar_d;
if (!gst_util_fraction_multiply (out_info->width, out_info->height,
out_info->par_n, out_info->par_d, &from_dar_n,
&from_dar_d) ||
!gst_util_fraction_multiply (from_dar_n, from_dar_d, new_h, new_w,
&par_n, &par_d)) {
par_n = glimage_sink->par_n;
par_d = glimage_sink->par_d;
out_info->par_n = par_n;
out_info->par_d = par_d;
out_info->width = new_w;
out_info->height = new_h;
GST_LOG_OBJECT (glimage_sink, "Set 3D output scale to %d,%d PAR %d/%d",
out_info->width, out_info->height, out_info->par_n, out_info->par_d);
s = gst_caps_get_structure (glimage_sink->in_caps, 0);
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment