sys/xvimage/xvimagesink.c: Improve the errors produced on bad output,...
sys/xvimage/xvimagesink.c: Improve the errors produced on bad output, including some human readable description strings. Original commit message from CVS: * sys/xvimage/xvimagesink.c: (gst_xvimagesink_xvimage_new), (gst_xvimagesink_get_xv_support), (gst_xvimagesink_xcontext_get), (gst_xvimagesink_get_format_from_caps), (gst_xvimagesink_setcaps), (gst_xvimagesink_show_frame), (gst_xvimagesink_buffer_alloc): Improve the errors produced on bad output, including some human readable description strings. Handle RGB Xv formats properly by transforming them into our big-endian caps description. Use gst_caps_truncate to ensure that we never try and choose a non-fixed caps in buffer_alloc. Handle the case where the XServer has a different idea about the size required for a particular frame and gives us too small a memory allocation. Use -1 to indicate 'no image format', because 0 is a valid XServer image format number. Put RGB Xv formats at the end of the caps, so that we always prefer YUV format frames. Iterate the available Xv Encodings to determine the maximum width and height, and then return that in our caps.