On Mon, Sep 29, 2003 at 04:07:17PM +0300, Tuukka Toivonen wrote: > On Mon, 29 Sep 2003, Ladislav Michl wrote: > > >Documentation/video4linux/API.html states that struct video_capture should > >be used to set capture window. This is explained in greater detail on > > It also says: > "The capture area is described by a struct video_window." Confusing. But > since struct video_window takes in list of clipping rectangles, necessary > for overlay (I think), one would expect you to be right in that VIDIOCSWIN > would define overlay window instead of capture window, unlike in current > drivers. I'd guess it should read as follows: Index: Documentation/video4linux/API.html =================================================================== RCS file: /home/cvs/linux/Documentation/video4linux/API.html,v retrieving revision 1.4.4.1 diff -u -r1.4.4.1 API.html --- Documentation/video4linux/API.html 26 Jun 2002 22:35:01 -0000 1.4.4.1 +++ Documentation/video4linux/API.html 29 Sep 2003 13:25:14 -0000 @@ -88,9 +88,9 @@ Setting the base address to NULL indicates there is no physical frame buffer access. <P> -<H3>Capture Windows</H3> -The capture area is described by a <b>struct video_window</b>. This defines -a capture area and the clipping information if relevant. The +<H3>Overlay Windows</H3> +The overlay area is described by a <b>struct video_window</b>. This defines +an overlay area and the clipping information if relevant. The <b>VIDIOCGWIN</b> ioctl recovers the current settings and the <b>VIDIOCSWIN</b> sets new values. A successful call to <b>VIDIOCSWIN</b> indicates that a suitable set of parameters have been chosen. They do not @@ -124,10 +124,18 @@ is activated by passing the <b>VIDIOCCAPTURE</b> ioctl a value of 1, and disabled by passing it a value of 0. <P> -Some capture devices can capture a subfield of the image they actually see. -This is indicated when VIDEO_TYPE_SUBCAPTURE is defined. +<H3>Capture Windows</H3> +Some capture devices can capture a subfield of the image they actually +see. This is indicated when VIDEO_TYPE_SUBCAPTURE is defined. The video_capture describes the time and special subfields to capture. -The video_capture structure contains the following fields. +The capture area is described by a <b>struct video_capture</b>. The +<b>VIDIOCGCAPTURE</b> ioctl recovers the current settings and the +<b>VIDIOCSCAPTURE</b> sets new values. A successful call to +<b>VIDIOCSCAPTURE</b> indicates that a suitable set of parameters have +been chosen. They do not indicate that exactly what was requested was +granted. The program should call <b>VIDIOCGCAPTURE</b> to check if the +nearest match was suitable. The video_capture structure contains the +following fields. <P> <TABLE> <TR><TD><b>x</b></TD><TD>X co-ordinate of source rectangle to grab</TD> > The API documentation is confusing but I would recommend you to stay > compatible with existing applications, even if that would mean breaking the > word in the documentation. better to follow documentation and add compacrapability hooks... > As you can see, there's good reason for V4L2. Have you looked it, is it > better? Better to check before it's fixed in stone. V4L2 is better (of course I have to read it several times and try to work with it for some time), but I'd like that driver to go into 2.4 kernel series. I don't mind maintaining two versions of driver nor fixing other stuff (I sent patches for this two years ago, but got no reply). > >I'm wondering how to break the circle. What I'm a supposed to do if I want > >to set full size overlay window, but capture image in 1/2 of size?? > > Use V4L2? VIDIOCSCAPTURE with decimation=2 to set capture VIDIOCSWIN to set overlay. there is no problem in API itself, but in its implementations... [snip] ladis
