You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

226 lines
8.8 KiB

<chapter> <title> The video output layer </title>
<sect1> <title> Data structures and main loop </title>
<para>
Important data structures are defined in <filename> include/video.h
</filename> and <filename> include/video_output.h</filename>. The main
data structure is <type> picture_t</type>, which describes everything a
video decoder thread needs. Please refer to this file for more
information. Typically, <parameter> p_data </parameter> will be a
pointer to YUV planar picture.
</para>
<para>
Note also the <type> subpicture_t </type> structure. In fact the VLC SPU
decoder only parses the SPU header, and converts the SPU graphical data
to an internal format which can be rendered much faster. So a part of
the "real" SPU decoder lies in
<filename> src/video_output/video_spu.c</filename>.
</para>
<para>
The <type> vout_thread_t </type> structure is much more complex, but
you needn't understand everything. Basically the video output thread
manages a heap of pictures and subpictures (5 by default). Every
picture has a status (displayed, destroyed, empty...) and eventually
a presentation time. The main job of the video output is an infinite
loop to : [this is subject to change in the near future]
</para>
<orderedlist>
<listitem> <para> Find the next picture to display in the heap.
</para> </listitem>
<listitem> <para> Find the current subpicture to display.
</para> </listitem>
<listitem> <para> Render the picture (if the video output plug-in
doesn't support YUV overlay). Rendering will call an optimized
YUV plug-in, which will also do the scaling, add subtitles and
an optional picture information field.
</para> </listitem>
<listitem> <para> Sleep until the specified date.
</para> </listitem>
<listitem> <para> Display the picture (plug-in function). For
outputs which display RGB data, it is often accomplished with
a buffer switching. <parameter> p_vout-&gt;p_buffer </parameter>
is an array of two buffers where the YUV transform takes place,
and p_vout-&gt;i_buffer_index indicates the currently displayed
buffer.
</para> </listitem>
<listitem> <para> Manage events.
</para> </listitem>
</orderedlist>
</sect1>
<sect1> <title> Methods used by video decoders </title>
<para>
The video output exports a bunch of functions so that decoders can send
their decoded data. The most important function is
<function>vout_CreatePicture</function> which allocates the picture
buffer to the size indicated by the video decoder. It then just needs
to feed <type> (void *) </type><parameter> p_picture-&gt;p_data </parameter>
with the decoded data, and call <function> vout_DisplayPicture </function>
and <function> vout_DatePicture </function> upon necessary.
</para>
<itemizedlist>
<listitem> <para> <type> picture_t * </type> <function>
vout_CreatePicture </function>
<parameter> ( vout_thread_t *p_vout, int i_type, int i_width,
int i_height ) </parameter> :
Returns an allocated picture buffer. <parameter> i_type </parameter>
will be for instance <constant> YUV_420_PICTURE</constant>, and
i_width and i_height are in pixels.
</para>
<warning> <para> If no picture is available in the heap,
<function> vout_CreatePicture </function> will return NULL.
</para> </warning>
</listitem>
<listitem> <para> <function> vout_LinkPicture </function>
<parameter> ( vout_thread_t *p_vout, picture_t *p_pic ) </parameter> :
Increases the refcount of the picture, so that it doesn't get
accidently freed while the decoder still needs it. For instance,
an I or P picture can still be needed after displaying to decode
interleaved B pictures.
</para> </listitem>
<listitem> <para> <function> vout_UnlinkPicture </function>
<parameter> ( vout_thread_t *p_vout, picture_t *p_pic ) </parameter> :
Decreases the refcount of the picture. An unlink must be done
for every link previously made.
</para> </listitem>
<listitem> <para> <function> vout_DatePicture </function>
<parameter> ( vout_thread_t *p_vout, picture_t *p_pic ) </parameter> :
Gives the picture a presentation date. You can start working on
a picture before knowing precisely at what time it will be
displayed. For instance to date an I or P picture, you must wait
until you have decoded all previous B pictures (which are
indeed placed after - decoding order != presentation order).
</para> </listitem>
<listitem> <para> <function> vout_DisplayPicture </function>
<parameter> ( vout_thread_t *p_vout, picture_t *p_pic ) </parameter> :
Tells the video output that a picture has been completely decoded
and is ready to be rendered. It can be called before or after
<function> vout_DatePicture</function>.
</para> </listitem>
<listitem> <para> <function> vout_DestroyPicture </function>
<parameter> ( vout_thread_t *p_vout, picture_t *p_pic ) </parameter> :
Marks the picture as empty (useful in case of a stream parsing
error).
</para> </listitem>
<listitem> <para> <type> subpicture_t * </type> <function>
vout_CreateSubPicture </function> <parameter> ( vout_thread_t *p_vout,
int i_channel, int i_type ) </parameter> :
Returns an allocated subpicture buffer. <parameter> i_channel
</parameter> is the ID of the subpicture channel, <parameter> i_type
</parameter> is <constant> DVD_SUBPICTURE </constant> or
<constant> TEXT_SUBPICTURE</constant>, <parameter> i_size
</parameter> is the length in bytes of the packet.
</para> </listitem>
<listitem> <para> <function> vout_DisplaySubPicture </function>
<parameter> ( vout_thread_t *p_vout, subpicture_t *p_subpic )
</parameter> :
Tells the video output that a subpicture has been completely
decoded. It obsoletes the previous subpicture.
</para> </listitem>
<listitem> <para> <function> vout_DestroySubPicture </function>
<parameter> ( vout_thread_t *p_vout, subpicture_t *p_subpic )
</parameter> :
Marks the subpicture as empty.
</para> </listitem>
</itemizedlist>
</sect1>
<sect1> <title> How to write a video output plug-in </title>
<para>
A video output takes care of the system calls to display the pictures and
manage the output window. Have a look at <filename>
plugins/x11/vout_x11.c</filename>. You must write the following
functions :
</para>
<orderedlist>
<listitem> <para> <type> int </type> <function> vout_Probe </function>
<parameter> ( probedata_t *p_data ) </parameter> :
Returns a score between 0 and 999 to indicate whether it can
run on the architecture. 999 is the best. <parameter> p_data
</parameter> is currently unused.
</para> </listitem>
<listitem> <para> <type> int </type> <function> vout_Create </function>
<parameter> ( vout_thread_t *p_vout ) </parameter> :
Basically, initializes and opens a new window. Returns TRUE if
it failed.
</para> </listitem>
<listitem> <para> <type> int </type> <function> vout_Init </function>
<parameter> ( vout_thread_t *p_vout ) </parameter> :
Creates optional picture buffers (for instance ximages or
xvimages). Returns TRUE if it failed.
</para> </listitem>
<listitem> <para> <function> vout_End </function> <parameter>
( vout_thread_t *p_vout ) </parameter> :
Frees optional picture buffers.
</para> </listitem>
<listitem> <para> <function> vout_Destroy </function> <parameter>
( vout_thread_t *p_vout ) </parameter> :
Unmaps the window and frees all allocated resources.
</para> </listitem>
<listitem> <para> <type> int </type> <function> vout_Manage
</function> <parameter> ( vout_thread_t *p_vout ) </parameter> :
Manages events (including for instance resize events).
</para> </listitem>
<listitem> <para> <function> vout_Display </function> <parameter>
( vout_thread_t *p_vout ) </parameter> :
Displays a previously rendered buffer.
</para> </listitem>
<listitem> <para> <function> vout_SetPalette </function>
<parameter> ( vout_thread_t *p_vout, u16 *red, u16 *green,
u16 *blue, u16 *transp ) </parameter> :
Sets the 8 bpp palette. <parameter> red, green </parameter>
and <parameter> blue </parameter> are arrays of 256 <type>
unsigned shorts</type>.
</para> </listitem>
</orderedlist>
</sect1>
<sect1> <title> How to write a YUV plug-in </title>
<para>
Look at the C source <filename> plugins/yuv/transforms_yuv.c</filename>.
You need to redefine just the same transformations. Basically, it is a
matrix multiply operation. Good luck.
</para>
</sect1>
</chapter>