<title>Pipeline manipulation</title>
<para>
This chapter will discuss how you can manipulate your pipeline in several
- ways from your application on. Parts of this chapter are downright
- hackish, so be assured that you'll need some programming knowledge
- before you start reading this.
+ ways from your application on. Parts of this chapter are very
+ lowlevel, so be assured that you'll need some programming knowledge
+ and a good understanding of &GStreamer; before you start reading this.
</para>
<para>
Topics that will be discussed here include how you can insert data into
to listen to a pipeline's data processing.
</para>
- <sect1 id="section-data-probe">
- <title>Data probing</title>
+ <sect1 id="section-using-probes">
+ <title>Using probes</title>
<para>
Probing is best envisioned as a pad listener. Technically, a probe is
- nothing more than a signal callback that can be attached to a pad.
- Those signals are by default not fired at all (since that may have a
- negative impact on performance), but can be enabled by attaching a
- probe using <function>gst_pad_add_buffer_probe ()</function>,
- <function>gst_pad_add_event_probe ()</function>, or
- <function>gst_pad_add_data_probe ()</function>.
- Those functions attach the signal handler and
- enable the actual signal emission. Similarly, one can use the
- <function>gst_pad_remove_buffer_probe ()</function>,
- <function>gst_pad_remove_event_probe ()</function>, or
- <function>gst_pad_remove_data_probe ()</function>
- to remove the signal handlers again.
+ nothing more than a callback that can be attached to a pad.
+ You can attach a probe using <function>gst_pad_add_probe ()</function>.
+ Similarly, one can use the
+ <function>gst_pad_remove_probe ()</function>
+ to remove the callback again. The probe notifies you of any activity
+ that happens on the pad, like buffers, events and queries. You can
+ define what kind of notifications you are interested in when you
+ add the probe.
</para>
<para>
- Probes run in pipeline threading context, so callbacks should try to
- not block and generally not do any weird stuff, since this could
- have a negative impact on pipeline performance or, in case of bugs,
- cause deadlocks or crashes. More precisely, one should usually not
- call any GUI-related functions from within a probe callback, nor try
- to change the state of the pipeline. An application may post custom
- messages on the pipeline's bus though to communicate with the main
- application thread and have it do things like stop the pipeline.
+ The probe can notify you of the following activity on pads:
</para>
- <para>
- In any case, most common buffer operations
- that elements can do in <function>_chain ()</function> functions, can
- be done in probe callbacks as well. The example below gives a short
- impression on how to use them (even if this usage is not entirely
- correct, but more on that below):
- </para>
- <programlisting><!-- example-begin probe.c -->
-#include <gst/gst.h>
+ <itemizedlist>
+ <listitem>
+ <para>
+ A buffer is pushed or pulled. You want to specify the
+ GST_PAD_PROBE_TYPE_BUFFER when registering the probe. Because the
+ pad can be scheduled in different ways, it is possible to also
+ specify in what scheduling mode you are interested with the
+ optional GST_PAD_PROBE_TYPE_PUSH and GST_PAD_PROBE_TYPE_PULL
+ flags.
+ </para>
+ <para>
+ You can use this probe to inspect, modify or drop the buffer.
+ See <xref linkend="section-data-probes"/>.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ A bufferlist is pushed. Use the GST_PAD_PROBE_TYPE_BUFFER_LIST
+ when registering the probe.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ An event travels over a pad. Use the GST_PAD_PROBE_TYPE_EVENT_DOWNSTREAM
+ and GST_PAD_PROBE_TYPE_EVENT_UPSTREAM flags to select downstream
+ and upstream events. There is also a convenience
+ GST_PAD_PROBE_TYPE_EVENT_BOTH to be notified of events going both
+ upstream and downstream. By default, flush events do not cause
+ a notification. You need to explicitly enable GST_PAD_PROBE_TYPE_EVENT_FLUSH
+ to receive callbacks from flushing events. Events are always
+ only notified in push mode.
+ </para>
+ <para>
+ You can use this probe to inspect, modify or drop the event.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ A query travels over a pad. Use the GST_PAD_PROBE_TYPE_QUERY_DOWNSTREAM
+ and GST_PAD_PROBE_TYPE_QUERY_UPSTREAM flags to select downstream
+ and upstream queries. The convenience GST_PAD_PROBE_TYPE_QUERY_BOTH
+ can also be used to select both directions. Query probes will be
+ notified twice, once when the query travels upstream/downstream and
+ once when the query result is returned. You can select in what stage
+ the callback will be called with the GST_PAD_PROBE_TYPE_PUSH and
+ GST_PAD_PROBE_TYPE_PULL, respectively when the query is performed
+ and when the query result is returned.
+ </para>
+ <para>
+ You can use this probe to inspect or modify the query. You can also
+ answer the query in the probe callback by placing the result value
+ in the query and by returning GST_PAD_PROBE_DROP from the
+ callback.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ In addition to notifying you of dataflow, you can also ask the
+ probe to block the dataflow when the callback returns. This is
+ called a blocking probe and is activated by specifying the
+ GST_PAD_PROBE_TYPE_BLOCK flag. You can use this flag with the
+ other flags to only block dataflow on selected activity. A pad
+ becomes unblocked again if you remove the probe or when you return
+ GST_PAD_PROBE_REMOVE from the callback. You can let only the
+ currently blocked item pass by returning GST_PAD_PROBE_PASS
+ from the callback, it will block again on the next item.
+ </para>
+ <para>
+ Blocking probes are used to temporarily block pads because they
+ are unlinked or because you are going to unlink them. If the
+ dataflow is not blocked, the pipeline would go into an error
+ state if data is pushed on an unlinked pad. We will se how
+ to use blocking probes to partially preroll a pipeline.
+ See also <xref linkend="section-preroll-probes"/>.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Be notified when no activity is happening on a pad. You install
+ this probe with the GST_PAD_PROBE_TYPE_IDLE flag. You can specify
+ GST_PAD_PROBE_TYPE_PUSH and/or GST_PAD_PROBE_TYPE_PULL to
+ only be notified depending on the pad scheduling mode.
+ The IDLE probe is also a blocking probe in that it will not let
+ any data pass on the pad for as long as the IDLE probe is
+ installed.
+ </para>
+ <para>
+ You can use idle probes to dynamically relink a pad. We will see
+ how to use idle probes to replace an element in the pipeline.
+ See also <xref linkend="section-dynamic-pipelines"/>.
+ </para>
+ </listitem>
+ </itemizedlist>
-static gboolean
-cb_have_data (GstPad *pad,
- GstBuffer *buffer,
- gpointer u_data)
+ <sect2 id="section-data-probes">
+ <title>Data probes</title>
+ <para>
+ Data probes allow you to be notified when there is data passing
+ on a pad. When adding the probe, specify the GST_PAD_PROBE_TYPE_BUFFER
+ and/or GST_PAD_PROBE_TYPE_BUFFER_LIST.
+ </para>
+ <para>
+ Data probes run in pipeline streaming thread context, so callbacks
+ should try to not block and generally not do any weird stuff, since
+ this could have a negative impact on pipeline performance or, in case
+ of bugs, cause deadlocks or crashes. More precisely, one should usually
+ not call any GUI-related functions from within a probe callback, nor try
+ to change the state of the pipeline. An application may post custom
+ messages on the pipeline's bus though to communicate with the main
+ application thread and have it do things like stop the pipeline.
+ </para>
+ <para>
+ In any case, most common buffer operations
+ that elements can do in <function>_chain ()</function> functions, can
+ be done in probe callbacks as well. The example below gives a short
+ impression on how to use them.
+ </para>
+ <programlisting>
+<!-- example-begin probe.c -->
+<![CDATA[
+#include <gst/gst.h>
+
+static GstPadProbeReturn
+cb_have_data (GstPad *pad,
+ GstPadProbeInfo *info,
+ gpointer user_data)
{
gint x, y;
- GstMapInfo info;
+ GstMapInfo map;
guint16 *ptr, t;
+ GstBuffer *buffer;
+
+ buffer = GST_PAD_PROBE_INFO_BUFFER (info);
+
+ buffer = gst_buffer_make_writable (buffer);
- gst_buffer_map (buffer, &info, GST_MAP_WRITE);
+ gst_buffer_map (buffer, &map, GST_MAP_WRITE);
- ptr = info.data;
+ ptr = (guint16 *) map.data;
/* invert data */
- for (y = 0; y < 288; y++) {
- for (x = 0; x < 384 / 2; x++) {
+ for (y = 0; y < 288; y++) {
+ for (x = 0; x < 384 / 2; x++) {
t = ptr[384 - 1 - x];
ptr[384 - 1 - x] = ptr[x];
ptr[x] = t;
}
ptr += 384;
}
- gst_buffer_unmap (buffer, &info);
+ gst_buffer_unmap (buffer, &map);
- return TRUE;
+ GST_PAD_PROBE_INFO_DATA (info) = buffer;
+
+ return GST_PAD_PROBE_OK;
}
gint
GstPad *pad;
/* init GStreamer */
- gst_init (&argc, &argv);
+ gst_init (&argc, &argv);
loop = g_main_loop_new (NULL, FALSE);
/* build */
filter = gst_element_factory_make ("capsfilter", "filter");
g_assert (filter != NULL); /* should always exist */
- csp = gst_element_factory_make ("ffmpegcolorspace", "csp");
+ csp = gst_element_factory_make ("videoconvert", "csp");
if (csp == NULL)
- g_error ("Could not create 'ffmpegcolorspace' element");
+ g_error ("Could not create 'videoconvert' element");
sink = gst_element_factory_make ("xvimagesink", "sink");
if (sink == NULL) {
gst_bin_add_many (GST_BIN (pipeline), src, filter, csp, sink, NULL);
gst_element_link_many (src, filter, csp, sink, NULL);
- filtercaps = gst_caps_new_simple ("video/x-raw-rgb",
+ filtercaps = gst_caps_new_simple ("video/x-raw",
+ "format", G_TYPE_STRING, "RGB16",
"width", G_TYPE_INT, 384,
"height", G_TYPE_INT, 288,
"framerate", GST_TYPE_FRACTION, 25, 1,
- "bpp", G_TYPE_INT, 16,
- "depth", G_TYPE_INT, 16,
- "endianness", G_TYPE_INT, G_BYTE_ORDER,
NULL);
g_object_set (G_OBJECT (filter), "caps", filtercaps, NULL);
gst_caps_unref (filtercaps);
- pad = gst_element_get_pad (src, "src");
- gst_pad_add_buffer_probe (pad, G_CALLBACK (cb_have_data), NULL);
+ pad = gst_element_get_static_pad (src, "src");
+ gst_pad_add_probe (pad, GST_PAD_PROBE_TYPE_BUFFER,
+ (GstPadProbeCallback) cb_have_data, NULL, NULL);
gst_object_unref (pad);
/* run */
return 0;
}
- <!-- example-end probe.c --></programlisting>
- <para>
- Compare that output with the output of <quote>gst-launch-0.10
- videotestsrc ! xvimagesink</quote>, just so you know what you're
- looking for.
- </para>
- <para>
- The above example is not really correct though. Strictly speaking, a
- pad probe callback is only allowed to modify the buffer content if the
- buffer is writable, and it is only allowed to modify buffer metadata like
- timestamps, caps, etc. if the buffer metadata is writable. Whether this
- is the case or not depends a lot on the pipeline and the elements
- involved. Often enough, this is the case, but sometimes it is not,
- and if it is not then unexpected modification of the data or metadata
- can introduce bugs that are very hard to debug and track down. You can
- check if a buffer and its metadata are writable with
- <function>gst_buffer_is_writable ()</function> and
- <function>gst_buffer_is_metadata_writable ()</function>. Since you
- can't pass back a different buffer than the one passed in, there is no
- point of making a buffer writable in the callback function.
- </para>
- <para>
- Pad probes are suited best for looking at data as it passes through
- the pipeline. If you need to modify data, you should write your own
- GStreamer element. Base classes like GstAudioFilter, GstVideoFilter or
- GstBaseTransform make this fairly easy.
- </para>
- <para>
- If you just want to inspect buffers as they pass through the pipeline,
- you don't even need to set up pad probes. You could also just insert
- an identity element into the pipeline and connect to its "handoff"
- signal. The identity element also provides a few useful debugging tools
- like the "dump" property or the "last-message" property (the latter is
- enabled by passing the '-v' switch to gst-launch).
- </para>
+]]>
+<!-- example-end probe.c -->
+ </programlisting>
+ <para>
+ Compare that output with the output of <quote>gst-launch-1.0
+ videotestsrc ! xvimagesink</quote>, just so you know what you're
+ looking for.
+ </para>
+ <para>
+ Strictly speaking, a pad probe callback is only allowed to modify the
+ buffer content if the buffer is writable. Whether this is the case or
+ not depends a lot on the pipeline and the elements involved. Often
+ enough, this is the case, but sometimes it is not, and if it is not
+ then unexpected modification of the data or metadata can introduce
+ bugs that are very hard to debug and track down. You can check if a
+ buffer is writable with <function>gst_buffer_is_writable ()</function>.
+ Since you can pass back a different buffer than the one passed in,
+ it is a good idea to make the buffer writable in the callback function
+ with <function>gst_buffer_make_writable ()</function>.
+ </para>
+ <para>
+ Pad probes are suited best for looking at data as it passes through
+ the pipeline. If you need to modify data, you should better write your
+ own GStreamer element. Base classes like GstAudioFilter, GstVideoFilter or
+ GstBaseTransform make this fairly easy.
+ </para>
+ <para>
+ If you just want to inspect buffers as they pass through the pipeline,
+ you don't even need to set up pad probes. You could also just insert
+ an identity element into the pipeline and connect to its "handoff"
+ signal. The identity element also provides a few useful debugging tools
+ like the "dump" property or the "last-message" property (the latter is
+ enabled by passing the '-v' switch to gst-launch and by setting the
+ silent property on the identity to FALSE).
+ </para>
+ </sect2>
+
+ <sect2 id="section-preroll-probes">
+ <title>Play a region of a media file</title>
+ <para>
+ In this example we will show you how to play back a region of
+ a media file. The goal is to only play the part of a file
+ from 2 seconds to 5 seconds and then EOS.
+ </para>
+ <para>
+ In a first step we will set a uridecodebin element to the PAUSED
+ state and make sure that we block all the source pads that are
+ created. When all the source pads are blocked, we have data on
+ all source pads and we say that the uridecodebin is prerolled.
+ </para>
+ <para>
+ In a prerolled pipeline we can ask for the duration of the media
+ and we can also perform seeks. We are interested in performing a
+ seek operation on the pipeline to select the range of media
+ that we are interested in.
+ </para>
+ <para>
+ After we configure the region we are interested in, we can link
+ the sink element, unblock the source pads and set the pipeline to
+ the playing state. You will see that exactly the requested
+ region is played by the sink before it goes to EOS.
+ </para>
+ <para>
+ What follows is an example application that loosly follows this
+ algorithm.
+ </para>
+ <programlisting>
+<!-- example-begin blockprobe.c -->
+<![CDATA[
+#include <gst/gst.h>
+
+static GMainLoop *loop;
+static volatile gint counter;
+static GstBus *bus;
+static gboolean prerolled = FALSE;
+static GstPad *sinkpad;
+
+static void
+dec_counter (GstElement * pipeline)
+{
+ if (prerolled)
+ return;
+
+ if (g_atomic_int_dec_and_test (&counter)) {
+ /* all probes blocked and no-more-pads signaled, post
+ * message on the bus. */
+ prerolled = TRUE;
+
+ gst_bus_post (bus, gst_message_new_application (
+ GST_OBJECT_CAST (pipeline),
+ gst_structure_new_empty ("ExPrerolled")));
+ }
+}
+
+/* called when a source pad of uridecodebin is blocked */
+static GstPadProbeReturn
+cb_blocked (GstPad *pad,
+ GstPadProbeInfo *info,
+ gpointer user_data)
+{
+ GstElement *pipeline = GST_ELEMENT (user_data);
+
+ if (prerolled)
+ return GST_PAD_PROBE_REMOVE;
+
+ dec_counter (pipeline);
+
+ return GST_PAD_PROBE_OK;
+}
+
+/* called when uridecodebin has a new pad */
+static void
+cb_pad_added (GstElement *element,
+ GstPad *pad,
+ gpointer user_data)
+{
+ GstElement *pipeline = GST_ELEMENT (user_data);
+
+ if (prerolled)
+ return;
+
+ g_atomic_int_inc (&counter);
+
+ gst_pad_add_probe (pad, GST_PAD_PROBE_TYPE_BLOCK_DOWNSTREAM,
+ (GstPadProbeCallback) cb_blocked, pipeline, NULL);
+
+ /* try to link to the video pad */
+ gst_pad_link (pad, sinkpad);
+}
+
+/* called when uridecodebin has created all pads */
+static void
+cb_no_more_pads (GstElement *element,
+ gpointer user_data)
+{
+ GstElement *pipeline = GST_ELEMENT (user_data);
+
+ if (prerolled)
+ return;
+
+ dec_counter (pipeline);
+}
+
+/* called when a new message is posted on the bus */
+static void
+cb_message (GstBus *bus,
+ GstMessage *message,
+ gpointer user_data)
+{
+ GstElement *pipeline = GST_ELEMENT (user_data);
+
+ switch (GST_MESSAGE_TYPE (message)) {
+ case GST_MESSAGE_ERROR:
+ g_print ("we received an error!\n");
+ g_main_loop_quit (loop);
+ break;
+ case GST_MESSAGE_EOS:
+ g_print ("we reached EOS\n");
+ g_main_loop_quit (loop);
+ break;
+ case GST_MESSAGE_APPLICATION:
+ {
+ if (gst_message_has_name (message, "ExPrerolled")) {
+ /* it's our message */
+ g_print ("we are all prerolled, do seek\n");
+ gst_element_seek (pipeline,
+ 1.0, GST_FORMAT_TIME,
+ GST_SEEK_FLAG_FLUSH | GST_SEEK_FLAG_ACCURATE,
+ GST_SEEK_TYPE_SET, 2 * GST_SECOND,
+ GST_SEEK_TYPE_SET, 5 * GST_SECOND);
+
+ gst_element_set_state (pipeline, GST_STATE_PLAYING);
+ }
+ break;
+ }
+ default:
+ break;
+ }
+}
+
+gint
+main (gint argc,
+ gchar *argv[])
+{
+ GstElement *pipeline, *src, *csp, *vs, *sink;
+
+ /* init GStreamer */
+ gst_init (&argc, &argv);
+ loop = g_main_loop_new (NULL, FALSE);
+
+ if (argc < 2) {
+ g_print ("usage: %s <uri>", argv[0]);
+ return -1;
+ }
+
+ /* build */
+ pipeline = gst_pipeline_new ("my-pipeline");
+
+ bus = gst_pipeline_get_bus (GST_PIPELINE (pipeline));
+ gst_bus_add_signal_watch (bus);
+ g_signal_connect (bus, "message", (GCallback) cb_message,
+ pipeline);
+
+ src = gst_element_factory_make ("uridecodebin", "src");
+ if (src == NULL)
+ g_error ("Could not create 'uridecodebin' element");
+
+ g_object_set (src, "uri", argv[1], NULL);
+
+ csp = gst_element_factory_make ("videoconvert", "csp");
+ if (csp == NULL)
+ g_error ("Could not create 'videoconvert' element");
+
+ vs = gst_element_factory_make ("videoscale", "vs");
+ if (csp == NULL)
+ g_error ("Could not create 'videoscale' element");
+
+ sink = gst_element_factory_make ("autovideosink", "sink");
+ if (sink == NULL)
+ g_error ("Could not create 'autovideosink' element");
+
+ gst_bin_add_many (GST_BIN (pipeline), src, csp, vs, sink, NULL);
+
+ /* can't link src yet, it has no pads */
+ gst_element_link_many (csp, vs, sink, NULL);
+
+ sinkpad = gst_element_get_static_pad (csp, "sink");
+
+ /* for each pad block that is installed, we will increment
+ * the counter. for each pad block that is signaled, we
+ * decrement the counter. When the counter is 0 we post
+ * an app message to tell the app that all pads are
+ * blocked. Start with 1 that is decremented when no-more-pads
+ * is signaled to make sure that we only post the message
+ * after no-more-pads */
+ g_atomic_int_set (&counter, 1);
+
+ g_signal_connect (src, "pad-added",
+ (GCallback) cb_pad_added, pipeline);
+ g_signal_connect (src, "no-more-pads",
+ (GCallback) cb_no_more_pads, pipeline);
+
+ gst_element_set_state (pipeline, GST_STATE_PAUSED);
+
+ g_main_loop_run (loop);
+
+ gst_element_set_state (pipeline, GST_STATE_NULL);
+
+ gst_object_unref (sinkpad);
+ gst_object_unref (bus);
+ gst_object_unref (pipeline);
+ g_main_loop_unref (loop);
+
+ return 0;
+}
+]]>
+<!-- example-end blockprobe.c -->
+</programlisting>
+ <para>
+ Note that we use a custom application message to signal the
+ main thread that the uridecidebin is prerolled. The main thread
+ will then issue a flushing seek to the requested region. The
+ flush will temporarily unblock the pad and reblock them when
+ new data arrives again. We detect this second block to remove
+ the probes. Then we set the pipeline to PLAYING and it should
+ play from 2 to 5 seconds, then EOS and exit the application.
+ </para>
+ </sect2>
</sect1>
<sect1 id="section-data-spoof">
Many people have expressed the wish to use their own sources to inject
data into a pipeline. Some people have also expressed the wish to grab
the output in a pipeline and take care of the actual output inside
- their application. While either of these methods are stongly
- discouraged, &GStreamer; offers hacks to do this. <emphasis>However,
- there is no support for those methods.</emphasis> If it doesn't work,
- you're on your own. Also, synchronization, thread-safety and other
- things that you've been able to take for granted so far are no longer
- guaranteed if you use any of those methods. It's always better to
- simply write a plugin and have the pipeline schedule and manage it.
+ their application. While either of these methods are strongly
+ discouraged, &GStreamer; offers support for this.
+ <emphasis>Beware! You need to know what you are doing.</emphasis> Since
+ you don't have any support from a base class you need to thoroughly
+ understand state changes and synchronization. If it doesn't work,
+ there are a million ways to shoot yourself in the foot. It's always
+ better to simply write a plugin and have the base class manage it.
See the Plugin Writer's Guide for more information on this topic. Also
see the next section, which will explain how to embed plugins statically
in your application.
</para>
- <note><para>
- <ulink type="http"
- url="http://gstreamer.freedesktop.org/data/doc/gstreamer/head/gst-plugins-base-libs/html/gstreamer-app.html">New
- API</ulink> was developed to make data insertion and extraction easy
- for applications. It can be found as GstAppSrc and GstAppSink in the
- <ulink type="http"
- url="http://gstreamer.freedesktop.org/modules/gst-plugins-base.html">
- gst-plugins-base</ulink> module.
- </para></note>
<para>
- After all those disclaimers, let's start. There's three possible
- elements that you can use for the above-mentioned purposes. Those are
- called <quote>fakesrc</quote> (an imaginary source),
- <quote>fakesink</quote> (an imaginary sink) and <quote>identity</quote>
- (an imaginary filter). The same method applies to each of those
- elements. Here, we will discuss how to use those elements to insert
- (using fakesrc) or grab (using fakesink or identity) data from a
+ There's two possible elements that you can use for the above-mentioned
+ purposes. Those are called <quote>appsrc</quote> (an imaginary source)
+ and <quote>appsink</quote> (an imaginary sink). The same method applies
+ to each of those elements. Here, we will discuss how to use those
+ elements to insert (using appsrc) or grab (using appsink) data from a
pipeline, and how to set negotiation.
</para>
<para>
- Those who're paying close attention will notice that the purpose
- of identity is almost identical to that of probes. Indeed, this is
- true. Probes allow for the same purpose, and a bunch more, and
- with less overhead plus dynamic removing/adding of handlers, but
- apart from those, probes and identity have the same purpose, just
- in a completely different implementation type.
+ Both appsrc and appsink provide 2 sets of API. One API uses standard
+ GObject (action) signals and properties. The same API is also
+ available as a regular C api. The C api is more performant but
+ requires you to link to the app library in order to use the elements.
</para>
- <sect2 id="section-spoof-handoff">
- <title>Inserting or grabbing data</title>
+ <sect2 id="section-spoof-appsrc">
+ <title>Inserting data with appsrc</title>
<para>
- The three before-mentioned elements (fakesrc, fakesink and identity)
- each have a <quote>handoff</quote> signal that will be called in
- the <function>_get ()</function>- (fakesrc) or <function>_chain
- ()</function>-function (identity, fakesink). In the signal handler,
- you can set (fakesrc) or get (identity, fakesink) data to/from the
- provided buffer. Note that in the case of fakesrc, you have to set
- the size of the provided buffer using the <quote>sizemax</quote>
- property. For both fakesrc and fakesink, you also have to set the
- <quote>signal-handoffs</quote> property for this method to work.
+ First we look at some examples for appsrc, which lets you insert data
+ into the pipeline from the application. Appsrc has some configuration
+ options that define how it will operate. You should decide about the
+ following configurations:
</para>
+ <itemizedlist>
+ <listitem>
+ <para>
+ Will the appsrc operate in push or pull mode. The stream-type
+ property can be used to control this. stream-type of
+ <quote>random-access</quote> will activate pull mode scheduling
+ while the other stream-types activate push mode.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ The caps of the buffers that appsrc will push out. This needs to
+ be configured with the caps property. The caps must be set to a
+ fixed caps and will be used to negotiate a format downstream.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ If the appsrc operates in live mode or not. This can be configured
+ with the is-live property. When operating in live-mode it is
+ important to configure the min-latency and max-latency in appsrc.
+ The min-latency should be set to the amount of time it takes between
+ capturing a buffer and when it is pushed inside appsrc.
+ In live mode, you should timestamp the buffers with the pipeline
+ running-time when the first byte of the buffer was captured before
+ feeding them to appsrc. You can let appsrc do the timestaping with
+ the do-timestamp property (but then the min-latency must be set
+ to 0 because it timestamps based on the running-time when the buffer
+ entered appsrc).
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ The format of the SEGMENT event that appsrc will push. The format
+ has implications for how the running-time of the buffers will
+ be calculated so you must be sure you understand this. For
+ live sources you probably want to set the format property to
+ GST_FORMAT_TIME. For non-live source it depends on the media type
+ that you are handling. If you plan to timestamp the buffers, you
+ should probably put a GST_FORMAT_TIME format, otherwise
+ GST_FORMAT_BYTES might be appropriate.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ If appsrc operates in random-access mode, it is important to configure
+ the size property of appsrc with the number of bytes in the stream.
+ This will allow downstream elements to know the size of the media and
+ alows them to seek to the end of the stream when needed.
+ </para>
+ </listitem>
+ </itemizedlist>
<para>
- Note that your handoff function should <emphasis>not</emphasis>
- block, since this will block pipeline iteration. Also, do not try
- to use all sort of weird hacks in such functions to accomplish
- something that looks like synchronization or so; it's not the right
- way and will lead to issues elsewhere. If you're doing any of this,
- you're basically misunderstanding the &GStreamer; design.
+ The main way of handling data to appsrc is by using the function
+ <function>gst_app_src_push_buffer ()</function> or by emiting the
+ push-buffer action signal. This will put the buffer onto a queue from
+ which appsrc will read from in its streaming thread. It is important
+ to note that data transport will not happen from the thread that
+ performed the push-buffer call.
</para>
- </sect2>
-
- <sect2 id="section-spoof-format">
- <title>Forcing a format</title>
- <para>
- Sometimes, when using fakesrc as a source in your pipeline, you'll
- want to set a specific format, for example a video size and format
- or an audio bitsize and number of channels. You can do this by
- forcing a specific <classname>GstCaps</classname> on the pipeline,
- which is possible by using <emphasis>filtered caps</emphasis>. You
- can set a filtered caps on a link by using the
- <quote>capsfilter</quote> element in between the two elements, and
- specifying a <classname>GstCaps</classname> as
- <quote>caps</quote> property on this element. It will then
- only allow types matching that specified capability set for
- negotiation. See also <xref linkend="section-caps-filter"/>.
+ <para>
+ The <quote>max-bytes</quote> property controls how much data can be
+ queued in appsrc before appsrc considers the queue full. A filled
+ internal queue will always signal the <quote>enough-data</quote>
+ signal, which signals the application that it should stop pushing
+ data into appsrc. The <quote>block</quote> property will cause appsrc to
+ block the push-buffer method until free data becomes available again.
+ </para>
+ <para>
+ When the internal queue is running out of data, the
+ <quote>need-data</quote> signal is emitted, which signals the application
+ that it should start pushing more data into appsrc.
</para>
- </sect2>
-
- <sect2 id="section-spoof-example">
- <title>Example application</title>
<para>
- This example application will generate black/white (it switches
- every second) video to an X-window output by using fakesrc as a
- source and using filtered caps to force a format. Since the depth
- of the image depends on your X-server settings, we use a colorspace
- conversion element to make sure that the output to your X server
- will have the correct bitdepth. You can also set timestamps on the
- provided buffers to override the fixed framerate.
+ In addition to the <quote>need-data</quote> and <quote>enough-data</quote>
+ signals, appsrc can emit the <quote>seek-data</quote> signal when the
+ <quote>stream-mode</quote> property is set to <quote>seekable</quote>
+ or <quote>random-access</quote>. The signal argument will contain the
+ new desired position in the stream expressed in the unit set with the
+ <quote>format</quote> property. After receiving the seek-data signal,
+ the application should push-buffers from the new position.
</para>
- <programlisting><!-- example-begin fakesrc.c -->
-#include <string.h> /* for memset () */
-#include <gst/gst.h>
+ <para>
+ When the last byte is pushed into appsrc, you must call
+ <function>gst_app_src_end_of_stream ()</function> to make it send
+ an EOS downstream.
+ </para>
+ <para>
+ These signals allow the application to operate appsrc in push and
+ pull mode as will be explained next.
+ </para>
+
+ <sect3 id="section-spoof-appsrc-push">
+ <title>Using appsrc in push mode</title>
+ <para>
+ When appsrc is configured in push mode (stream-type is stream or
+ seekable), the application repeatedly calls the push-buffer method
+ with a new buffer. Optionally, the queue size in the appsrc can be
+ controlled with the enough-data and need-data signals by respectively
+ stopping/starting the push-buffer calls. The value of the
+ min-percent property defines how empty the internal appsrc queue
+ needs to be before the need-data signal will be fired. You can set
+ this to some value >0 to avoid completely draining the queue.
+ </para>
+ <para>
+ When the stream-type is set to seekable, don't forget to implement
+ a seek-data callback.
+ </para>
+ <para>
+ Use this model when implementing various network protocols or
+ hardware devices.
+ </para>
+ </sect3>
+
+ <sect3 id="section-spoof-appsrc-pull">
+ <title>Using appsrc in pull mode</title>
+ <para>
+ In the pull model, data is fed to appsrc from the need-data signal
+ handler. You should push exactly the amount of bytes requested in the
+ need-data signal. You are only allowed to push less bytes when you are
+ at the end of the stream.
+ </para>
+ <para>
+ Use this model for file access or other randomly accessable sources.
+ </para>
+ </sect3>
+
+ <sect3 id="section-spoof-appsrc-ex">
+ <title>Appsrc example</title>
+ <para>
+ This example application will generate black/white (it switches
+ every second) video to an Xv-window output by using appsrc as a
+ source with caps to force a format. We use a colorspace
+ conversion element to make sure that we feed the right format to
+ your X server. We configure a video stream with a variable framerate
+ (0/1) and we set the timestamps on the outgoing buffers in such
+ a way that we play 2 frames per second.
+ </para>
+ <para>
+ Note how we use the pull mode method of pushing new buffers into
+ appsrc although appsrc is running in push mode.
+ </para>
+ <programlisting>
+<!-- example-begin appsrc.c -->
+<![CDATA[
+#include <gst/gst.h>
+
+static GMainLoop *loop;
static void
-cb_handoff (GstElement *fakesrc,
- GstBuffer *buffer,
- GstPad *pad,
- gpointer user_data)
+cb_need_data (GstElement *appsrc,
+ guint unused_size,
+ gpointer user_data)
{
static gboolean white = FALSE;
- GstMapInfo info;
-
- gst_buffer_map (buffer, &info, GST_MAP_WRITE);
+ static GstClockTime timestamp = 0;
+ GstBuffer *buffer;
+ guint size;
+ GstFlowReturn ret;
+
+ size = 385 * 288 * 2;
+
+ buffer = gst_buffer_new_allocate (NULL, size, NULL);
/* this makes the image black/white */
- memset (info.data, white ? 0xff : 0x0, info.size);
+ gst_buffer_memset (buffer, 0, white ? 0xff : 0x0, size);
+
white = !white;
- gst_buffer_unmap (buffer, &info);
+ GST_BUFFER_PTS (buffer) = timestamp;
+ GST_BUFFER_DURATION (buffer) = gst_util_uint64_scale_int (1, GST_SECOND, 2);
+
+ timestamp += GST_BUFFER_DURATION (buffer);
+
+ g_signal_emit_by_name (appsrc, "push-buffer", buffer, &ret);
+
+ if (ret != GST_FLOW_OK) {
+ /* something wrong, stop pushing */
+ g_main_loop_quit (loop);
+ }
}
gint
main (gint argc,
gchar *argv[])
{
- GstElement *pipeline, *fakesrc, *flt, *conv, *videosink;
- GMainLoop *loop;
+ GstElement *pipeline, *appsrc, *conv, *videosink;
/* init GStreamer */
- gst_init (&argc, &argv);
+ gst_init (&argc, &argv);
loop = g_main_loop_new (NULL, FALSE);
/* setup pipeline */
pipeline = gst_pipeline_new ("pipeline");
- fakesrc = gst_element_factory_make ("fakesrc", "source");
- flt = gst_element_factory_make ("capsfilter", "flt");
- conv = gst_element_factory_make ("ffmpegcolorspace", "conv");
+ appsrc = gst_element_factory_make ("appsrc", "source");
+ conv = gst_element_factory_make ("videoconvert", "conv");
videosink = gst_element_factory_make ("xvimagesink", "videosink");
/* setup */
- g_object_set (G_OBJECT (flt), "caps",
- gst_caps_new_simple ("video/x-raw-rgb",
+ g_object_set (G_OBJECT (appsrc), "caps",
+ gst_caps_new_simple ("video/x-raw",
+ "format", G_TYPE_STRING, "RGB16",
"width", G_TYPE_INT, 384,
"height", G_TYPE_INT, 288,
- "framerate", GST_TYPE_FRACTION, 1, 1,
- "bpp", G_TYPE_INT, 16,
- "depth", G_TYPE_INT, 16,
- "endianness", G_TYPE_INT, G_BYTE_ORDER,
+ "framerate", GST_TYPE_FRACTION, 0, 1,
NULL), NULL);
- gst_bin_add_many (GST_BIN (pipeline), fakesrc, flt, conv, videosink, NULL);
- gst_element_link_many (fakesrc, flt, conv, videosink, NULL);
+ gst_bin_add_many (GST_BIN (pipeline), appsrc, conv, videosink, NULL);
+ gst_element_link_many (appsrc, conv, videosink, NULL);
- /* setup fake source */
- g_object_set (G_OBJECT (fakesrc),
- "signal-handoffs", TRUE,
- "sizemax", 384 * 288 * 2,
- "sizetype", 2, NULL);
- g_signal_connect (fakesrc, "handoff", G_CALLBACK (cb_handoff), NULL);
+ /* setup appsrc */
+ g_object_set (G_OBJECT (appsrc),
+ "stream-type", 0,
+ "format", GST_FORMAT_TIME, NULL);
+ g_signal_connect (appsrc, "need-data", G_CALLBACK (cb_need_data), NULL);
/* play */
gst_element_set_state (pipeline, GST_STATE_PLAYING);
/* clean up */
gst_element_set_state (pipeline, GST_STATE_NULL);
gst_object_unref (GST_OBJECT (pipeline));
+ g_main_loop_unref (loop);
+
+ return 0;
+ }
+]]>
+<!-- example-end appsrc.c -->
+ </programlisting>
+ </sect3>
+ </sect2>
+
+ <sect2 id="section-spoof-appsink">
+ <title>Grabbing data with appsink</title>
+ <para>
+ Unlike appsrc, appsink is a little easier to use. It also supports
+ a pull and push based model of getting data from the pipeline.
+ </para>
+ <para>
+ The normal way of retrieving samples from appsink is by using the
+ <function>gst_app_sink_pull_sample()</function> and
+ <function>gst_app_sink_pull_preroll()</function> methods or by using
+ the <quote>pull-sample</quote> and <quote>pull-preroll</quote>
+ signals. These methods block until a sample becomes available in the
+ sink or when the sink is shut down or reaches EOS.
+ </para>
+ <para>
+ Appsink will internally use a queue to collect buffers from the
+ streaming thread. If the application is not pulling samples fast
+ enough, this queue will consume a lot of memory over time. The
+ <quote>max-buffers</quote> property can be used to limit the queue
+ size. The <quote>drop</quote> property controls whether the
+ streaming thread blocks or if older buffers are dropped when the
+ maximum queue size is reached. Note that blocking the streaming thread
+ can negatively affect real-time performance and should be avoided.
+ </para>
+ <para>
+ If a blocking behaviour is not desirable, setting the
+ <quote>emit-signals</quote> property to TRUE will make appsink emit
+ the <quote>new-sample</quote> and <quote>new-preroll</quote> signals
+ when a sample can be pulled without blocking.
+ </para>
+ <para>
+ The <quote>caps</quote> property on appsink can be used to control
+ the formats that appsink can receive. This property can contain
+ non-fixed caps, the format of the pulled samples can be obtained by
+ getting the sample caps.
+ </para>
+ <para>
+ If one of the pull-preroll or pull-sample methods return NULL, the
+ appsink is stopped or in the EOS state. You can check for the EOS state
+ with the <quote>eos</quote> property or with the
+ <function>gst_app_sink_is_eos()</function> method.
+ </para>
+ <para>
+ The eos signal can also be used to be informed when the EOS state is
+ reached to avoid polling.
+ </para>
+ <para>
+ Consider configuring the following properties in the appsink:
+ </para>
+ <itemizedlist>
+ <listitem>
+ <para>
+ The <quote>sync</quote> property if you want to have the sink
+ base class synchronize the buffer against the pipeline clock
+ before handing you the sample.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Enable Quality-of-Service with the <quote>qos</quote> property.
+ If you are dealing with raw video frames and let the base class
+ sycnhronize on the clock, it might be a good idea to also let
+ the base class send QOS events upstream.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ The caps property that contains the accepted caps. Upstream elements
+ will try to convert the format so that it matches the configured
+ caps on appsink. You must still check the
+ <classname>GstSample</classname> to get the actual caps of the
+ buffer.
+ </para>
+ </listitem>
+ </itemizedlist>
+
+ <sect3 id="section-spoof-appsink-ex">
+ <title>Appsink example</title>
+ <para>
+ What follows is an example on how to capture a snapshot of a video
+ stream using appsink.
+ </para>
+ <programlisting>
+<!-- example-begin appsink.c -->
+<![CDATA[
+#include <gst/gst.h>
+#ifdef HAVE_GTK
+#include <gtk/gtk.h>
+#endif
+
+#include <stdlib.h>
+
+#define CAPS "video/x-raw,format=RGB,width=160,pixel-aspect-ratio=1/1"
+
+int
+main (int argc, char *argv[])
+{
+ GstElement *pipeline, *sink;
+ gint width, height;
+ GstSample *sample;
+ gchar *descr;
+ GError *error = NULL;
+ gint64 duration, position;
+ GstStateChangeReturn ret;
+ gboolean res;
+ GstMapInfo map;
+
+ gst_init (&argc, &argv);
+
+ if (argc != 2) {
+ g_print ("usage: %s <uri>\n Writes snapshot.png in the current directory\n",
+ argv[0]);
+ exit (-1);
+ }
+
+ /* create a new pipeline */
+ descr =
+ g_strdup_printf ("uridecodebin uri=%s ! videoconvert ! videoscale ! "
+ " appsink name=sink caps=\"" CAPS "\"", argv[1]);
+ pipeline = gst_parse_launch (descr, &error);
+
+ if (error != NULL) {
+ g_print ("could not construct pipeline: %s\n", error->message);
+ g_clear_error (&error);
+ exit (-1);
+ }
+
+ /* get sink */
+ sink = gst_bin_get_by_name (GST_BIN (pipeline), "sink");
+
+ /* set to PAUSED to make the first frame arrive in the sink */
+ ret = gst_element_set_state (pipeline, GST_STATE_PAUSED);
+ switch (ret) {
+ case GST_STATE_CHANGE_FAILURE:
+ g_print ("failed to play the file\n");
+ exit (-1);
+ case GST_STATE_CHANGE_NO_PREROLL:
+ /* for live sources, we need to set the pipeline to PLAYING before we can
+ * receive a buffer. We don't do that yet */
+ g_print ("live sources not supported yet\n");
+ exit (-1);
+ default:
+ break;
+ }
+ /* This can block for up to 5 seconds. If your machine is really overloaded,
+ * it might time out before the pipeline prerolled and we generate an error. A
+ * better way is to run a mainloop and catch errors there. */
+ ret = gst_element_get_state (pipeline, NULL, NULL, 5 * GST_SECOND);
+ if (ret == GST_STATE_CHANGE_FAILURE) {
+ g_print ("failed to play the file\n");
+ exit (-1);
+ }
+
+ /* get the duration */
+ gst_element_query_duration (pipeline, GST_FORMAT_TIME, &duration);
+
+ if (duration != -1)
+ /* we have a duration, seek to 5% */
+ position = duration * 5 / 100;
+ else
+ /* no duration, seek to 1 second, this could EOS */
+ position = 1 * GST_SECOND;
+
+ /* seek to the a position in the file. Most files have a black first frame so
+ * by seeking to somewhere else we have a bigger chance of getting something
+ * more interesting. An optimisation would be to detect black images and then
+ * seek a little more */
+ gst_element_seek_simple (pipeline, GST_FORMAT_TIME,
+ GST_SEEK_FLAG_KEY_UNIT | GST_SEEK_FLAG_FLUSH, position);
+
+ /* get the preroll buffer from appsink, this block untils appsink really
+ * prerolls */
+ g_signal_emit_by_name (sink, "pull-preroll", &sample, NULL);
+
+ /* if we have a buffer now, convert it to a pixbuf. It's possible that we
+ * don't have a buffer because we went EOS right away or had an error. */
+ if (sample) {
+ GstBuffer *buffer;
+ GstCaps *caps;
+ GstStructure *s;
+
+ /* get the snapshot buffer format now. We set the caps on the appsink so
+ * that it can only be an rgb buffer. The only thing we have not specified
+ * on the caps is the height, which is dependant on the pixel-aspect-ratio
+ * of the source material */
+ caps = gst_sample_get_caps (sample);
+ if (!caps) {
+ g_print ("could not get snapshot format\n");
+ exit (-1);
+ }
+ s = gst_caps_get_structure (caps, 0);
+
+ /* we need to get the final caps on the buffer to get the size */
+ res = gst_structure_get_int (s, "width", &width);
+ res |= gst_structure_get_int (s, "height", &height);
+ if (!res) {
+ g_print ("could not get snapshot dimension\n");
+ exit (-1);
+ }
+
+ /* create pixmap from buffer and save, gstreamer video buffers have a stride
+ * that is rounded up to the nearest multiple of 4 */
+ buffer = gst_sample_get_buffer (sample);
+ gst_buffer_map (buffer, &map, GST_MAP_READ);
+#ifdef HAVE_GTK
+ pixbuf = gdk_pixbuf_new_from_data (map.data,
+ GDK_COLORSPACE_RGB, FALSE, 8, width, height,
+ GST_ROUND_UP_4 (width * 3), NULL, NULL);
+
+ /* save the pixbuf */
+ gdk_pixbuf_save (pixbuf, "snapshot.png", "png", &error, NULL);
+#endif
+ gst_buffer_unmap (buffer, &map);
+ gst_sample_unref (sample);
+ } else {
+ g_print ("could not make snapshot\n");
+ }
+
+ /* cleanup and exit */
+ gst_element_set_state (pipeline, GST_STATE_NULL);
+ gst_object_unref (pipeline);
+
+ exit (0);
+}
+]]>
+<!-- example-end appsink.c -->
+</programlisting>
+ </sect3>
+ </sect2>
+ </sect1>
+
+ <sect1 id="section-spoof-format">
+ <title>Forcing a format</title>
+ <para>
+ Sometimes you'll want to set a specific format, for example a video
+ size and format or an audio bitsize and number of channels. You can
+ do this by forcing a specific <classname>GstCaps</classname> on
+ the pipeline, which is possible by using
+ <emphasis>filtered caps</emphasis>. You can set a filtered caps on
+ a link by using the <quote>capsfilter</quote> element in between the
+ two elements, and specifying a <classname>GstCaps</classname> as
+ <quote>caps</quote> property on this element. It will then
+ only allow types matching that specified capability set for
+ negotiation. See also <xref linkend="section-caps-filter"/>.
+ </para>
+
+ <sect2 id="section-dynamic-format">
+ <title>Changing format in a PLAYING pipeline</title>
+ <para>
+ It is also possible to dynamically change the format in a pipeline
+ while PLAYING. This can simply be done by changing the caps
+ property on a capsfilter. The capsfilter will send a RECONFIGURE
+ event upstream that will make the upstream element attempt to
+ renegotiate a new format and allocator. This only works if
+ the upstream element is not using fixed caps on the source pad.
+ </para>
+ <para>
+ Below is an example of how you can change the caps of a pipeline
+ while in the PLAYING state:
+ </para>
+ <programlisting>
+<!-- example-begin dynformat.c -->
+<![CDATA[
+#include <stdlib.h>
+
+#include <gst/gst.h>
+
+#define MAX_ROUND 100
+
+int
+main (int argc, char **argv)
+{
+ GstElement *pipe, *filter;
+ GstCaps *caps;
+ gint width, height;
+ gint xdir, ydir;
+ gint round;
+ GstMessage *message;
+
+ gst_init (&argc, &argv);
+
+ pipe = gst_parse_launch_full ("videotestsrc ! capsfilter name=filter ! "
+ "ximagesink", NULL, GST_PARSE_FLAG_NONE, NULL);
+ g_assert (pipe != NULL);
+
+ filter = gst_bin_get_by_name (GST_BIN (pipe), "filter");
+ g_assert (filter);
+
+ width = 320;
+ height = 240;
+ xdir = ydir = -10;
+
+ for (round = 0; round < MAX_ROUND; round++) {
+ gchar *capsstr;
+ g_print ("resize to %dx%d (%d/%d) \r", width, height, round, MAX_ROUND);
+
+ /* we prefer our fixed width and height but allow other dimensions to pass
+ * as well */
+ capsstr = g_strdup_printf ("video/x-raw, width=(int)%d, height=(int)%d",
+ width, height);
+
+ caps = gst_caps_from_string (capsstr);
+ g_free (capsstr);
+ g_object_set (filter, "caps", caps, NULL);
+ gst_caps_unref (caps);
+
+ if (round == 0)
+ gst_element_set_state (pipe, GST_STATE_PLAYING);
+
+ width += xdir;
+ if (width >= 320)
+ xdir = -10;
+ else if (width < 200)
+ xdir = 10;
+
+ height += ydir;
+ if (height >= 240)
+ ydir = -10;
+ else if (height < 150)
+ ydir = 10;
+
+ message =
+ gst_bus_poll (GST_ELEMENT_BUS (pipe), GST_MESSAGE_ERROR,
+ 50 * GST_MSECOND);
+ if (message) {
+ g_print ("got error \n");
+
+ gst_message_unref (message);
+ }
+ }
+ g_print ("done \n");
+
+ gst_object_unref (filter);
+ gst_element_set_state (pipe, GST_STATE_NULL);
+ gst_object_unref (pipe);
return 0;
}
- <!-- example-end fakesrc.c --></programlisting>
+]]>
+<!-- example-end dynformat.c -->
+ </programlisting>
+ <para>
+ Note how we use <function>gst_bus_poll()</function> with a
+ small timeout to get messages and also introduce a short
+ sleep.
+ </para>
+ <para>
+ It is possible to set multiple caps for the capsfilter separated
+ with a ;. The capsfilter will try to renegotiate to the first
+ possible format from the list.
+ </para>
</sect2>
</sect1>
- <sect1 id="section-data-manager">
- <title>Embedding static elements in your application</title>
+ <sect1 id="section-dynamic-pipelines">
+ <title>Dynamically changing the pipeline</title>
<para>
- The <ulink type="http"
- url="http://gstreamer.freedesktop.org/data/doc/gstreamer/head/pwg/html/index.html">Plugin
- Writer's Guide</ulink> describes in great detail how to write elements
- for the &GStreamer; framework. In this section, we will solely discuss
- how to embed such elements statically in your application. This can be
- useful for application-specific elements that have no use elsewhere in
- &GStreamer;.
+ In this section we talk about some techniques for dynamically
+ modifying the pipeline. We are talking specifically about changing
+ the pipeline while it is in the PLAYING state without interrupting
+ the flow.
</para>
<para>
- Dynamically loaded plugins contain a structure that's defined using
- <function>GST_PLUGIN_DEFINE ()</function>. This structure is loaded
- when the plugin is loaded by the &GStreamer; core. The structure
- contains an initialization function (usually called
- <function>plugin_init</function>) that will be called right after that.
- It's purpose is to register the elements provided by the plugin with
- the &GStreamer; framework. If you want to embed elements directly in
- your application, the only thing you need to do is to replace
- <function>GST_PLUGIN_DEFINE ()</function> with
- <function>GST_PLUGIN_DEFINE_STATIC ()</function>. This will cause the
- elements to be registered when your application loads, and the elements
- will from then on be available like any other element, without them
- having to be dynamically loadable libraries. In the example below, you
- would be able to call <function>gst_element_factory_make
- ("my-element-name", "some-name")</function> to create an instance of the
- element.
+ There are some important things to consider when building dynamic
+ pipelines:
</para>
+ <itemizedlist>
+ <listitem>
+ <para>
+ When removing elements from the pipeline, make sure that there
+ is no dataflow on unlinked pads because that will cause a fatal
+ pipeline error. Always block source pads (in push mode) or
+ sink pads (in pull mode) before unlinking pads.
+ See also <xref linkend="section-dynamic-changing"/>.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ When adding elements to a pipeline, make sure to put the element
+ into the right state, usually the same state as the parent, before
+ allowing dataflow the element. When an element is newly created,
+ it is in the NULL state and will return an error when it
+ receives data.
+ See also <xref linkend="section-dynamic-changing"/>.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ When adding elements to a pipeline, &GStreamer; will by default
+ set the clock and base-time on the element to the current values
+ of the pipeline. This means that the element will be able to
+ construct the same pipeline running-time as the other elements
+ in the pipeline. This means that sinks will synchronize buffers
+ like the other sinks in the pipeline and that sources produce
+ buffers with a running-time that matches the other sources.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ When unlinking elements from an upstream chain, always make sure
+ to flush any queued data in the element by sending an EOS event
+ down the element sink pad(s) and by waiting that the EOS leaves
+ the elements (with an event probe).
+ See also <xref linkend="section-dynamic-changing"/>.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ A live source will produce buffers with a running-time of the
+ current running-time in the pipeline.
+ </para>
+ <para>
+ A pipeline without a live source produces buffers with a
+ running-time starting from 0. Likewise, after a flushing seek,
+ those pipelines reset the running-time back to 0.
+ </para>
+ <para>
+ The running-time can be changed with
+ <function>gst_pad_set_offset ()</function>. It is important to
+ know the running-time of the elements in the pipeline in order
+ to maintain synchronization.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Adding elements might change the state of the pipeline. Adding a
+ non-prerolled sink, for example, brings the pipeline back to the
+ prerolling state. Removing a non-prerolled sink, for example, might
+ change the pipeline to PAUSED and PLAYING state.
+ </para>
+ <para>
+ Adding a live source cancels the preroll stage and put the pipeline
+ to the playing state. Adding a live source or other live elements
+ might also change the latency of a pipeline.
+ </para>
+ <para>
+ Adding or removing elements to the pipeline might change the clock
+ selection of the pipeline. If the newly added element provides a clock,
+ it might be worth changing the clock in the pipeline to the new
+ clock. If, on the other hand, the element that provides the clock
+ for the pipeline is removed, a new clock has to be selected.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Adding and removing elements might cause upstream or downstream
+ elements to renegotiate caps and or allocators. You don't really
+ need to do anything from the application, plugins largely
+ adapt themself to the new pipeline topology in order to optimize
+ their formats and allocation strategy.
+ </para>
+ <para>
+ What is important is that when you add, remove or change elements
+ in the pipeline, it is possible that the pipeline needs to
+ negotiate a new format and this can fail. Usually you can fix this
+ by inserting the right converter elements where needed.
+ See also <xref linkend="section-dynamic-changing"/>.
+ </para>
+ </listitem>
+ </itemizedlist>
- <programlisting>
-/*
- * Here, you would write the actual plugin code.
- */
+ <para>
+ &GStreamer; offers support for doing about any dynamic pipeline
+ modification but it requires you to know a bit of details before
+ you can do this without causing pipeline errors. In the following
+ sections we will demonstrate a couple of typical use-cases.
+ </para>
-[..]
+ <sect2 id="section-dynamic-changing">
+ <title>Changing elements in a pipeline</title>
+ <para>
+ In the next example we look at the following chain of elements:
+ </para>
+ <programlisting>
+ - ----. .----------. .---- -
+ element1 | | element2 | | element3
+ src -> sink src -> sink
+ - ----' '----------' '---- -
+ </programlisting>
+ <para>
+ We want to change element2 by element4 while the pipeline is in
+ the PLAYING state. Let's say that element2 is a visualization and
+ that you want to switch the visualization in the pipeline.
+ </para>
+ <para>
+ We can't just unlink element2's sinkpad from element1's source
+ pad because that would leave element1's source pad
+ unlinked and would cause a streaming error in the pipeline when
+ data is pushed on the source pad.
+ The technique is to block the dataflow from element1's source pad
+ before we change element2 by element4 and then resume dataflow
+ as shown in the following steps:
+ </para>
+ <itemizedlist>
+ <listitem>
+ <para>
+ Block element1's source pad with a blocking pad probe. When the
+ pad is blocked, the probe callback will be called.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Inside the block callback nothing is flowing between element1
+ and element2 and nothing will flow until unblocked.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Unlink element1 and element2.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Make sure data is flushed out of element2. Some elements might
+ internally keep some data, you need to make sure not to lose data
+ by forcing it out of element2. You can do this by pushing EOS into
+ element2, like this:
+ </para>
+ <itemizedlist>
+ <listitem>
+ <para>
+ Put an event probe on element2's source pad.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Send EOS to element2's sinkpad. This makes sure the all the
+ data inside element2 is forced out.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Wait for the EOS event to appear on element2's source pad.
+ When the EOS is received, drop it and remove the event
+ probe.
+ </para>
+ </listitem>
+ </itemizedlist>
+ </listitem>
+ <listitem>
+ <para>
+ Unlink element2 and element3. You can now also remove element2
+ from the pipeline and set the state to NULL.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Add element4 to the pipeline, if not already added. Link element4
+ and element3. Link element1 and element4.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Make sure element4 is in the same state as the rest of the elements
+ in the pipeline. It should be at least in the PAUSED state before
+ it can receive buffers and events.
+ </para>
+ </listitem>
+ <listitem>
+ <para>
+ Unblock element1's source pad probe. This will let new data into
+ element4 and continue streaming.
+ </para>
+ </listitem>
+ </itemizedlist>
+ <para>
+ The above algorithm works when the source pad is blocked, i.e. when
+ there is dataflow in the pipeline. If there is no dataflow, there is
+ also no point in changing the element (just yet) so this algorithm can
+ be used in the PAUSED state as well.
+ </para>
+ <para>
+ Let show you how this works with an example. This example changes the
+ video effect on a simple pipeline every second.
+ </para>
+ <programlisting>
+<!-- example-begin effectswitch.c -->
+<![CDATA[
+#include <gst/gst.h>
+
+static gchar *opt_effects = NULL;
+
+#define DEFAULT_EFFECTS "identity,exclusion,navigationtest," \
+ "agingtv,videoflip,vertigotv,gaussianblur,shagadelictv,edgetv"
+
+static GstPad *blockpad;
+static GstElement *conv_before;
+static GstElement *conv_after;
+static GstElement *cur_effect;
+static GstElement *pipeline;
+
+static GQueue effects = G_QUEUE_INIT;
+
+static GstPadProbeReturn
+event_probe_cb (GstPad * pad, GstPadProbeInfo * info, gpointer user_data)
+{
+ GMainLoop *loop = user_data;
+ GstElement *next;
+
+ if (GST_EVENT_TYPE (GST_PAD_PROBE_INFO_DATA (info)) != GST_EVENT_EOS)
+ return GST_PAD_PROBE_OK;
+
+ gst_pad_remove_probe (pad, GST_PAD_PROBE_INFO_ID (info));
+
+ /* push current event back into the queue */
+ g_queue_push_tail (&effects, gst_object_ref (cur_effect));
+ /* take next effect from the queue */
+ next = g_queue_pop_head (&effects);
+ if (next == NULL) {
+ GST_DEBUG_OBJECT (pad, "no more effects");
+ g_main_loop_quit (loop);
+ return GST_PAD_PROBE_DROP;
+ }
+
+ g_print ("Switching from '%s' to '%s'..\n", GST_OBJECT_NAME (cur_effect),
+ GST_OBJECT_NAME (next));
+
+ gst_element_set_state (cur_effect, GST_STATE_NULL);
+
+ /* remove unlinks automatically */
+ GST_DEBUG_OBJECT (pipeline, "removing %" GST_PTR_FORMAT, cur_effect);
+ gst_bin_remove (GST_BIN (pipeline), cur_effect);
+
+ GST_DEBUG_OBJECT (pipeline, "adding %" GST_PTR_FORMAT, next);
+ gst_bin_add (GST_BIN (pipeline), next);
+
+ GST_DEBUG_OBJECT (pipeline, "linking..");
+ gst_element_link_many (conv_before, next, conv_after, NULL);
+
+ gst_element_set_state (next, GST_STATE_PLAYING);
+
+ cur_effect = next;
+ GST_DEBUG_OBJECT (pipeline, "done");
+
+ return GST_PAD_PROBE_DROP;
+}
+
+static GstPadProbeReturn
+pad_probe_cb (GstPad * pad, GstPadProbeInfo * info, gpointer user_data)
+{
+ GstPad *srcpad, *sinkpad;
+
+ GST_DEBUG_OBJECT (pad, "pad is blocked now");
+
+ /* remove the probe first */
+ gst_pad_remove_probe (pad, GST_PAD_PROBE_INFO_ID (info));
+
+ /* install new probe for EOS */
+ srcpad = gst_element_get_static_pad (cur_effect, "src");
+ gst_pad_add_probe (srcpad, GST_PAD_PROBE_TYPE_BLOCK |
+ GST_PAD_PROBE_TYPE_EVENT_DOWNSTREAM, event_probe_cb, user_data, NULL);
+ gst_object_unref (srcpad);
+
+ /* push EOS into the element, the probe will be fired when the
+ * EOS leaves the effect and it has thus drained all of its data */
+ sinkpad = gst_element_get_static_pad (cur_effect, "sink");
+ gst_pad_send_event (sinkpad, gst_event_new_eos ());
+ gst_object_unref (sinkpad);
+
+ return GST_PAD_PROBE_OK;
+}
+
+static gboolean
+timeout_cb (gpointer user_data)
+{
+ gst_pad_add_probe (blockpad, GST_PAD_PROBE_TYPE_BLOCK_DOWNSTREAM,
+ pad_probe_cb, user_data, NULL);
+
+ return TRUE;
+}
static gboolean
-register_elements (GstPlugin *plugin)
+bus_cb (GstBus * bus, GstMessage * msg, gpointer user_data)
{
- return gst_element_register (plugin, "my-element-name",
- GST_RANK_NONE, MY_PLUGIN_TYPE);
+ GMainLoop *loop = user_data;
+
+ switch (GST_MESSAGE_TYPE (msg)) {
+ case GST_MESSAGE_ERROR:{
+ GError *err = NULL;
+ gchar *dbg;
+
+ gst_message_parse_error (msg, &err, &dbg);
+ gst_object_default_error (msg->src, err, dbg);
+ g_clear_error (&err);
+ g_free (dbg);
+ g_main_loop_quit (loop);
+ break;
+ }
+ default:
+ break;
+ }
+ return TRUE;
}
-GST_PLUGIN_DEFINE_STATIC (
- GST_VERSION_MAJOR,
- GST_VERSION_MINOR,
- "my-private-plugins",
- "Private elements of my application",
- register_elements,
- VERSION,
- "LGPL",
- "my-application",
- "http://www.my-application.net/"
-)
- </programlisting>
+int
+main (int argc, char **argv)
+{
+ GOptionEntry options[] = {
+ {"effects", 'e', 0, G_OPTION_ARG_STRING, &opt_effects,
+ "Effects to use (comma-separated list of element names)", NULL},
+ {NULL}
+ };
+ GOptionContext *ctx;
+ GError *err = NULL;
+ GMainLoop *loop;
+ GstElement *src, *q1, *q2, *effect, *filter1, *filter2, *sink;
+ gchar **effect_names, **e;
+
+ ctx = g_option_context_new ("");
+ g_option_context_add_main_entries (ctx, options, NULL);
+ g_option_context_add_group (ctx, gst_init_get_option_group ());
+ if (!g_option_context_parse (ctx, &argc, &argv, &err)) {
+ g_print ("Error initializing: %s\n", err->message);
+ g_clear_error (&err);
+ g_option_context_free (ctx);
+ return 1;
+ }
+ g_option_context_free (ctx);
+
+ if (opt_effects != NULL)
+ effect_names = g_strsplit (opt_effects, ",", -1);
+ else
+ effect_names = g_strsplit (DEFAULT_EFFECTS, ",", -1);
+
+ for (e = effect_names; e != NULL && *e != NULL; ++e) {
+ GstElement *el;
+
+ el = gst_element_factory_make (*e, NULL);
+ if (el) {
+ g_print ("Adding effect '%s'\n", *e);
+ g_queue_push_tail (&effects, el);
+ }
+ }
+
+ pipeline = gst_pipeline_new ("pipeline");
+
+ src = gst_element_factory_make ("videotestsrc", NULL);
+ g_object_set (src, "is-live", TRUE, NULL);
+
+ filter1 = gst_element_factory_make ("capsfilter", NULL);
+ gst_util_set_object_arg (G_OBJECT (filter1), "caps",
+ "video/x-raw, width=320, height=240, "
+ "format={ I420, YV12, YUY2, UYVY, AYUV, Y41B, Y42B, "
+ "YVYU, Y444, v210, v216, NV12, NV21, UYVP, A420, YUV9, YVU9, IYU1 }");
+
+ q1 = gst_element_factory_make ("queue", NULL);
+
+ blockpad = gst_element_get_static_pad (q1, "src");
+
+ conv_before = gst_element_factory_make ("videoconvert", NULL);
+
+ effect = g_queue_pop_head (&effects);
+ cur_effect = effect;
+
+ conv_after = gst_element_factory_make ("videoconvert", NULL);
+
+ q2 = gst_element_factory_make ("queue", NULL);
+
+ filter2 = gst_element_factory_make ("capsfilter", NULL);
+ gst_util_set_object_arg (G_OBJECT (filter2), "caps",
+ "video/x-raw, width=320, height=240, "
+ "format={ RGBx, BGRx, xRGB, xBGR, RGBA, BGRA, ARGB, ABGR, RGB, BGR }");
+
+ sink = gst_element_factory_make ("ximagesink", NULL);
+
+ gst_bin_add_many (GST_BIN (pipeline), src, filter1, q1, conv_before, effect,
+ conv_after, q2, sink, NULL);
+
+ gst_element_link_many (src, filter1, q1, conv_before, effect, conv_after,
+ q2, sink, NULL);
+
+ gst_element_set_state (pipeline, GST_STATE_PLAYING);
+
+ loop = g_main_loop_new (NULL, FALSE);
+
+ gst_bus_add_watch (GST_ELEMENT_BUS (pipeline), bus_cb, loop);
+
+ g_timeout_add_seconds (1, timeout_cb, loop);
+
+ g_main_loop_run (loop);
+
+ gst_element_set_state (pipeline, GST_STATE_NULL);
+ gst_object_unref (pipeline);
+
+ return 0;
+}
+]]>
+<!-- example-end effectswitch.c -->
+ </programlisting>
+ <para>
+ Note how we added videoconvert elements before and after the effect.
+ This is needed because some elements might operate in different
+ colorspaces than other elements. By inserting the conversion elements
+ you ensure that the right format can be negotiated at any time.
+ </para>
+ </sect2>
</sect1>
+
</chapter>
projects / platform / upstream / gstreamer.git / blobdiff