mirrors/gstreamer
mirrors/gstreamer
1
1
Fork 0
mirror of https://gitlab.freedesktop.org/gstreamer/gstreamer.git synced 2024-11-27 20:21:24 +00:00
Code Issues 1 Projects Releases Wiki Activity

baseaudiocodec: ... and also rename to baseaudiodecoder

Browse source
This commit is contained in:
Mark Nauwelaerts 2011-03-01 14:08:18 +01:00 committed by Tim-Philipp Müller
parent dfd7616f60
commit 90d99f23c6
2 changed files with 1159 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

939
gst-libs/gst/audio/gstbaseaudiodecoder.c Normal file
View file

@ -0,0 +1,939 @@
/* GStreamer
* Copyright (C) 2009 Igalia S.L.
* Author: Iago Toral Quiroga <itoral@igalia.com>
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Library General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Library General Public License for more details.
*
* You should have received a copy of the GNU Library General Public
* License along with this library; if not, write to the
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
* Boston, MA 02111-1307, USA.
*/
/**
* SECTION:gstbaseaudiodecoder
* @short_description: Base class for codec elements
* @see_also: #GstBaseTransform, #GstBaseSource, #GstBaseSink
*
* #GstBaseAudioDecoder is the base class for codec elements ion GStreamer. It is
* a layer on top of #GstElement that provides simplified interface to plugin
* writers, hangling many details for you. Its way of operation is explained
* below.
*
* Subclasses are responsible for specifying the codec's source pad caps. For
* that purpose they should provide an implementation of ::negotiate_src_caps.
* If the subclass provides an implementation of this method, it will be
* invoked by #GstBaseAudioDecoder on its sink_setcaps function. Otherwise, if
* the subclass does not provide an implementation of this method, the subclass
* will be responsible for calling gst_base_audio_decoder_set_src_caps() to
* complete the caps negotiation before any buffers are pushed out.
*
* Each buffer received on the codec's sink pad is pushed to its input
* adapter. When there is enough data present in the input adapter
* (configured in the #GstBaseAudioDecoder:input-buffer-size
* property), the method ::process_data is called on the subclass. Subclasses
* must provide an implementation of this method, which would read from the
* input adapter, encode or decode the data, and push it to the output adapter.
* If #GstBaseAudioDecoder:input-buffer-size is set to 0 ::process_data will be
* invoked as soon as there is any data on the input adapter.
*
* Similarly, when there is enough data present on the output adapter,
* (configured in the #GstBaseAudioDecoder:output-buffer-size property),
* buffers will be pushed out through the codec's source pad. If
* #GstBaseAudioDecoder:output-buffer-size is set to 0 a buffer will be pushed
* out as soon as there is any data present on the output adapter. Notice
* that if no implementation of ::negotiate_src_caps has been provided by the
* subclass, it must call gst_base_audio_decoder_set_src_caps() to complete
* the caps negotiation process or otherwise attempting to push buffers
* through the codec's source pad will fail.
*
* It is possible for subclasses to take control on how and when buffers
* are pushed out by overriding the ::push_data method. If subclasses
* provide an implementation of this method #GstBaseAudioDecoder will
* not push buffers out by itself, instead, whenever there* is data present
* in the output adapter, it will invoke ::push_data on subclass, which
* will implement there any logic necessary for pushing buffers out when
* appropriate. In this mode of operation, the property
* ::output_buffer_size is ignored in #GstBaseAudioDecoder. In any case,
* buffers should be pushed using gst_base_audio_decoder_push_buffer().
*
* #GstBaseAudioDecoder checks for discontinuities and handles them
* appropriately when pushing buffers out (setting the discontinuous
* flag on the output buffers when necessary). Subclasses can check if
* the data present on the adapters represents a discontinuity by checking
* the discont field of #GstBaseAudioDecoder. Also, subclasses can provide
* an implementation for the ::handle_discont method, which will be invoked
* whenever a discontinuity is detected on the source stream.
*
* Because data is not processed immediately and is stored in adapters,
* depending on how the actual codec operates it may be possible to
* receive an end-of-stream event before all the data in the adapters
* has been processed and pushed out. If this can happen, the subclass
* must provide implementation of the ::flush_input method, which should
* then read the data present int the input adapter, process it and
* store the result in the output adapter. The subclass may also want
* provide an implementation for the ::flush_output method, which would
* take care of reading the data from the output adapter and push it
* out through the codec's source pad. If no implementation is provided
* for the ::flush_out method, #GstBaseAudioDecoder will create a single
* buffer with all the data present in the output adapter and push it
* out. If a subclass needs to force a flush on the adapters for some
* reason, it should call gst_base_audio_decoder_flush(), which will then
* invoke ::flush_input and/or ::flush_output appropriately.
*
* Subclasses may provide an implementation for the ::start, ::stop
* and ::reset methods when needed. This methods will be called
* from #GstBaseAudioDecoder when needed (on state changes,
* discontinuities, etc), so they must never invoke the
* implementation on the parent class. When a subclass needs to
* start, stop or reset the codec itself, it should use the public
* functions gst_base_audio_decoder_{start,stop,reset}(), which call
* the corresponding methods on the parent class, which will then
* call the functions provided by the subclass (if any).
*
* #GstBaseAudioDecoder also provides an sink event handler.
* Subclasses that want to be notified on these events, can provide
* an implementation of the ::event function, which will be called after
* #GstBaseAudioDecoder has processed the event itself.
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include "gstbaseaudiodecoder.h"
#include <gst/audio/audio.h>
#include <string.h>
/*
* FIXME: maybe we need more work with the segments (see ac3 decoder)
*/
GST_DEBUG_CATEGORY (baseaudiodecoder_debug);
#define GST_CAT_DEFAULT baseaudiodecoder_debug
/* ----- Signals and properties ----- */
enum
{
LAST_SIGNAL
};
enum
{
PROP_0,
PROP_INPUT_BUFFER_SIZE,
PROP_OUTPUT_BUFFER_SIZE
};
/* ----- Function prototypes ----- */
static void gst_base_audio_decoder_finalize (GObject * object);
static void gst_base_audio_decoder_set_property (GObject * object,
guint prop_id, const GValue * value, GParamSpec * pspec);
static void gst_base_audio_decoder_get_property (GObject * object,
guint prop_id, GValue * value, GParamSpec * pspec);
static GstStateChangeReturn gst_base_audio_decoder_change_state (GstElement *
element, GstStateChange transition);
static gboolean gst_base_audio_decoder_sink_event (GstPad * pad,
GstEvent * event);
static gboolean gst_base_audio_decoder_sink_setcaps (GstPad * pad,
GstCaps * caps);
static GstFlowReturn gst_base_audio_decoder_chain (GstPad * pad,
GstBuffer * buf);
static void gst_base_audio_decoder_handle_discont (GstBaseAudioDecoder * codec,
GstBuffer * buf);
/* ----- GObject setup ----- */
GST_BOILERPLATE (GstBaseAudioDecoder, gst_base_audio_decoder, GstElement,
GST_TYPE_ELEMENT);
static void
gst_base_audio_decoder_base_init (gpointer g_class)
{
GST_DEBUG_CATEGORY_INIT (baseaudiodecoder_debug, "baseaudiodecoder", 0,
"Base Audio Codec Classes");
}
static void
gst_base_audio_decoder_class_init (GstBaseAudioDecoderClass * klass)
{
GObjectClass *gobject_class;
GstElementClass *element_class;
gobject_class = G_OBJECT_CLASS (klass);
element_class = GST_ELEMENT_CLASS (klass);
gobject_class->set_property = gst_base_audio_decoder_set_property;
gobject_class->get_property = gst_base_audio_decoder_get_property;
gobject_class->finalize = gst_base_audio_decoder_finalize;
element_class->change_state = gst_base_audio_decoder_change_state;
klass->start = NULL;
klass->stop = NULL;
klass->reset = NULL;
klass->event = NULL;
klass->handle_discont = NULL;
klass->flush_input = NULL;
klass->flush_output = NULL;
klass->process_data = NULL;
klass->push_data = NULL;
klass->negotiate_src_caps = NULL;
/* Properties */
g_object_class_install_property (gobject_class, PROP_INPUT_BUFFER_SIZE,
g_param_spec_uint ("input-buffer-size", "Input buffer size",
"Size of the input buffers in bytes (0 for not setting a "
"particular size)", 0, G_MAXUINT, 0, G_PARAM_READWRITE));
g_object_class_install_property (gobject_class, PROP_OUTPUT_BUFFER_SIZE,
g_param_spec_uint ("output-buffer-size", "Output buffer size",
"Size of the output buffers in bytes (0 for not setting a "
"particular size)", 0, G_MAXUINT, 0, G_PARAM_READWRITE));
}
static void
gst_base_audio_decoder_init (GstBaseAudioDecoder * codec,
GstBaseAudioDecoderClass * klass)
{
GstPadTemplate *pad_template;
GST_DEBUG ("gst_base_audio_decoder_init");
/* Setup sink pad */
pad_template =
gst_element_class_get_pad_template (GST_ELEMENT_CLASS (klass), "sink");
g_return_if_fail (pad_template != NULL);
codec->sinkpad = gst_pad_new_from_template (pad_template, "sink");
gst_pad_set_event_function (codec->sinkpad,
gst_base_audio_decoder_sink_event);
gst_pad_set_setcaps_function (codec->sinkpad,
gst_base_audio_decoder_sink_setcaps);
gst_pad_set_chain_function (codec->sinkpad, gst_base_audio_decoder_chain);
gst_element_add_pad (GST_ELEMENT (codec), codec->sinkpad);
/* Setup source pad */
pad_template =
gst_element_class_get_pad_template (GST_ELEMENT_CLASS (klass), "src");
g_return_if_fail (pad_template != NULL);
codec->srcpad = gst_pad_new_from_template (pad_template, "src");
gst_pad_use_fixed_caps (codec->srcpad);
gst_element_add_pad (GST_ELEMENT (codec), codec->srcpad);
/* Setup adapters */
codec->input_adapter = gst_adapter_new ();
codec->output_adapter = gst_adapter_new ();
codec->input_buffer_size = 0;
codec->output_buffer_size = 0;
/* Setup state */
memset (&codec->state, 0, sizeof (GstAudioState));
gst_segment_init (&codec->state.segment, GST_FORMAT_TIME);
codec->started = FALSE;
codec->bytes_in = 0;
codec->bytes_out = 0;
codec->discont = TRUE;
codec->caps_set = FALSE;
codec->first_ts = -1;
codec->last_ts = -1;
}
static void
gst_base_audio_decoder_get_property (GObject * object, guint prop_id,
GValue * value, GParamSpec * pspec)
{
GstBaseAudioDecoder *codec;
codec = GST_BASE_AUDIO_DECODER (object);
switch (prop_id) {
case PROP_INPUT_BUFFER_SIZE:
g_value_set_uint (value, codec->input_buffer_size);
break;
case PROP_OUTPUT_BUFFER_SIZE:
g_value_set_uint (value, codec->output_buffer_size);
break;
default:
G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
break;
}
}
static void
gst_base_audio_decoder_set_property (GObject * object, guint prop_id,
const GValue * value, GParamSpec * pspec)
{
GstBaseAudioDecoder *codec;
codec = GST_BASE_AUDIO_DECODER (object);
switch (prop_id) {
case PROP_INPUT_BUFFER_SIZE:
codec->input_buffer_size = g_value_get_uint (value);
break;
case PROP_OUTPUT_BUFFER_SIZE:
codec->output_buffer_size = g_value_get_uint (value);
break;
default:
G_OBJECT_WARN_INVALID_PROPERTY_ID (object, prop_id, pspec);
break;
}
}
static void
gst_base_audio_decoder_finalize (GObject * object)
{
GstBaseAudioDecoder *codec;
g_return_if_fail (GST_IS_BASE_AUDIO_DECODER (object));
codec = GST_BASE_AUDIO_DECODER (object);
if (codec->input_adapter) {
g_object_unref (codec->input_adapter);
}
if (codec->output_adapter) {
g_object_unref (codec->output_adapter);
}
if (codec->codec_data) {
gst_buffer_unref (codec->codec_data);
}
G_OBJECT_CLASS (parent_class)->finalize (object);
}
/* ----- Private element implementation ----- */
static void
gst_base_audio_decoder_read_state_from_caps (GstBaseAudioDecoder * codec,
GstCaps * caps)
{
GstStructure *structure;
const GValue *codec_data;
structure = gst_caps_get_structure (caps, 0);
if (codec->codec_data) {
gst_buffer_unref (codec->codec_data);
codec->codec_data = NULL;
}
codec_data = gst_structure_get_value (structure, "codec_data");
if (codec_data && G_VALUE_TYPE (codec_data) == GST_TYPE_BUFFER) {
codec->codec_data = gst_value_get_buffer (codec_data);
}
gst_structure_get_int (structure, "channels", &codec->state.channels);
gst_structure_get_int (structure, "rate", &codec->state.rate);
gst_structure_get_int (structure, "depth", &codec->state.sample_depth);
gst_structure_get_int (structure, "width", &codec->state.bytes_per_sample);
codec->state.bytes_per_sample /= 8;
codec->state.frame_size =
codec->state.bytes_per_sample * codec->state.channels;
}
static gboolean
gst_base_audio_decoder_sink_event (GstPad * pad, GstEvent * event)
{
GstBaseAudioDecoder *codec;
GstBaseAudioDecoderClass *codec_class;
gboolean ret = FALSE;
codec = GST_BASE_AUDIO_DECODER (gst_pad_get_parent (pad));
codec_class = GST_BASE_AUDIO_DECODER_GET_CLASS (codec);
switch (GST_EVENT_TYPE (event)) {
case GST_EVENT_EOS:
/* Flush any data still present in the adapters */
gst_base_audio_decoder_flush (codec);
ret = gst_pad_push_event (codec->srcpad, event);
break;
case GST_EVENT_FLUSH_STOP:
gst_base_audio_decoder_reset (codec);
ret = gst_pad_push_event (codec->srcpad, event);
break;
case GST_EVENT_NEWSEGMENT:
{
gboolean update;
GstFormat format;
gdouble rate, arate;
gint64 start, stop, time;
gst_event_parse_new_segment_full (event, &update, &rate, &arate, &format,
&start, &stop, &time);
if (format != GST_FORMAT_TIME)
goto newseg_wrong_format;
if (rate <= 0.0)
goto newseg_wrong_rate;
GST_DEBUG ("news egment %lld %lld", start, time);
gst_segment_set_newsegment_full (&codec->state.segment,
update, rate, arate, format, start, stop, time);
ret = gst_pad_push_event (codec->srcpad, event);
break;
}
default:
ret = gst_pad_push_event (codec->srcpad, event);
break;
}
/* Let the subclass see the event too */
if (codec_class->event) {
if (!codec_class->event (codec, event)) {
ret = FALSE;
goto subclass_event_error;
}
}
done:
gst_object_unref (codec);
return ret;
newseg_wrong_format:
GST_DEBUG ("received non TIME newsegment");
gst_event_unref (event);
goto done;
newseg_wrong_rate:
GST_DEBUG ("negative rates not supported");
gst_event_unref (event);
goto done;
subclass_event_error:
GST_DEBUG ("codec implementation failed to proces event");
gst_event_unref (event);
goto done;
}
static gboolean
gst_base_audio_decoder_sink_setcaps (GstPad * pad, GstCaps * caps)
{
GstBaseAudioDecoder *codec;
GstBaseAudioDecoderClass *codec_class;
codec = GST_BASE_AUDIO_DECODER (gst_pad_get_parent (pad));
codec_class = GST_BASE_AUDIO_DECODER_GET_CLASS (codec);
GST_DEBUG ("gst_base_audio_decoder_sink_setcaps %" GST_PTR_FORMAT, caps);
/* Let the subclass provide the source caps and we will set them
on the codec's source pad */
if (codec_class->negotiate_src_caps) {
GstCaps *src_caps;
src_caps = codec_class->negotiate_src_caps (codec, caps);
if (!gst_base_audio_decoder_set_src_caps (codec, src_caps)) {
GST_DEBUG ("Caps negotiation failed!");
g_object_unref (codec);
gst_caps_unref (src_caps);
return FALSE;
}
gst_caps_unref (src_caps);
} else {
/* If the subclass does not provide a negotiate_src_caps method, then
it will be responsible for calling gst_base_audio_decoder_set_src_caps
with appropriate caps before we try to push buffers out */
GST_DEBUG ("Subclass does not provide negotiate_src_caps, is that ok?");
}
gst_base_audio_decoder_start (codec);
g_object_unref (codec);
return TRUE;
}
static GstStateChangeReturn
gst_base_audio_decoder_change_state (GstElement * element,
GstStateChange transition)
{
GstBaseAudioDecoder *codec;
GstBaseAudioDecoderClass *codec_class;
GstStateChangeReturn ret;
codec = GST_BASE_AUDIO_DECODER (element);
codec_class = GST_BASE_AUDIO_DECODER_GET_CLASS (element);
switch (transition) {
case GST_STATE_CHANGE_NULL_TO_READY:
if (!gst_base_audio_decoder_start (codec)) {
goto start_failed;
}
break;
case GST_STATE_CHANGE_READY_TO_PAUSED:
if (!gst_base_audio_decoder_reset (codec)) {
goto reset_failed;
}
break;
case GST_STATE_CHANGE_PAUSED_TO_PLAYING:
break;
default:
break;
}
ret = parent_class->change_state (element, transition);
switch (transition) {
case GST_STATE_CHANGE_PLAYING_TO_PAUSED:
break;
case GST_STATE_CHANGE_PAUSED_TO_READY:
if (!gst_base_audio_decoder_stop (codec)) {
goto stop_failed;
}
break;
case GST_STATE_CHANGE_READY_TO_NULL:
break;
default:
break;
}
return ret;
start_failed:
{
GST_ELEMENT_ERROR (codec, LIBRARY, INIT, (NULL), ("Failed to start codec"));
return GST_STATE_CHANGE_FAILURE;
}
reset_failed:
{
GST_ELEMENT_ERROR (codec, LIBRARY, INIT, (NULL), ("Failed to reset codec"));
return GST_STATE_CHANGE_FAILURE;
}
stop_failed:
{
GST_ELEMENT_ERROR (codec, LIBRARY, INIT, (NULL), ("Failed to stop codec"));
return GST_STATE_CHANGE_FAILURE;
}
}
static void
gst_base_audio_decoder_handle_discont (GstBaseAudioDecoder * codec,
GstBuffer * buffer)
{
GstBaseAudioDecoderClass *codec_class;
codec_class = GST_BASE_AUDIO_DECODER_GET_CLASS (codec);
/* Reset codec on discont */
if (codec->started) {
gst_base_audio_decoder_reset (codec);
}
codec->discont = TRUE;
/* Let the subclass do its stuff too if that is needed */
if (codec_class->handle_discont) {
codec_class->handle_discont (codec, buffer);
}
}
static GstFlowReturn
gst_base_audio_decoder_chain (GstPad * pad, GstBuffer * buf)
{
GstBaseAudioDecoder *codec;
GstBaseAudioDecoderClass *codec_class;
GstBuffer *outbuf;
GstFlowReturn ret;
guint bytes_ready;
guint64 timestamp;
codec = GST_BASE_AUDIO_DECODER (gst_pad_get_parent (pad));
codec_class = GST_BASE_AUDIO_DECODER_GET_CLASS (codec);
GST_DEBUG ("gst_base_audio_decoder_chain");
/* Make sure we have started our codec */
if (G_UNLIKELY (!codec->started)) {
if (G_UNLIKELY (!gst_base_audio_decoder_start (codec))) {
GST_ELEMENT_ERROR (codec, LIBRARY, INIT, (NULL),
("Failed to start codec"));
gst_object_unref (codec);
return GST_FLOW_ERROR;
}
}
/* Handle timestamps */
timestamp = GST_BUFFER_TIMESTAMP (buf);
if (GST_CLOCK_TIME_IS_VALID (timestamp)) {
GST_DEBUG ("buffer timestamp %" GST_TIME_FORMAT " duration:%"
GST_TIME_FORMAT, GST_TIME_ARGS (timestamp),
GST_TIME_ARGS (GST_BUFFER_DURATION (buf)));
if (gst_adapter_available (codec->input_adapter) == 0) {
codec->first_ts = timestamp;
}
codec->last_ts = timestamp;
}
/* Check for discontinuity */
if (G_UNLIKELY (GST_BUFFER_FLAG_IS_SET (buf, GST_BUFFER_FLAG_DISCONT))) {
GST_DEBUG ("received DISCONT buffer");
gst_base_audio_decoder_handle_discont (codec, buf);
}
/* Push buffer to the input adapter so the codec can
take data from it as needed */
codec->bytes_in += GST_BUFFER_SIZE (buf);
gst_adapter_push (codec->input_adapter, buf);
GST_DEBUG ("Input buffer size: %ld bytes", GST_BUFFER_SIZE (buf));
/* Check if we have enough data to be processed. While we have
enough data on the input adapter, instruct the element to
process it */
ret = GST_FLOW_OK;
bytes_ready = gst_adapter_available (codec->input_adapter);
while (ret == GST_FLOW_OK && bytes_ready > 0 &&
bytes_ready >= codec->input_buffer_size) {
GST_DEBUG ("Processing data");
ret = codec_class->process_data (codec);
bytes_ready = gst_adapter_available (codec->input_adapter);
GST_DEBUG ("%ld bytes remaining on the input", bytes_ready);
}
/* FIXME: is it possible that we have enough data in the output
adapter but we have to wait for more data before we can
push buffers out? In that case we need a custom GST_FLOW.
Not sure if we could handle pushing buffers here in that
case though, since we always push in output_buffer_size
blocks. */
/* If no error was raised, check if we can push buffers out */
if (G_LIKELY (ret == GST_FLOW_OK)) {
bytes_ready = gst_adapter_available (codec->output_adapter);
GST_DEBUG ("Processed input correctly");
GST_DEBUG ("%ld bytes on the output", bytes_ready);
/* If the subclass wants to control how buffers are pushed out
let it do it */
if (bytes_ready > 0 && codec_class->push_data) {
GST_DEBUG ("Calling push_data on the subclass");
codec_class->push_data (codec);
} else if (bytes_ready > 0 && bytes_ready >= codec->output_buffer_size) {
/* We have enough data in the output adapter, so take a buffer, apply
clipping, push it out and repeat while we have enough data */
guint bytes_to_push;
bytes_to_push =
codec->output_buffer_size ? codec->output_buffer_size : bytes_ready;
do {
GST_DEBUG ("Pushing a buffer out (%ld bytes)", bytes_to_push);
outbuf = gst_adapter_take_buffer (codec->output_adapter, bytes_to_push);
/* Set buffer timestamp/duration if needed (and possible) */
if (!GST_BUFFER_TIMESTAMP_IS_VALID (outbuf) && codec->first_ts != -1) {
GST_DEBUG ("Computing output buffer timestamp");
GST_BUFFER_TIMESTAMP (outbuf) = codec->first_ts;
}
if (!GST_BUFFER_DURATION_IS_VALID (outbuf) && codec->state.frame_size) {
guint nsamples;
GST_DEBUG ("Computing output buffer duration");
nsamples = GST_BUFFER_SIZE (outbuf) / codec->state.frame_size;
GST_BUFFER_DURATION (outbuf) =
gst_util_uint64_scale_int (GST_SECOND, nsamples,
codec->state.rate);
}
if (codec->first_ts != -1) {
codec->first_ts += GST_BUFFER_DURATION (outbuf);
if (codec->first_ts > codec->last_ts) {
codec->last_ts = codec->first_ts;
}
}
GST_DEBUG ("out buffer timestamp %" GST_TIME_FORMAT " duration:%"
GST_TIME_FORMAT, GST_TIME_ARGS (GST_BUFFER_TIMESTAMP (outbuf)),
GST_TIME_ARGS (GST_BUFFER_DURATION (outbuf)));
/* Clip buffer */
if (codec->state.segment.format == GST_FORMAT_TIME ||
codec->state.segment.format == GST_FORMAT_DEFAULT) {
GST_DEBUG ("Clipping buffer");
outbuf = gst_audio_buffer_clip (outbuf, &codec->state.segment,
codec->state.rate, codec->state.frame_size);
}
/* Set DISCONT flag on the output buffer if needed */
if (G_LIKELY (outbuf)) {
if (G_UNLIKELY (codec->discont)) {
GST_BUFFER_FLAG_SET (outbuf, GST_BUFFER_FLAG_DISCONT);
codec->discont = FALSE;
GST_DEBUG ("Buffer is discont");
}
ret = gst_base_audio_decoder_push_buffer (codec, outbuf);
}
/* See if we can push another buffer */
bytes_ready = gst_adapter_available (codec->output_adapter);
GST_DEBUG ("%ld bytes left on the output", bytes_ready);
} while (ret == GST_FLOW_OK && bytes_ready >= bytes_to_push);
} else {
/* We need more data before we can push a buffer out */
GST_DEBUG ("Not pushing out, need more data");
ret = GST_FLOW_OK;
}
} else {
/* We got an error */
GST_DEBUG ("Got error while processing data");
}
GST_DEBUG ("chain-done");
return ret;
}
/* ----- Element public API ----- */
/**
* gst_base_audio_decoder_reset:
* @codec: The #GstBaseAudioDecoder instance.
*
* Resets the codec.
*
* This method will also invoke the subclass's reset virtual method
* if available. Niotice that reseting the codec will clear the
* input and output adapters.
*
* Returns: TRUE if the start operation was successful.
*/
gboolean
gst_base_audio_decoder_reset (GstBaseAudioDecoder * codec)
{
GstBaseAudioDecoderClass *codec_class;
GST_DEBUG ("gst_base_audio_decoder_reset");
codec_class = GST_BASE_AUDIO_DECODER_GET_CLASS (codec);
gst_adapter_clear (codec->input_adapter);
gst_adapter_clear (codec->output_adapter);
/* FIXME: is this needed? */
gst_segment_init (&codec->state.segment, GST_FORMAT_TIME);
codec->first_ts = -1;
codec->last_ts = -1;
if (codec_class->reset) {
codec_class->reset (codec);
}
return TRUE;
}
/**
* gst_base_audio_decoder_stop:
* @codec: The #GstBaseAudioDecoder instance.
*
* Stop the codec. Normally this will be used for closing resource.
*
* This method will also invoke the subclass's stop virtual method
* if available.
*
* Returns: TRUE if the start operation was successful.
*/
gboolean
gst_base_audio_decoder_stop (GstBaseAudioDecoder * codec)
{
GstBaseAudioDecoderClass *codec_class;
GST_DEBUG ("gst_base_audio_decoder_stop");
codec_class = GST_BASE_AUDIO_DECODER_GET_CLASS (codec);
gst_base_audio_decoder_reset (codec);
codec->bytes_in = 0;
codec->bytes_out = 0;
if (codec_class->stop) {
codec_class->stop (codec);
}
codec->started = FALSE;
return TRUE;
}
/**
* gst_base_audio_decoder_start:
* @codec: The #GstBaseAudioDecoder instance.
*
* Setup the codec so it can start processing data. Normally
* this will be used for opening resources needed for operation.
*
* This method will also invoke the subclass's start virtual method
* if available.
*
* Returns: TRUE if the start operation was successful.
*/
gboolean
gst_base_audio_decoder_start (GstBaseAudioDecoder * codec)
{
GstBaseAudioDecoderClass *codec_class;
GST_DEBUG ("gst_base_audio_decoder_start");
codec_class = GST_BASE_AUDIO_DECODER_GET_CLASS (codec);
gst_base_audio_decoder_reset (codec);
codec->bytes_in = 0;
codec->bytes_out = 0;
if (codec_class->start) {
codec_class->start (codec);
}
codec->started = TRUE;
return TRUE;
}
/**
* gst_base_audio_decoder_flush:
* @codec: The #GstBaseAudioDecoder instance.
*
* Flushes the input and output adapters. Subclasses should provide
* a flush_input implementation to allow flushing the input adapter.
* For the output adapter subclasses should provide a flush_output
* implementation. If no flush_output implementation is provided
* the output adapter will be flushed by pushing a single buffer
* containing all the data present in the output adapter.
*
* It is guaranteed that any data present in the adapters will be cleared
* after calling this method even if the operation flush
* operation was not successfull.
*
* Returns: TRUE if the flush operation was successful (any data present in
* the adapters was properly processed).
*/
gboolean
gst_base_audio_decoder_flush (GstBaseAudioDecoder * codec)
{
GstFlowReturn ret_i = GST_FLOW_OK;
GstFlowReturn ret_o = GST_FLOW_OK;
guint bytes;
GstBaseAudioDecoderClass *codec_class;
GST_DEBUG ("gst_base_audio_decoder_flush");
codec_class = GST_BASE_AUDIO_DECODER_GET_CLASS (codec);
/* Flush input adapter */
bytes = gst_adapter_available (codec->input_adapter);
if (bytes > 0) {
GST_DEBUG ("Flushing input adapter");
/* If the subclass provides a flush_input implementation, use that.
Otherwise we will clear the adapter and lose the data */
if (codec_class->flush_input) {
ret_i = codec_class->flush_input (codec);
if (ret_i != GST_FLOW_OK) {
GST_DEBUG ("failed to flush input");
}
} else {
GST_DEBUG ("Received EOS but cannot flush input, data will be lost");
ret_i = GST_FLOW_ERROR;
}
gst_adapter_clear (codec->input_adapter);
}
/* Flush output adapter */
bytes = gst_adapter_available (codec->output_adapter);
if (bytes > 0) {
/* If the subclass provides a flush_output implementation, use that.
Otherwise just push a single buffer with the adapter contents */
GST_DEBUG ("Flushing output adapter");
if (codec_class->flush_output) {
ret_o = codec_class->flush_output (codec);
if (ret_o != GST_FLOW_OK) {
GST_DEBUG ("failed to flush output (flush_output)");
}
} else {
GstBuffer *outbuf =
gst_adapter_take_buffer (codec->output_adapter, bytes);
ret_o = gst_base_audio_decoder_push_buffer (codec, outbuf);
gst_buffer_unref (outbuf);
if (ret_o != GST_FLOW_OK) {
GST_DEBUG ("Forced output flush failed");
}
}
gst_adapter_clear (codec->output_adapter);
}
return (ret_i == GST_FLOW_OK && ret_o == GST_FLOW_OK);
}
/**
* gst_base_audio_decoder_set_src_caps:
* @codec: #GstBaseAudioDecoder instance
* @caps: The caps to set on the source pad of @codec.
*
* Attempts to set @caps as the source caps of @codec. If the new caps
* are accepted on the source pad, this will issue a flush on the adapters
* to ensure that any data received with the old caps is processed first
* and a reset of the codec.
*
* Returns: TRUE if caps were set successfully.
*/
gboolean
gst_base_audio_decoder_set_src_caps (GstBaseAudioDecoder * codec,
GstCaps * caps)
{
gboolean ret;
GST_DEBUG ("gst_base_audio_decoder_set_src_caps %" GST_PTR_FORMAT, caps);
/* First, check if the pad accepts the new caps */
if (!gst_pad_accept_caps (codec->srcpad, caps)) {
GST_DEBUG ("pad does not accept new caps");
return FALSE;
}
/* If we have data in our adapters we should probably flush first */
gst_base_audio_decoder_flush (codec);
/* Set the caps on the pad */
ret = gst_pad_set_caps (codec->srcpad, caps);
/* And update the state of the codec from the caps */
if (ret) {
gst_base_audio_decoder_read_state_from_caps (codec, caps);
codec->caps_set = TRUE;
}
return ret;
}
/**
* gst_base_audio_decoder_push_buffer:
* @codec: #GstBaseAudioDecoder instance
* @buffer: a #GstBuffer.
*
* Pushes a buffer through the source pad.
*
* Returns: a #GstFlowReturn indicating the result of the push operation.
*/
GstFlowReturn
gst_base_audio_decoder_push_buffer (GstBaseAudioDecoder * codec,
GstBuffer * buffer)
{
codec->bytes_out += GST_BUFFER_SIZE (buffer);
return gst_pad_push (codec->srcpad, buffer);
}

220
gst-libs/gst/audio/gstbaseaudiodecoder.h Normal file
View file

@ -0,0 +1,220 @@
/* GStreamer
* Copyright (C) 2009 Igalia S.L.
* Author: Iago Toral Quiroga <itoral@igalia.com>
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Library General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Library General Public License for more details.
*
* You should have received a copy of the GNU Library General Public
* License along with this library; if not, write to the
* Free Software Foundation, Inc., 59 Temple Place - Suite 330,
* Boston, MA 02111-1307, USA.
*/
#ifndef _GST_BASE_AUDIO_DECODER_H_
#define _GST_BASE_AUDIO_DECODER_H_
#ifndef GST_USE_UNSTABLE_API
#warning "GstBaseAudioDecoder is unstable API and may change in future."
#warning "You can define GST_USE_UNSTABLE_API to avoid this warning."
#endif
#include <gst/gst.h>
#include <gst/base/gstadapter.h>
G_BEGIN_DECLS
#define GST_TYPE_BASE_AUDIO_DECODER \
(gst_base_audio_decoder_get_type())
#define GST_BASE_AUDIO_DECODER(obj) \
(G_TYPE_CHECK_INSTANCE_CAST((obj),GST_TYPE_BASE_AUDIO_DECODER,GstBaseAudioDecoder))
#define GST_BASE_AUDIO_DECODER_CLASS(klass) \
(G_TYPE_CHECK_CLASS_CAST((klass),GST_TYPE_BASE_AUDIO_DECODER,GstBaseAudioDecoderClass))
#define GST_BASE_AUDIO_DECODER_GET_CLASS(obj) \
(G_TYPE_INSTANCE_GET_CLASS((obj),GST_TYPE_BASE_AUDIO_DECODER,GstBaseAudioDecoderClass))
#define GST_IS_BASE_AUDIO_DECODER(obj) \
(G_TYPE_CHECK_INSTANCE_TYPE((obj),GST_TYPE_BASE_AUDIO_DECODER))
#define GST_IS_BASE_AUDIO_DECODER_CLASS(obj) \
(G_TYPE_CHECK_CLASS_TYPE((klass),GST_TYPE_BASE_AUDIO_DECODER))
/**
* GST_BASE_AUDIO_DECODER_SINK_NAME:
*
* The name of the templates for the sink pad.
*/
#define GST_BASE_AUDIO_DECODER_SINK_NAME "sink"
/**
* GST_BASE_AUDIO_DECODER_SRC_NAME:
*
* The name of the templates for the source pad.
*/
#define GST_BASE_AUDIO_DECODER_SRC_NAME "src"
/**
* GST_BASE_AUDIO_DECODER_SRC_PAD:
* @obj: base audio codec instance
*
* Gives the pointer to the source #GstPad object of the element.
*/
#define GST_BASE_AUDIO_DECODER_SRC_PAD(obj) (((GstBaseAudioDecoder *) (obj))->srcpad)
/**
* GST_BASE_AUDIO_DECODER_SINK_PAD:
* @obj: base audio codec instance
*
* Gives the pointer to the sink #GstPad object of the element.
*/
#define GST_BASE_AUDIO_DECODER_SINK_PAD(obj) (((GstBaseAudioDecoder *) (obj))->sinkpad)
/**
* GST_BASE_AUDIO_DECODER_INPUT_ADAPTER:
* @obj: base audio codec instance
*
* Gives the pointer to the input #GstAdapter object of the element.
*/
#define GST_BASE_AUDIO_DECODER_INPUT_ADAPTER(obj) (((GstBaseAudioDecoder *) (obj))->input_adapter)
/**
* GST_BASE_AUDIO_DECODER_OUTPUT_ADAPTER:
* @obj: base audio codec instance
*
* Gives the pointer to the output #GstAdapter object of the element.
*/
#define GST_BASE_AUDIO_DECODER_OUTPUT_ADAPTER(obj) (((GstBaseAudioDecoder *) (obj))->output_adapter)
typedef struct _GstBaseAudioDecoder GstBaseAudioDecoder;
typedef struct _GstBaseAudioDecoderClass GstBaseAudioDecoderClass;
typedef struct _GstAudioState GstAudioState;
struct _GstAudioState
{
gint channels;
gint rate;
gint bytes_per_sample;
gint sample_depth;
gint frame_size;
GstSegment segment;
};
/**
* GstBaseAudioDecoder:
* @element: the parent element.
* @caps_set: whether caps have been set on the codec's source pad.
* @sinkpad: the sink pad.
* @srcpad: the source pad.
* @input_adapter: the input adapter that will be filled with the input buffers.
* @output_adapter: the output adapter. Subclasses will read from the input
* adapter, process the data and fill the output adapter with the result.
* @input_buffer_size: The minimum amount of data that should be present on the
* input adapter for the codec to process it.
* @output_buffer_size: The minimum amount of data that should be present on the
* output adapter for the codec to push buffers out.
* @bytes_in: total bytes that have been received.
* @bytes_out: total bytes that have been pushed out.
* @discont: whether the next buffer to push represents a discontinuity in the
* stream.
* @state: Audio stream information. See #GstAudioState for details.
* @codec_data: The codec data.
* @started: Whether the codec has been started and is ready to process data
* or not.
* @first_ts: timestamp of the first buffer in the input adapter.
* @last_ts: timestamp of the last buffer in the input adapter.
*
* The opaque #GstBaseAudioDecoder data structure.
*/
struct _GstBaseAudioDecoder
{
GstElement element;
/*< private >*/
gboolean caps_set;
/*< protected >*/
GstPad *sinkpad;
GstPad *srcpad;
GstAdapter *input_adapter;
GstAdapter *output_adapter;
guint input_buffer_size;
guint output_buffer_size;
guint64 bytes_in;
guint64 bytes_out;
gboolean discont;
GstAudioState state;
GstBuffer *codec_data;
gboolean started;
guint64 first_ts;
guint64 last_ts;
};
/**
* GstBaseAudioDecoderClass:
* @parent_class: Element parent class
* @start: Start processing. Ideal for opening resources in the subclass
* @stop: Stop processing. Subclasses should use this to close resources.
* @reset: Resets the codec. Called on discontinuities, etc.
* @event: Override this to handle events arriving on the sink pad.
* @handle_discont: Override to be notified on discontinuities.
* @flush_input: Subclasses may implement this to flush the input adapter,
* processing any data present in it and filling the output adapter with the
* result. This could be necessary if it is possible for the codec to
* receive an end-of-stream event before all the data in the input
* adapter has been processed.
* @flush_output: Subclasses may implement this to flush the output adapter,
* pushing buffers out through the codec's source pad when the end-of-stream
* event is received and there is data waiting to be processed in the
* adapters.
* @process_data: Subclasses must implement this. They should read from the
* input adapter, encode/decode the data present in it and fill the
* output adapter with the result.
* @push_data: Normally, #GstBaseAudioDecoder will handle pushing buffers out.
* However, it is possible for developers to take control of when and how
* buffers are pushed out by overriding this method. If subclasses provide
* an implementation, #GstBaseAudioDecoder will not push any buffers,
* instead, whenever there is data on the output adapter, it will call this
* method on the subclass, which would be the sole responsible for
* pushing the buffers out when appropriate.
* @negotiate_src_caps: Subclasses can implement this method to provide
* appropriate caps to be set on the codec's source pad. If they don't
* provide this, they will be responsible for calling
* gst_base_audio_decoder_set_src_caps when appropriate.
*/
struct _GstBaseAudioDecoderClass
{
GstElementClass parent_class;
gboolean (*start) (GstBaseAudioDecoder *codec);
gboolean (*stop) (GstBaseAudioDecoder *codec);
gboolean (*reset) (GstBaseAudioDecoder *codec);
GstFlowReturn (*event) (GstBaseAudioDecoder *codec, GstEvent *event);
void (*handle_discont) (GstBaseAudioDecoder *codec, GstBuffer *buffer);
gboolean (*flush_input) (GstBaseAudioDecoder *codec);
gboolean (*flush_output) (GstBaseAudioDecoder *codec);
GstFlowReturn (*process_data) (GstBaseAudioDecoder *codec);
GstFlowReturn (*push_data) (GstBaseAudioDecoder *codec);
GstCaps * (*negotiate_src_caps) (GstBaseAudioDecoder *codec,
GstCaps *sink_caps);
};
GType gst_base_audio_decoder_get_type (void);
gboolean gst_base_audio_decoder_reset (GstBaseAudioDecoder *codec);
gboolean gst_base_audio_decoder_stop (GstBaseAudioDecoder *codec);
gboolean gst_base_audio_decoder_start (GstBaseAudioDecoder *codec);
gboolean gst_base_audio_decoder_flush (GstBaseAudioDecoder *codec);
gboolean gst_base_audio_decoder_set_src_caps (GstBaseAudioDecoder *codec,
GstCaps *caps);
GstFlowReturn gst_base_audio_decoder_push_buffer (GstBaseAudioDecoder *codec,
GstBuffer *buffer);
G_END_DECLS
#endif