gstreamer/gst-libs/gst/audio/gstaudiometa.c
Havard Graff 0f866832b1 audio: add GstAudioLevelMeta
Will be used to implement RTP extension https://tools.ietf.org/html/rfc6464

Co-authored-by: Guillaume Desmottes <guillaume.desmottes@collabora.com>
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/706>
2021-02-04 10:25:24 +01:00

622 lines
18 KiB
C

/* GStreamer
* Copyright (C) <2011> Wim Taymans <wim.taymans@gmail.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., 51 Franklin St, Fifth Floor,
* Boston, MA 02110-1301, USA.
*/
/**
* SECTION:gstaudiometa
* @title: GstAudio meta
* @short_description: Buffer metadata for audio downmix matrix handling
*
* #GstAudioDownmixMeta defines an audio downmix matrix to be send along with
* audio buffers. These functions in this module help to create and attach the
* meta as well as extracting it.
*/
#ifdef HAVE_CONFIG_H
#include "config.h"
#endif
#include <string.h>
#include "gstaudiometa.h"
static gboolean
gst_audio_downmix_meta_init (GstMeta * meta, gpointer params,
GstBuffer * buffer)
{
GstAudioDownmixMeta *dmeta = (GstAudioDownmixMeta *) meta;
dmeta->from_position = dmeta->to_position = NULL;
dmeta->from_channels = dmeta->to_channels = 0;
dmeta->matrix = NULL;
return TRUE;
}
static void
gst_audio_downmix_meta_free (GstMeta * meta, GstBuffer * buffer)
{
GstAudioDownmixMeta *dmeta = (GstAudioDownmixMeta *) meta;
g_free (dmeta->from_position);
if (dmeta->matrix) {
g_free (*dmeta->matrix);
g_free (dmeta->matrix);
}
}
static gboolean
gst_audio_downmix_meta_transform (GstBuffer * dest, GstMeta * meta,
GstBuffer * buffer, GQuark type, gpointer data)
{
GstAudioDownmixMeta *smeta, *dmeta;
smeta = (GstAudioDownmixMeta *) meta;
if (GST_META_TRANSFORM_IS_COPY (type)) {
dmeta = gst_buffer_add_audio_downmix_meta (dest, smeta->from_position,
smeta->from_channels, smeta->to_position, smeta->to_channels,
(const gfloat **) smeta->matrix);
if (!dmeta)
return FALSE;
} else {
/* return FALSE, if transform type is not supported */
return FALSE;
}
return TRUE;
}
/**
* gst_buffer_get_audio_downmix_meta_for_channels:
* @buffer: a #GstBuffer
* @to_position: (array length=to_channels): the channel positions of
* the destination
* @to_channels: The number of channels of the destination
*
* Find the #GstAudioDownmixMeta on @buffer for the given destination
* channel positions.
*
* Returns: (transfer none): the #GstAudioDownmixMeta on @buffer.
*/
GstAudioDownmixMeta *
gst_buffer_get_audio_downmix_meta_for_channels (GstBuffer * buffer,
const GstAudioChannelPosition * to_position, gint to_channels)
{
gpointer state = NULL;
GstMeta *meta;
const GstMetaInfo *info = GST_AUDIO_DOWNMIX_META_INFO;
while ((meta = gst_buffer_iterate_meta (buffer, &state))) {
if (meta->info->api == info->api) {
GstAudioDownmixMeta *ameta = (GstAudioDownmixMeta *) meta;
if (ameta->to_channels == to_channels &&
memcmp (ameta->to_position, to_position,
sizeof (GstAudioChannelPosition) * to_channels) == 0)
return ameta;
}
}
return NULL;
}
/**
* gst_buffer_add_audio_downmix_meta:
* @buffer: a #GstBuffer
* @from_position: (array length=from_channels): the channel positions
* of the source
* @from_channels: The number of channels of the source
* @to_position: (array length=to_channels): the channel positions of
* the destination
* @to_channels: The number of channels of the destination
* @matrix: The matrix coefficients.
*
* Attaches #GstAudioDownmixMeta metadata to @buffer with the given parameters.
*
* @matrix is an two-dimensional array of @to_channels times @from_channels
* coefficients, i.e. the i-th output channels is constructed by multiplicating
* the input channels with the coefficients in @matrix[i] and taking the sum
* of the results.
*
* Returns: (transfer none): the #GstAudioDownmixMeta on @buffer.
*/
GstAudioDownmixMeta *
gst_buffer_add_audio_downmix_meta (GstBuffer * buffer,
const GstAudioChannelPosition * from_position, gint from_channels,
const GstAudioChannelPosition * to_position, gint to_channels,
const gfloat ** matrix)
{
GstAudioDownmixMeta *meta;
gint i;
g_return_val_if_fail (from_position != NULL, NULL);
g_return_val_if_fail (from_channels > 0, NULL);
g_return_val_if_fail (to_position != NULL, NULL);
g_return_val_if_fail (to_channels > 0, NULL);
g_return_val_if_fail (matrix != NULL, NULL);
meta =
(GstAudioDownmixMeta *) gst_buffer_add_meta (buffer,
GST_AUDIO_DOWNMIX_META_INFO, NULL);
meta->from_channels = from_channels;
meta->to_channels = to_channels;
meta->from_position =
g_new (GstAudioChannelPosition, meta->from_channels + meta->to_channels);
meta->to_position = meta->from_position + meta->from_channels;
memcpy (meta->from_position, from_position,
sizeof (GstAudioChannelPosition) * meta->from_channels);
memcpy (meta->to_position, to_position,
sizeof (GstAudioChannelPosition) * meta->to_channels);
meta->matrix = g_new (gfloat *, meta->to_channels);
meta->matrix[0] = g_new (gfloat, meta->from_channels * meta->to_channels);
memcpy (meta->matrix[0], matrix[0], sizeof (gfloat) * meta->from_channels);
for (i = 1; i < meta->to_channels; i++) {
meta->matrix[i] = meta->matrix[0] + i * meta->from_channels;
memcpy (meta->matrix[i], matrix[i], sizeof (gfloat) * meta->from_channels);
}
return meta;
}
GType
gst_audio_downmix_meta_api_get_type (void)
{
static volatile GType type;
static const gchar *tags[] =
{ GST_META_TAG_AUDIO_STR, GST_META_TAG_AUDIO_CHANNELS_STR, NULL };
if (g_once_init_enter (&type)) {
GType _type = gst_meta_api_type_register ("GstAudioDownmixMetaAPI", tags);
g_once_init_leave (&type, _type);
}
return type;
}
const GstMetaInfo *
gst_audio_downmix_meta_get_info (void)
{
static const GstMetaInfo *audio_downmix_meta_info = NULL;
if (g_once_init_enter ((GstMetaInfo **) & audio_downmix_meta_info)) {
const GstMetaInfo *meta =
gst_meta_register (GST_AUDIO_DOWNMIX_META_API_TYPE,
"GstAudioDownmixMeta", sizeof (GstAudioDownmixMeta),
gst_audio_downmix_meta_init, gst_audio_downmix_meta_free,
gst_audio_downmix_meta_transform);
g_once_init_leave ((GstMetaInfo **) & audio_downmix_meta_info,
(GstMetaInfo *) meta);
}
return audio_downmix_meta_info;
}
static gboolean
gst_audio_clipping_meta_init (GstMeta * meta, gpointer params,
GstBuffer * buffer)
{
GstAudioClippingMeta *cmeta = (GstAudioClippingMeta *) meta;
cmeta->format = GST_FORMAT_UNDEFINED;
cmeta->start = cmeta->end = 0;
return TRUE;
}
static gboolean
gst_audio_clipping_meta_transform (GstBuffer * dest, GstMeta * meta,
GstBuffer * buffer, GQuark type, gpointer data)
{
GstAudioClippingMeta *smeta, *dmeta;
smeta = (GstAudioClippingMeta *) meta;
if (GST_META_TRANSFORM_IS_COPY (type)) {
GstMetaTransformCopy *copy = data;
if (copy->region)
return FALSE;
dmeta =
gst_buffer_add_audio_clipping_meta (dest, smeta->format, smeta->start,
smeta->end);
if (!dmeta)
return FALSE;
} else {
/* TODO: Could implement an automatic transform for resampling */
/* return FALSE, if transform type is not supported */
return FALSE;
}
return TRUE;
}
/**
* gst_buffer_add_audio_clipping_meta:
* @buffer: a #GstBuffer
* @format: GstFormat of @start and @stop, GST_FORMAT_DEFAULT is samples
* @start: Amount of audio to clip from start of buffer
* @end: Amount of to clip from end of buffer
*
* Attaches #GstAudioClippingMeta metadata to @buffer with the given parameters.
*
* Returns: (transfer none): the #GstAudioClippingMeta on @buffer.
*
* Since: 1.8
*/
GstAudioClippingMeta *
gst_buffer_add_audio_clipping_meta (GstBuffer * buffer,
GstFormat format, guint64 start, guint64 end)
{
GstAudioClippingMeta *meta;
g_return_val_if_fail (format != GST_FORMAT_UNDEFINED, NULL);
meta =
(GstAudioClippingMeta *) gst_buffer_add_meta (buffer,
GST_AUDIO_CLIPPING_META_INFO, NULL);
meta->format = format;
meta->start = start;
meta->end = end;
return meta;
}
GType
gst_audio_clipping_meta_api_get_type (void)
{
static volatile GType type;
static const gchar *tags[] =
{ GST_META_TAG_AUDIO_STR, GST_META_TAG_AUDIO_RATE_STR, NULL };
if (g_once_init_enter (&type)) {
GType _type = gst_meta_api_type_register ("GstAudioClippingMetaAPI", tags);
g_once_init_leave (&type, _type);
}
return type;
}
const GstMetaInfo *
gst_audio_clipping_meta_get_info (void)
{
static const GstMetaInfo *audio_clipping_meta_info = NULL;
if (g_once_init_enter ((GstMetaInfo **) & audio_clipping_meta_info)) {
const GstMetaInfo *meta =
gst_meta_register (GST_AUDIO_CLIPPING_META_API_TYPE,
"GstAudioClippingMeta", sizeof (GstAudioClippingMeta),
gst_audio_clipping_meta_init, NULL,
gst_audio_clipping_meta_transform);
g_once_init_leave ((GstMetaInfo **) & audio_clipping_meta_info,
(GstMetaInfo *) meta);
}
return audio_clipping_meta_info;
}
static gboolean
gst_audio_meta_init (GstMeta * meta, gpointer params, GstBuffer * buffer)
{
GstAudioMeta *ameta = (GstAudioMeta *) meta;
gst_audio_info_init (&ameta->info);
ameta->samples = 0;
ameta->offsets = NULL;
return TRUE;
}
static void
gst_audio_meta_free (GstMeta * meta, GstBuffer * buffer)
{
GstAudioMeta *ameta = (GstAudioMeta *) meta;
if (ameta->offsets && ameta->offsets != ameta->priv_offsets_arr)
g_slice_free1 (ameta->info.channels * sizeof (gsize), ameta->offsets);
}
static gboolean
gst_audio_meta_transform (GstBuffer * dest, GstMeta * meta,
GstBuffer * buffer, GQuark type, gpointer data)
{
GstAudioMeta *smeta, *dmeta;
smeta = (GstAudioMeta *) meta;
if (GST_META_TRANSFORM_IS_COPY (type)) {
dmeta = gst_buffer_add_audio_meta (dest, &smeta->info, smeta->samples,
smeta->offsets);
if (!dmeta)
return FALSE;
} else {
/* return FALSE, if transform type is not supported */
return FALSE;
}
return TRUE;
}
/**
* gst_buffer_add_audio_meta:
* @buffer: a #GstBuffer
* @info: the audio properties of the buffer
* @samples: the number of valid samples in the buffer
* @offsets: (nullable): the offsets (in bytes) where each channel plane starts
* in the buffer or %NULL to calculate it (see below); must be %NULL also
* when @info->layout is %GST_AUDIO_LAYOUT_INTERLEAVED
*
* Allocates and attaches a #GstAudioMeta on @buffer, which must be writable
* for that purpose. The fields of the #GstAudioMeta are directly populated
* from the arguments of this function.
*
* When @info->layout is %GST_AUDIO_LAYOUT_NON_INTERLEAVED and @offsets is
* %NULL, the offsets are calculated with a formula that assumes the planes are
* tightly packed and in sequence:
* offsets[channel] = channel * @samples * sample_stride
*
* It is not allowed for channels to overlap in memory,
* i.e. for each i in [0, channels), the range
* [@offsets[i], @offsets[i] + @samples * sample_stride) must not overlap
* with any other such range. This function will assert if the parameters
* specified cause this restriction to be violated.
*
* It is, obviously, also not allowed to specify parameters that would cause
* out-of-bounds memory access on @buffer. This is also checked, which means
* that you must add enough memory on the @buffer before adding this meta.
*
* Returns: (transfer none): the #GstAudioMeta that was attached on the @buffer
*
* Since: 1.16
*/
GstAudioMeta *
gst_buffer_add_audio_meta (GstBuffer * buffer, const GstAudioInfo * info,
gsize samples, gsize offsets[])
{
GstAudioMeta *meta;
gint i;
gsize plane_size;
g_return_val_if_fail (GST_IS_BUFFER (buffer), FALSE);
g_return_val_if_fail (info != NULL, NULL);
g_return_val_if_fail (GST_AUDIO_INFO_IS_VALID (info), NULL);
g_return_val_if_fail (GST_AUDIO_INFO_FORMAT (info) !=
GST_AUDIO_FORMAT_UNKNOWN, NULL);
g_return_val_if_fail (info->layout == GST_AUDIO_LAYOUT_NON_INTERLEAVED
|| !offsets, NULL);
meta =
(GstAudioMeta *) gst_buffer_add_meta (buffer, GST_AUDIO_META_INFO, NULL);
meta->info = *info;
meta->samples = samples;
plane_size = samples * info->finfo->width / 8;
if (info->layout == GST_AUDIO_LAYOUT_NON_INTERLEAVED) {
#ifndef G_DISABLE_CHECKS
gsize max_offset = 0;
gint j;
#endif
if (G_UNLIKELY (info->channels > 8))
meta->offsets = g_slice_alloc (info->channels * sizeof (gsize));
else
meta->offsets = meta->priv_offsets_arr;
if (offsets) {
for (i = 0; i < info->channels; i++) {
meta->offsets[i] = offsets[i];
#ifndef G_DISABLE_CHECKS
max_offset = MAX (max_offset, offsets[i]);
for (j = 0; j < info->channels; j++) {
if (i != j && !(offsets[j] + plane_size <= offsets[i]
|| offsets[i] + plane_size <= offsets[j])) {
g_critical ("GstAudioMeta properties would cause channel memory "
"areas to overlap! offsets: %" G_GSIZE_FORMAT " (%d), %"
G_GSIZE_FORMAT " (%d) with plane size %" G_GSIZE_FORMAT,
offsets[i], i, offsets[j], j, plane_size);
gst_buffer_remove_meta (buffer, (GstMeta *) meta);
return NULL;
}
}
#endif
}
} else {
/* default offsets assume channels are laid out sequentially in memory */
for (i = 0; i < info->channels; i++)
meta->offsets[i] = i * plane_size;
#ifndef G_DISABLE_CHECKS
max_offset = meta->offsets[info->channels - 1];
#endif
}
#ifndef G_DISABLE_CHECKS
if (max_offset + plane_size > gst_buffer_get_size (buffer)) {
g_critical ("GstAudioMeta properties would cause "
"out-of-bounds memory access on the buffer: max_offset %"
G_GSIZE_FORMAT ", samples %" G_GSIZE_FORMAT ", bps %u, buffer size %"
G_GSIZE_FORMAT, max_offset, samples, info->finfo->width / 8,
gst_buffer_get_size (buffer));
gst_buffer_remove_meta (buffer, (GstMeta *) meta);
return NULL;
}
#endif
}
return meta;
}
GType
gst_audio_meta_api_get_type (void)
{
static volatile GType type;
static const gchar *tags[] = {
GST_META_TAG_AUDIO_STR, GST_META_TAG_AUDIO_CHANNELS_STR,
GST_META_TAG_AUDIO_RATE_STR, NULL
};
if (g_once_init_enter (&type)) {
GType _type = gst_meta_api_type_register ("GstAudioMetaAPI", tags);
g_once_init_leave (&type, _type);
}
return type;
}
const GstMetaInfo *
gst_audio_meta_get_info (void)
{
static const GstMetaInfo *audio_meta_info = NULL;
if (g_once_init_enter ((GstMetaInfo **) & audio_meta_info)) {
const GstMetaInfo *meta = gst_meta_register (GST_AUDIO_META_API_TYPE,
"GstAudioMeta", sizeof (GstAudioMeta),
gst_audio_meta_init,
gst_audio_meta_free,
gst_audio_meta_transform);
g_once_init_leave ((GstMetaInfo **) & audio_meta_info,
(GstMetaInfo *) meta);
}
return audio_meta_info;
}
/**
* gst_audio_level_meta_api_get_type:
*
* Return the #GType associated with #GstAudioLevelMeta.
*
* Returns: a #GType
*
* Since: 1.20
*/
GType
gst_audio_level_meta_api_get_type (void)
{
static volatile GType type = 0;
static const gchar *tags[] = { NULL };
if (g_once_init_enter (&type)) {
GType _type = gst_meta_api_type_register ("GstAudioLevelMetaAPI", tags);
g_once_init_leave (&type, _type);
}
return type;
}
static gboolean
gst_audio_level_meta_init (GstMeta * meta, gpointer params, GstBuffer * buffer)
{
GstAudioLevelMeta *dmeta = (GstAudioLevelMeta *) meta;
dmeta->level = 127;
dmeta->voice_activity = FALSE;
return TRUE;
}
static gboolean
gst_audio_level_meta_transform (GstBuffer * dst, GstMeta * meta,
GstBuffer * src, GQuark type, gpointer data)
{
if (GST_META_TRANSFORM_IS_COPY (type)) {
GstAudioLevelMeta *smeta = (GstAudioLevelMeta *) meta;
GstAudioLevelMeta *dmeta;
dmeta = gst_buffer_add_audio_level_meta (dst, smeta->level,
smeta->voice_activity);
if (dmeta == NULL)
return FALSE;
} else {
/* return FALSE, if transform type is not supported */
return FALSE;
}
return TRUE;
}
/**
* gst_audio_level_meta_get_info:
*
* Return the #GstMetaInfo associated with #GstAudioLevelMeta.
*
* Returns: (transfer none): a #GstMetaInfo
*
* Since: 1.20
*/
const GstMetaInfo *
gst_audio_level_meta_get_info (void)
{
static const GstMetaInfo *audio_level_meta_info = NULL;
if (g_once_init_enter (&audio_level_meta_info)) {
const GstMetaInfo *meta = gst_meta_register (GST_AUDIO_LEVEL_META_API_TYPE,
"GstAudioLevelMeta",
sizeof (GstAudioLevelMeta),
gst_audio_level_meta_init,
(GstMetaFreeFunction) NULL,
gst_audio_level_meta_transform);
g_once_init_leave (&audio_level_meta_info, meta);
}
return audio_level_meta_info;
}
/**
* gst_buffer_add_audio_level_meta:
* @buffer: a #GstBuffer
* @level: the -dBov from 0-127 (127 is silence).
* @voice_activity: whether the buffer contains voice activity.
*
* Attaches audio level information to @buffer. (RFC 6464)
*
* Returns: (transfer none) (nullable): the #GstAudioLevelMeta on @buffer.
*
* Since: 1.20
*/
GstAudioLevelMeta *
gst_buffer_add_audio_level_meta (GstBuffer * buffer, guint8 level,
gboolean voice_activity)
{
GstAudioLevelMeta *meta;
g_return_val_if_fail (buffer != NULL, NULL);
meta = (GstAudioLevelMeta *) gst_buffer_add_meta (buffer,
GST_AUDIO_LEVEL_META_INFO, NULL);
if (!meta)
return NULL;
meta->level = level;
meta->voice_activity = voice_activity;
return meta;
}
/**
* gst_buffer_get_audio_level_meta:
* @buffer: a #GstBuffer
*
* Find the #GstAudioLevelMeta on @buffer.
*
* Returns: (transfer none) (nullable): the #GstAudioLevelMeta or %NULL when
* there is no such metadata on @buffer.
*
* Since: 1.20
*/
GstAudioLevelMeta *
gst_buffer_get_audio_level_meta (GstBuffer * buffer)
{
return (GstAudioLevelMeta *) gst_buffer_get_meta (buffer,
gst_audio_level_meta_api_get_type ());
}