| Index: modules/include/module_common_types.h
|
| diff --git a/modules/include/module_common_types.h b/modules/include/module_common_types.h
|
| index bc5c347a4a93baadca7a294b8bf4b18d60e6697b..c3ad9935554295c3713abb087368ffc23e8a39a5 100644
|
| --- a/modules/include/module_common_types.h
|
| +++ b/modules/include/module_common_types.h
|
| @@ -330,7 +330,18 @@ class AudioFrame {
|
| // ResetWithoutMuting() to skip this wasteful zeroing.
|
| void ResetWithoutMuting();
|
|
|
| - void UpdateFrame(int id, uint32_t timestamp, const int16_t* data,
|
| + // TODO(solenberg): Remove once downstream users of AudioFrame have updated.
|
| + RTC_DEPRECATED
|
| + void UpdateFrame(int id, uint32_t timestamp, const int16_t* data,
|
| + size_t samples_per_channel, int sample_rate_hz,
|
| + SpeechType speech_type, VADActivity vad_activity,
|
| + size_t num_channels = 1) {
|
| + RTC_UNUSED(id);
|
| + UpdateFrame(timestamp, data, samples_per_channel, sample_rate_hz,
|
| + speech_type, vad_activity, num_channels);
|
| + }
|
| +
|
| + void UpdateFrame(uint32_t timestamp, const int16_t* data,
|
| size_t samples_per_channel, int sample_rate_hz,
|
| SpeechType speech_type, VADActivity vad_activity,
|
| size_t num_channels = 1);
|
| @@ -366,7 +377,6 @@ class AudioFrame {
|
| RTC_DEPRECATED AudioFrame& operator>>=(const int rhs);
|
| RTC_DEPRECATED AudioFrame& operator+=(const AudioFrame& rhs);
|
|
|
| - int id_;
|
| // RTP timestamp of the first sample in the AudioFrame.
|
| uint32_t timestamp_ = 0;
|
| // Time since the first frame in milliseconds.
|
| @@ -414,7 +424,6 @@ inline void AudioFrame::Reset() {
|
| }
|
|
|
| inline void AudioFrame::ResetWithoutMuting() {
|
| - id_ = -1;
|
| // TODO(wu): Zero is a valid value for |timestamp_|. We should initialize
|
| // to an invalid value, or add a new member to indicate invalidity.
|
| timestamp_ = 0;
|
| @@ -428,15 +437,13 @@ inline void AudioFrame::ResetWithoutMuting() {
|
| profile_timestamp_ms_ = 0;
|
| }
|
|
|
| -inline void AudioFrame::UpdateFrame(int id,
|
| - uint32_t timestamp,
|
| +inline void AudioFrame::UpdateFrame(uint32_t timestamp,
|
| const int16_t* data,
|
| size_t samples_per_channel,
|
| int sample_rate_hz,
|
| SpeechType speech_type,
|
| VADActivity vad_activity,
|
| size_t num_channels) {
|
| - id_ = id;
|
| timestamp_ = timestamp;
|
| samples_per_channel_ = samples_per_channel;
|
| sample_rate_hz_ = sample_rate_hz;
|
| @@ -457,7 +464,6 @@ inline void AudioFrame::UpdateFrame(int id,
|
| inline void AudioFrame::CopyFrom(const AudioFrame& src) {
|
| if (this == &src) return;
|
|
|
| - id_ = src.id_;
|
| timestamp_ = src.timestamp_;
|
| elapsed_time_ms_ = src.elapsed_time_ms_;
|
| ntp_time_ms_ = src.ntp_time_ms_;
|
|
|