NetEqImpl::GetDecoderFormat: Return RTP clockrate, not codec sample rate

Well, in fact we need to return both. But return codec sample rate
separately and let the SdpAudioFormat contain the RTP clockrate,
otherwise we're essentially lying to our callers.

Bug: webrtc:11028
Change-Id: I40f36cb9db6b9824404ade6b0515a8312ff97009
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/156307
Commit-Queue: Karl Wiberg <kwiberg@webrtc.org>
Reviewed-by: Ivo Creusen <ivoc@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29444}
diff --git a/audio/channel_receive.cc b/audio/channel_receive.cc
index 5bb568e..486dcb1 100644
--- a/audio/channel_receive.cc
+++ b/audio/channel_receive.cc
@@ -888,6 +888,9 @@
   // TODO(ossu): Zero clockrate can only happen if we've added an external
   // decoder for a format we don't support internally. Remove once that way of
   // adding decoders is gone!
+  // TODO(kwiberg): `decoder->second.clockrate_hz` is an RTP clockrate as it
+  // should, but `acm_receiver_.last_output_sample_rate_hz()` is a codec sample
+  // rate, which is not always the same thing.
   return (decoder && decoder->second.clockrate_hz != 0)
              ? decoder->second.clockrate_hz
              : acm_receiver_.last_output_sample_rate_hz();
diff --git a/modules/audio_coding/acm2/acm_receiver.cc b/modules/audio_coding/acm2/acm_receiver.cc
index 1c8d88d..4019615 100644
--- a/modules/audio_coding/acm2/acm_receiver.cc
+++ b/modules/audio_coding/acm2/acm_receiver.cc
@@ -73,7 +73,7 @@
   if (!last_decoder_) {
     return absl::nullopt;
   }
-  return last_decoder_->second.clockrate_hz;
+  return last_decoder_->sample_rate_hz;
 }
 
 int AcmReceiver::last_output_sample_rate_hz() const {
@@ -89,7 +89,7 @@
 
   int payload_type = rtp_header.payloadType;
   auto format = neteq_->GetDecoderFormat(payload_type);
-  if (format && absl::EqualsIgnoreCase(format->name, "red")) {
+  if (format && absl::EqualsIgnoreCase(format->sdp_format.name, "red")) {
     // This is a RED packet. Get the format of the audio codec.
     payload_type = incoming_payload[0] & 0x7f;
     format = neteq_->GetDecoderFormat(payload_type);
@@ -102,15 +102,17 @@
 
   {
     rtc::CritScope lock(&crit_sect_);
-    if (absl::EqualsIgnoreCase(format->name, "cn")) {
-      if (last_decoder_ && last_decoder_->second.num_channels > 1) {
+    if (absl::EqualsIgnoreCase(format->sdp_format.name, "cn")) {
+      if (last_decoder_ && last_decoder_->num_channels > 1) {
         // This is a CNG and the audio codec is not mono, so skip pushing in
         // packets into NetEq.
         return 0;
       }
     } else {
-      RTC_DCHECK(format);
-      last_decoder_ = std::make_pair(payload_type, *format);
+      last_decoder_ = DecoderInfo{/*payload_type=*/payload_type,
+                                  /*sample_rate_hz=*/format->sample_rate_hz,
+                                  /*num_channels=*/format->num_channels,
+                                  /*sdp_format=*/std::move(format->sdp_format)};
     }
   }  // |crit_sect_| is released.
 
@@ -221,8 +223,8 @@
   if (!last_decoder_) {
     return absl::nullopt;
   }
-  RTC_DCHECK_NE(-1, last_decoder_->first);  // Payload type should be valid.
-  return last_decoder_;
+  RTC_DCHECK_NE(-1, last_decoder_->payload_type);
+  return std::make_pair(last_decoder_->payload_type, last_decoder_->sdp_format);
 }
 
 void AcmReceiver::GetNetworkStatistics(NetworkStatistics* acm_stat) const {
diff --git a/modules/audio_coding/acm2/acm_receiver.h b/modules/audio_coding/acm2/acm_receiver.h
index f07f8a9..1512656 100644
--- a/modules/audio_coding/acm2/acm_receiver.h
+++ b/modules/audio_coding/acm2/acm_receiver.h
@@ -203,11 +203,17 @@
   void GetDecodingCallStatistics(AudioDecodingCallStats* stats) const;
 
  private:
+  struct DecoderInfo {
+    int payload_type;
+    int sample_rate_hz;
+    int num_channels;
+    SdpAudioFormat sdp_format;
+  };
+
   uint32_t NowInTimestamp(int decoder_sampling_rate) const;
 
   rtc::CriticalSection crit_sect_;
-  absl::optional<std::pair<int, SdpAudioFormat>> last_decoder_
-      RTC_GUARDED_BY(crit_sect_);
+  absl::optional<DecoderInfo> last_decoder_ RTC_GUARDED_BY(crit_sect_);
   ACMResampler resampler_ RTC_GUARDED_BY(crit_sect_);
   std::unique_ptr<int16_t[]> last_audio_buffer_ RTC_GUARDED_BY(crit_sect_);
   CallStatistics call_stats_ RTC_GUARDED_BY(crit_sect_);
diff --git a/modules/audio_coding/neteq/include/neteq.h b/modules/audio_coding/neteq/include/neteq.h
index c6af751..b53b5ad 100644
--- a/modules/audio_coding/neteq/include/neteq.h
+++ b/modules/audio_coding/neteq/include/neteq.h
@@ -143,6 +143,13 @@
 
   enum ReturnCodes { kOK = 0, kFail = -1 };
 
+  // Return type for GetDecoderFormat.
+  struct DecoderFormat {
+    int sample_rate_hz;
+    int num_channels;
+    SdpAudioFormat sdp_format;
+  };
+
   // Creates a new NetEq object, with parameters set in |config|. The |config|
   // object will only have to be valid for the duration of the call to this
   // method.
@@ -265,7 +272,7 @@
 
   // Returns the decoder info for the given payload type. Returns empty if no
   // such payload type was registered.
-  virtual absl::optional<SdpAudioFormat> GetDecoderFormat(
+  virtual absl::optional<DecoderFormat> GetDecoderFormat(
       int payload_type) const = 0;
 
   // Flushes both the packet buffer and the sync buffer.
diff --git a/modules/audio_coding/neteq/neteq_impl.cc b/modules/audio_coding/neteq/neteq_impl.cc
index 751fc45..37036e3 100644
--- a/modules/audio_coding/neteq/neteq_impl.cc
+++ b/modules/audio_coding/neteq/neteq_impl.cc
@@ -392,21 +392,23 @@
   return last_output_sample_rate_hz_;
 }
 
-absl::optional<SdpAudioFormat> NetEqImpl::GetDecoderFormat(
+absl::optional<NetEq::DecoderFormat> NetEqImpl::GetDecoderFormat(
     int payload_type) const {
   rtc::CritScope lock(&crit_sect_);
   const DecoderDatabase::DecoderInfo* const di =
       decoder_database_->GetDecoderInfo(payload_type);
-  if (!di) {
-    return absl::nullopt;  // Payload type not registered.
+  if (di) {
+    const AudioDecoder* const decoder = di->GetDecoder();
+    // TODO(kwiberg): Why the special case for RED?
+    return DecoderFormat{
+        /*sample_rate_hz=*/di->IsRed() ? 8000 : di->SampleRateHz(),
+        /*num_channels=*/
+        decoder ? rtc::dchecked_cast<int>(decoder->Channels()) : 1,
+        /*sdp_format=*/di->GetFormat()};
+  } else {
+    // Payload type not registered.
+    return absl::nullopt;
   }
-
-  SdpAudioFormat format = di->GetFormat();
-  // TODO(solenberg): This is legacy but messed up - mixing RTP rate and SR.
-  format.clockrate_hz = di->IsRed() ? 8000 : di->SampleRateHz();
-  const AudioDecoder* const decoder = di->GetDecoder();
-  format.num_channels = decoder ? decoder->Channels() : 1;
-  return format;
 }
 
 void NetEqImpl::FlushBuffers() {
diff --git a/modules/audio_coding/neteq/neteq_impl.h b/modules/audio_coding/neteq/neteq_impl.h
index 8ecb9b6..842869f 100644
--- a/modules/audio_coding/neteq/neteq_impl.h
+++ b/modules/audio_coding/neteq/neteq_impl.h
@@ -182,7 +182,7 @@
 
   int last_output_sample_rate_hz() const override;
 
-  absl::optional<SdpAudioFormat> GetDecoderFormat(
+  absl::optional<DecoderFormat> GetDecoderFormat(
       int payload_type) const override;
 
   // Flushes both the packet buffer and the sync buffer.