blob: 353942bdc8d20db4ded2e1ca3cac4f870b095771 [file] [log] [blame]
* Copyright (c) 2021 The WebRTC project authors. All Rights Reserved.
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
#include <cstdint>
#include <memory>
#include "absl/container/inlined_vector.h"
#include "api/video/encoded_frame.h"
#include "modules/rtp_rtcp/source/rtp_packet_received.h"
namespace webrtc {
// The RtpVideoFrameAssembler takes RtpPacketReceived and assembles them into
// complete frames. A frame is considered complete when all packets of the frame
// has been received, the bitstream data has successfully extracted, an ID has
// been assigned, and all dependencies are known. Frame IDs are strictly
// monotonic in decode order, dependencies are expressed as frame IDs.
class RtpVideoFrameAssembler {
// FrameVector is just a vector-like type of std::unique_ptr<EncodedFrame>.
// The vector type may change without notice.
using FrameVector = absl::InlinedVector<std::unique_ptr<EncodedFrame>, 3>;
enum PayloadFormat { kRaw, kH264, kVp8, kVp9, kAv1, kGeneric };
explicit RtpVideoFrameAssembler(PayloadFormat payload_format);
RtpVideoFrameAssembler(const RtpVideoFrameAssembler& other) = delete;
RtpVideoFrameAssembler& operator=(const RtpVideoFrameAssembler& other) =
// Typically when a packet is inserted zero or one frame is completed. In the
// case of RTP packets being inserted out of order then sometime multiple
// frames could be completed from a single packet, hence the 'FrameVector'
// return type.
FrameVector InsertPacket(const RtpPacketReceived& packet);
class Impl;
std::unique_ptr<Impl> impl_;
} // namespace webrtc