From 41c44cde417e82efb1fd9162e11c068d643537af Mon Sep 17 00:00:00 2001
From: Jianjun Zhu <jianjun.zhu@intel.com>
Date: Tue, 20 Feb 2024 10:05:34 +0800
Subject: [PATCH] Add some comments for H265 RTP depacketizer.

This CL helps readers to understand which part of the spec
VideoRtpDepacketizerH265 implements.

Bug: webrtc:13485
Change-Id: Ie78a6ce781e6af559d59b1b07ce2854115368a86
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/340008
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#41768}
---
 .../rtp_rtcp/source/video_rtp_depacketizer_h265.cc | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/modules/rtp_rtcp/source/video_rtp_depacketizer_h265.cc b/modules/rtp_rtcp/source/video_rtp_depacketizer_h265.cc
index d2646e82da..cea55078e5 100644
--- a/modules/rtp_rtcp/source/video_rtp_depacketizer_h265.cc
+++ b/modules/rtp_rtcp/source/video_rtp_depacketizer_h265.cc
@@ -29,6 +29,8 @@
 #include "rtc_base/checks.h"
 #include "rtc_base/logging.h"
 
+// RTP Payload Format for HEVC: https://datatracker.ietf.org/doc/html/rfc7798
+
 namespace webrtc {
 namespace {
 
@@ -55,6 +57,10 @@ bool ParseApStartOffsets(const uint8_t* nalu_ptr,
   return true;
 }
 
+// Single NALU packet structure
+// https://datatracker.ietf.org/doc/html/rfc7798#section-4.4.1
+// Aggregation Packet (AP) strcture
+// https://datatracker.ietf.org/doc/html/rfc7798#section-4.4.2
 absl::optional<VideoRtpDepacketizer::ParsedRtpPayload> ProcessApOrSingleNalu(
     rtc::CopyOnWriteBuffer rtp_payload) {
   // Skip the single NALU header (payload header), aggregated packet case will
@@ -122,6 +128,10 @@ absl::optional<VideoRtpDepacketizer::ParsedRtpPayload> ProcessApOrSingleNalu(
       case H265::NaluType::kIdrNLp:
       case H265::NaluType::kCra:
       case H265::NaluType::kRsvIrapVcl23:
+        // Mark IRAP(Intra Random Access Point) frames as key frames. Their NALU
+        // types are in the range of BLA_W_LP (16) to RSV_IRAP_VCL23 (23),
+        // inclusive.
+        // https://datatracker.ietf.org/doc/html/rfc7798#section-3.1.1
         parsed_payload->video_header.frame_type =
             VideoFrameType::kVideoFrameKey;
         ABSL_FALLTHROUGH_INTENDED;
@@ -170,6 +180,8 @@ absl::optional<VideoRtpDepacketizer::ParsedRtpPayload> ProcessApOrSingleNalu(
   return parsed_payload;
 }
 
+// Fragmentation Unit (FU) structure:
+// https://datatracker.ietf.org/doc/html/rfc7798#section-4.4.3
 absl::optional<VideoRtpDepacketizer::ParsedRtpPayload> ParseFuNalu(
     rtc::CopyOnWriteBuffer rtp_payload) {
   if (rtp_payload.size() < kH265FuHeaderSizeBytes + kH265NalHeaderSizeBytes) {
@@ -204,6 +216,8 @@ absl::optional<VideoRtpDepacketizer::ParsedRtpPayload> ParseFuNalu(
 
   if (original_nal_type >= H265::NaluType::kBlaWLp &&
       original_nal_type <= H265::NaluType::kRsvIrapVcl23) {
+    // IRAP picture.
+    // https://datatracker.ietf.org/doc/html/rfc7798#section-3.1.1
     parsed_payload->video_header.frame_type = VideoFrameType::kVideoFrameKey;
   } else {
     parsed_payload->video_header.frame_type = VideoFrameType::kVideoFrameDelta;