dragonpilot - 基於 openpilot 的開源駕駛輔助系統
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

242 lines
7.9 KiB

webrtcd: webrtc streaming server (audio/video/cereal) (#30186) * WebRTCClient and WebRTCServer abstractions * webrtc client implementation * Interactive test scripts * Send localDescriptions as offer/asnwer, as they are different * Tracks need to be added after setting remote description for multi-cam streaming to work * Remove WebRTCStreamingMetadata * Wait for tracks * Move stuff to separate files, rename some things * Refactor everything, create WebRTCStreamBuilder for both offer and answers * ta flight done time to grind * wait for incoming tracks and channels * Dummy track and frame reader track. Fix timing. * dt based on camera type * first trial of the new api * Fix audio track * methods for checking for incoming tracks * Web migration part 2 * Fixes for stream api * use rtc description for web.py * experimental cereal proxy * remove old code from bodyav * fix is_started * serialize session description * fix audio * messaging channel wrapper * fix audiotrack * h264 codec preference * Add codec preference to tracks * override sdp codecs * add logging * Move cli stuff to separate file * slight cleanup * Fix audio track * create codec_mime inside force_codec function * fix incoming media estimation * move builders to __init__ * stream updates following builders * Update example script * web.py support for new builder * web speaker fixes * StreamingMediaInfo API * Move things around * should_add_data_channel rename * is_connected_and_ready * fix linter errors * make cli executable * remove dumb comments * logging support * fix parse_info_from_offer * improve type annotations * satisfy linters * Support for waiting for disconnection * Split device tracks into video/audio files. Move audio speaker to audio.py * default dt for dummy video track * Fix cli * new speaker fixes * Remove almost all functionality from web.py * webrtcd * continue refactoring web.py * after handling joystick reset in controlsd with #30409, controls are not necessary anymore * ping endpoint * Update js files to at least support what worked previously * Fixes after some tests on the body * Streaming fixes * Remove the use of WebRTCStreamBuilder. Subclass use is now required * Add todo * delete all streams on shutdown * Replace lastPing with lastChannelMessageTime * Update ping text only if rtc is still on * That should affect the chart too * Fix paths in web * use protocol in SSLContext * remove warnings since aiortc is not used directly anymore * check if task is done in stop * remove channel handler wrapper, since theres only one channel * Move things around * Moved webrtc abstractions to separate repository * Moved webrtcd to tools/webrtc * Update imports * Add bodyrtc as dependency * Add webrtcd to process_config * Remove usage of DummyVideoStreamTrack * Add main to webrtcd * Move webrtcd to system * Fix imports * Move cereal proxy logic outside of runner * Incoming proxy abstractions * Add some tests * Make it executable * Fix process config * Fix imports * Additional tests. Add tests to pyproject.toml * Update poetry lock * New line * Bump aiortc to 1.6.0 * Added teleoprtc_repo as submodule, and linked its source dir * Add init file to webrtc module * Handle aiortc warnings * Ignore deprecation warnings * Ignore resource warning too * Ignore the warnings * find free port for test_webrtcd * Start process inside the test case * random sleep test * test 2 * Test endpoint function instead * Update comment * Add system/webrtc to release * default arguments for body fields * Add teleoprtc to release * Bump teleoprtc * Exclude teleoprtc from static analysis * Use separate event loop for stream session tests
1 year ago
#!/usr/bin/env python3
import argparse
import asyncio
import json
import uuid
import logging
from dataclasses import dataclass, field
from typing import Any, List, Optional, Union
# aiortc and its dependencies have lots of internal warnings :(
import warnings
warnings.filterwarnings("ignore", category=DeprecationWarning)
import aiortc
from aiortc.mediastreams import VideoStreamTrack, AudioStreamTrack
from aiortc.contrib.media import MediaBlackhole
from aiortc.exceptions import InvalidStateError
webrtcd: webrtc streaming server (audio/video/cereal) (#30186) * WebRTCClient and WebRTCServer abstractions * webrtc client implementation * Interactive test scripts * Send localDescriptions as offer/asnwer, as they are different * Tracks need to be added after setting remote description for multi-cam streaming to work * Remove WebRTCStreamingMetadata * Wait for tracks * Move stuff to separate files, rename some things * Refactor everything, create WebRTCStreamBuilder for both offer and answers * ta flight done time to grind * wait for incoming tracks and channels * Dummy track and frame reader track. Fix timing. * dt based on camera type * first trial of the new api * Fix audio track * methods for checking for incoming tracks * Web migration part 2 * Fixes for stream api * use rtc description for web.py * experimental cereal proxy * remove old code from bodyav * fix is_started * serialize session description * fix audio * messaging channel wrapper * fix audiotrack * h264 codec preference * Add codec preference to tracks * override sdp codecs * add logging * Move cli stuff to separate file * slight cleanup * Fix audio track * create codec_mime inside force_codec function * fix incoming media estimation * move builders to __init__ * stream updates following builders * Update example script * web.py support for new builder * web speaker fixes * StreamingMediaInfo API * Move things around * should_add_data_channel rename * is_connected_and_ready * fix linter errors * make cli executable * remove dumb comments * logging support * fix parse_info_from_offer * improve type annotations * satisfy linters * Support for waiting for disconnection * Split device tracks into video/audio files. Move audio speaker to audio.py * default dt for dummy video track * Fix cli * new speaker fixes * Remove almost all functionality from web.py * webrtcd * continue refactoring web.py * after handling joystick reset in controlsd with #30409, controls are not necessary anymore * ping endpoint * Update js files to at least support what worked previously * Fixes after some tests on the body * Streaming fixes * Remove the use of WebRTCStreamBuilder. Subclass use is now required * Add todo * delete all streams on shutdown * Replace lastPing with lastChannelMessageTime * Update ping text only if rtc is still on * That should affect the chart too * Fix paths in web * use protocol in SSLContext * remove warnings since aiortc is not used directly anymore * check if task is done in stop * remove channel handler wrapper, since theres only one channel * Move things around * Moved webrtc abstractions to separate repository * Moved webrtcd to tools/webrtc * Update imports * Add bodyrtc as dependency * Add webrtcd to process_config * Remove usage of DummyVideoStreamTrack * Add main to webrtcd * Move webrtcd to system * Fix imports * Move cereal proxy logic outside of runner * Incoming proxy abstractions * Add some tests * Make it executable * Fix process config * Fix imports * Additional tests. Add tests to pyproject.toml * Update poetry lock * New line * Bump aiortc to 1.6.0 * Added teleoprtc_repo as submodule, and linked its source dir * Add init file to webrtc module * Handle aiortc warnings * Ignore deprecation warnings * Ignore resource warning too * Ignore the warnings * find free port for test_webrtcd * Start process inside the test case * random sleep test * test 2 * Test endpoint function instead * Update comment * Add system/webrtc to release * default arguments for body fields * Add teleoprtc to release * Bump teleoprtc * Exclude teleoprtc from static analysis * Use separate event loop for stream session tests
1 year ago
from aiohttp import web
import capnp
from teleoprtc import WebRTCAnswerBuilder
from teleoprtc.info import parse_info_from_offer
from openpilot.system.webrtc.device.video import LiveStreamVideoStreamTrack
from openpilot.system.webrtc.device.audio import AudioInputStreamTrack, AudioOutputSpeaker
from cereal import messaging
webrtcd: webrtc streaming server (audio/video/cereal) (#30186) * WebRTCClient and WebRTCServer abstractions * webrtc client implementation * Interactive test scripts * Send localDescriptions as offer/asnwer, as they are different * Tracks need to be added after setting remote description for multi-cam streaming to work * Remove WebRTCStreamingMetadata * Wait for tracks * Move stuff to separate files, rename some things * Refactor everything, create WebRTCStreamBuilder for both offer and answers * ta flight done time to grind * wait for incoming tracks and channels * Dummy track and frame reader track. Fix timing. * dt based on camera type * first trial of the new api * Fix audio track * methods for checking for incoming tracks * Web migration part 2 * Fixes for stream api * use rtc description for web.py * experimental cereal proxy * remove old code from bodyav * fix is_started * serialize session description * fix audio * messaging channel wrapper * fix audiotrack * h264 codec preference * Add codec preference to tracks * override sdp codecs * add logging * Move cli stuff to separate file * slight cleanup * Fix audio track * create codec_mime inside force_codec function * fix incoming media estimation * move builders to __init__ * stream updates following builders * Update example script * web.py support for new builder * web speaker fixes * StreamingMediaInfo API * Move things around * should_add_data_channel rename * is_connected_and_ready * fix linter errors * make cli executable * remove dumb comments * logging support * fix parse_info_from_offer * improve type annotations * satisfy linters * Support for waiting for disconnection * Split device tracks into video/audio files. Move audio speaker to audio.py * default dt for dummy video track * Fix cli * new speaker fixes * Remove almost all functionality from web.py * webrtcd * continue refactoring web.py * after handling joystick reset in controlsd with #30409, controls are not necessary anymore * ping endpoint * Update js files to at least support what worked previously * Fixes after some tests on the body * Streaming fixes * Remove the use of WebRTCStreamBuilder. Subclass use is now required * Add todo * delete all streams on shutdown * Replace lastPing with lastChannelMessageTime * Update ping text only if rtc is still on * That should affect the chart too * Fix paths in web * use protocol in SSLContext * remove warnings since aiortc is not used directly anymore * check if task is done in stop * remove channel handler wrapper, since theres only one channel * Move things around * Moved webrtc abstractions to separate repository * Moved webrtcd to tools/webrtc * Update imports * Add bodyrtc as dependency * Add webrtcd to process_config * Remove usage of DummyVideoStreamTrack * Add main to webrtcd * Move webrtcd to system * Fix imports * Move cereal proxy logic outside of runner * Incoming proxy abstractions * Add some tests * Make it executable * Fix process config * Fix imports * Additional tests. Add tests to pyproject.toml * Update poetry lock * New line * Bump aiortc to 1.6.0 * Added teleoprtc_repo as submodule, and linked its source dir * Add init file to webrtc module * Handle aiortc warnings * Ignore deprecation warnings * Ignore resource warning too * Ignore the warnings * find free port for test_webrtcd * Start process inside the test case * random sleep test * test 2 * Test endpoint function instead * Update comment * Add system/webrtc to release * default arguments for body fields * Add teleoprtc to release * Bump teleoprtc * Exclude teleoprtc from static analysis * Use separate event loop for stream session tests
1 year ago
class CerealOutgoingMessageProxy:
def __init__(self, sm: messaging.SubMaster):
self.sm = sm
self.channels: List[aiortc.RTCDataChannel] = []
def add_channel(self, channel: aiortc.RTCDataChannel):
self.channels.append(channel)
def to_json(self, msg_content: Any):
if isinstance(msg_content, capnp._DynamicStructReader):
msg_dict = msg_content.to_dict()
elif isinstance(msg_content, capnp._DynamicListReader):
msg_dict = [self.to_json(msg) for msg in msg_content]
elif isinstance(msg_content, bytes):
msg_dict = msg_content.decode()
else:
msg_dict = msg_content
return msg_dict
def update(self):
# this is blocking in async context...
self.sm.update(0)
for service, updated in self.sm.updated.items():
if not updated:
continue
msg_dict = self.to_json(self.sm[service])
mono_time, valid = self.sm.logMonoTime[service], self.sm.valid[service]
outgoing_msg = {"type": service, "logMonoTime": mono_time, "valid": valid, "data": msg_dict}
encoded_msg = json.dumps(outgoing_msg).encode()
for channel in self.channels:
channel.send(encoded_msg)
class CerealIncomingMessageProxy:
def __init__(self, pm: messaging.PubMaster):
self.pm = pm
def send(self, message: bytes):
msg_json = json.loads(message)
msg_type, msg_data = msg_json["type"], msg_json["data"]
size = None
if not isinstance(msg_data, dict):
size = len(msg_data)
msg = messaging.new_message(msg_type, size=size)
setattr(msg, msg_type, msg_data)
self.pm.send(msg_type, msg)
class CerealProxyRunner:
def __init__(self, proxy: CerealOutgoingMessageProxy):
self.proxy = proxy
self.is_running = False
self.task = None
self.logger = logging.getLogger("webrtcd")
def start(self):
assert self.task is None
self.task = asyncio.create_task(self.run())
def stop(self):
if self.task is None or self.task.done():
return
self.task.cancel()
self.task = None
async def run(self):
while True:
try:
self.proxy.update()
except InvalidStateError:
self.logger.warning("Cereal outgoing proxy invalid state (connection closed)")
break
webrtcd: webrtc streaming server (audio/video/cereal) (#30186) * WebRTCClient and WebRTCServer abstractions * webrtc client implementation * Interactive test scripts * Send localDescriptions as offer/asnwer, as they are different * Tracks need to be added after setting remote description for multi-cam streaming to work * Remove WebRTCStreamingMetadata * Wait for tracks * Move stuff to separate files, rename some things * Refactor everything, create WebRTCStreamBuilder for both offer and answers * ta flight done time to grind * wait for incoming tracks and channels * Dummy track and frame reader track. Fix timing. * dt based on camera type * first trial of the new api * Fix audio track * methods for checking for incoming tracks * Web migration part 2 * Fixes for stream api * use rtc description for web.py * experimental cereal proxy * remove old code from bodyav * fix is_started * serialize session description * fix audio * messaging channel wrapper * fix audiotrack * h264 codec preference * Add codec preference to tracks * override sdp codecs * add logging * Move cli stuff to separate file * slight cleanup * Fix audio track * create codec_mime inside force_codec function * fix incoming media estimation * move builders to __init__ * stream updates following builders * Update example script * web.py support for new builder * web speaker fixes * StreamingMediaInfo API * Move things around * should_add_data_channel rename * is_connected_and_ready * fix linter errors * make cli executable * remove dumb comments * logging support * fix parse_info_from_offer * improve type annotations * satisfy linters * Support for waiting for disconnection * Split device tracks into video/audio files. Move audio speaker to audio.py * default dt for dummy video track * Fix cli * new speaker fixes * Remove almost all functionality from web.py * webrtcd * continue refactoring web.py * after handling joystick reset in controlsd with #30409, controls are not necessary anymore * ping endpoint * Update js files to at least support what worked previously * Fixes after some tests on the body * Streaming fixes * Remove the use of WebRTCStreamBuilder. Subclass use is now required * Add todo * delete all streams on shutdown * Replace lastPing with lastChannelMessageTime * Update ping text only if rtc is still on * That should affect the chart too * Fix paths in web * use protocol in SSLContext * remove warnings since aiortc is not used directly anymore * check if task is done in stop * remove channel handler wrapper, since theres only one channel * Move things around * Moved webrtc abstractions to separate repository * Moved webrtcd to tools/webrtc * Update imports * Add bodyrtc as dependency * Add webrtcd to process_config * Remove usage of DummyVideoStreamTrack * Add main to webrtcd * Move webrtcd to system * Fix imports * Move cereal proxy logic outside of runner * Incoming proxy abstractions * Add some tests * Make it executable * Fix process config * Fix imports * Additional tests. Add tests to pyproject.toml * Update poetry lock * New line * Bump aiortc to 1.6.0 * Added teleoprtc_repo as submodule, and linked its source dir * Add init file to webrtc module * Handle aiortc warnings * Ignore deprecation warnings * Ignore resource warning too * Ignore the warnings * find free port for test_webrtcd * Start process inside the test case * random sleep test * test 2 * Test endpoint function instead * Update comment * Add system/webrtc to release * default arguments for body fields * Add teleoprtc to release * Bump teleoprtc * Exclude teleoprtc from static analysis * Use separate event loop for stream session tests
1 year ago
except Exception as ex:
self.logger.error("Cereal outgoing proxy failure: %s", ex)
await asyncio.sleep(0.01)
class StreamSession:
def __init__(self, sdp: str, cameras: List[str], incoming_services: List[str], outgoing_services: List[str], debug_mode: bool = False):
config = parse_info_from_offer(sdp)
builder = WebRTCAnswerBuilder(sdp)
assert len(cameras) == config.n_expected_camera_tracks, "Incoming stream has misconfigured number of video tracks"
for cam in cameras:
track = LiveStreamVideoStreamTrack(cam) if not debug_mode else VideoStreamTrack()
builder.add_video_stream(cam, track)
if config.expected_audio_track:
track = AudioInputStreamTrack() if not debug_mode else AudioStreamTrack()
builder.add_audio_stream(track)
if config.incoming_audio_track:
self.audio_output_cls = AudioOutputSpeaker if not debug_mode else MediaBlackhole
builder.offer_to_receive_audio_stream()
self.stream = builder.stream()
self.identifier = str(uuid.uuid4())
self.outgoing_bridge = CerealOutgoingMessageProxy(messaging.SubMaster(outgoing_services))
self.incoming_bridge = CerealIncomingMessageProxy(messaging.PubMaster(incoming_services))
self.outgoing_bridge_runner = CerealProxyRunner(self.outgoing_bridge)
self.audio_output: Optional[Union[AudioOutputSpeaker, MediaBlackhole]] = None
self.run_task: Optional[asyncio.Task] = None
self.logger = logging.getLogger("webrtcd")
self.logger.info("New stream session (%s), cameras %s, audio in %s out %s, incoming services %s, outgoing services %s",
self.identifier, cameras, config.incoming_audio_track, config.expected_audio_track, incoming_services, outgoing_services)
def start(self):
self.run_task = asyncio.create_task(self.run())
def stop(self):
if self.run_task.done():
return
self.run_task.cancel()
self.run_task = None
asyncio.run(self.post_run_cleanup())
async def get_answer(self):
return await self.stream.start()
async def message_handler(self, message: bytes):
try:
self.incoming_bridge.send(message)
except Exception as ex:
self.logger.error("Cereal incoming proxy failure: %s", ex)
async def run(self):
try:
await self.stream.wait_for_connection()
if self.stream.has_messaging_channel():
self.stream.set_message_handler(self.message_handler)
channel = self.stream.get_messaging_channel()
self.outgoing_bridge_runner.proxy.add_channel(channel)
self.outgoing_bridge_runner.start()
if self.stream.has_incoming_audio_track():
track = self.stream.get_incoming_audio_track(buffered=False)
self.audio_output = self.audio_output_cls()
self.audio_output.addTrack(track)
self.audio_output.start()
self.logger.info("Stream session (%s) connected", self.identifier)
await self.stream.wait_for_disconnection()
await self.post_run_cleanup()
self.logger.info("Stream session (%s) ended", self.identifier)
except Exception as ex:
self.logger.error("Stream session failure: %s", ex)
async def post_run_cleanup(self):
await self.stream.stop()
self.outgoing_bridge_runner.stop()
if self.audio_output:
self.audio_output.stop()
@dataclass
class StreamRequestBody:
sdp: str
cameras: List[str]
bridge_services_in: List[str] = field(default_factory=list)
bridge_services_out: List[str] = field(default_factory=list)
async def get_stream(request: web.Request):
stream_dict, debug_mode = request.app['streams'], request.app['debug']
raw_body = await request.json()
body = StreamRequestBody(**raw_body)
session = StreamSession(body.sdp, body.cameras, body.bridge_services_in, body.bridge_services_out, debug_mode)
answer = await session.get_answer()
session.start()
stream_dict[session.identifier] = session
return web.json_response({"sdp": answer.sdp, "type": answer.type})
async def on_shutdown(app: web.Application):
for session in app['streams'].values():
session.stop()
del app['streams']
def webrtcd_thread(host: str, port: int, debug: bool):
logging.basicConfig(level=logging.CRITICAL, handlers=[logging.StreamHandler()])
logging_level = logging.DEBUG if debug else logging.INFO
logging.getLogger("WebRTCStream").setLevel(logging_level)
logging.getLogger("webrtcd").setLevel(logging_level)
app = web.Application()
app['streams'] = dict()
app['debug'] = debug
app.on_shutdown.append(on_shutdown)
app.router.add_post("/stream", get_stream)
web.run_app(app, host=host, port=port)
def main():
parser = argparse.ArgumentParser(description="WebRTC daemon")
parser.add_argument("--host", type=str, default="0.0.0.0", help="Host to listen on")
parser.add_argument("--port", type=int, default=5001, help="Port to listen on")
parser.add_argument("--debug", action="store_true", help="Enable debug mode")
args = parser.parse_args()
webrtcd_thread(args.host, args.port, args.debug)
if __name__=="__main__":
main()