Switch HTTP streamer to a thread and pipe its output to a local pipe where the fetch chunk can access it the same way our transcode processes would be accessed. Simplifies the code.

This commit is contained in:
SergeantPanda 2025-10-12 09:42:15 -05:00
parent 74280baa85
commit 404d2f82a3
3 changed files with 194 additions and 91 deletions

View file

@ -103,5 +103,9 @@ class ConfigHelper:
@staticmethod
def chunk_timeout():
"""Get chunk timeout in seconds (used for both socket and HTTP read timeouts)"""
"""
Get chunk timeout in seconds (used for both socket and HTTP read timeouts).
This controls how long we wait for each chunk before timing out.
Set this higher (e.g., 30s) for slow providers that may have intermittent delays.
"""
return ConfigHelper.get('CHUNK_TIMEOUT', 5) # Default 5 seconds

View file

@ -0,0 +1,138 @@
"""
HTTP Stream Reader - Thread-based HTTP stream reader that writes to a pipe.
This allows us to use the same fetch_chunk() path for both transcode and HTTP streams.
"""
import threading
import os
import requests
from requests.adapters import HTTPAdapter
from .utils import get_logger
logger = get_logger()
class HTTPStreamReader:
"""Thread-based HTTP stream reader that writes to a pipe"""
def __init__(self, url, user_agent=None, chunk_size=8192):
self.url = url
self.user_agent = user_agent
self.chunk_size = chunk_size
self.session = None
self.response = None
self.thread = None
self.pipe_read = None
self.pipe_write = None
self.running = False
def start(self):
"""Start the HTTP stream reader thread"""
# Create a pipe (works on Windows and Unix)
self.pipe_read, self.pipe_write = os.pipe()
# Start the reader thread
self.running = True
self.thread = threading.Thread(target=self._read_stream, daemon=True)
self.thread.start()
logger.info(f"Started HTTP stream reader thread for {self.url}")
return self.pipe_read
def _read_stream(self):
"""Thread worker that reads HTTP stream and writes to pipe"""
try:
# Build headers
headers = {}
if self.user_agent:
headers['User-Agent'] = self.user_agent
logger.info(f"HTTP reader connecting to {self.url}")
# Create session
self.session = requests.Session()
# Disable retries for faster failure detection
adapter = HTTPAdapter(max_retries=0, pool_connections=1, pool_maxsize=1)
self.session.mount('http://', adapter)
self.session.mount('https://', adapter)
# Stream the URL
self.response = self.session.get(
self.url,
headers=headers,
stream=True,
timeout=(5, 30) # 5s connect, 30s read
)
if self.response.status_code != 200:
logger.error(f"HTTP {self.response.status_code} from {self.url}")
return
logger.info(f"HTTP reader connected successfully, streaming data...")
# Stream chunks to pipe
chunk_count = 0
for chunk in self.response.iter_content(chunk_size=self.chunk_size):
if not self.running:
break
if chunk:
try:
# Write binary data to pipe
os.write(self.pipe_write, chunk)
chunk_count += 1
# Log progress periodically
if chunk_count % 1000 == 0:
logger.debug(f"HTTP reader streamed {chunk_count} chunks")
except OSError as e:
logger.error(f"Pipe write error: {e}")
break
logger.info("HTTP stream ended")
except requests.exceptions.RequestException as e:
logger.error(f"HTTP reader request error: {e}")
except Exception as e:
logger.error(f"HTTP reader unexpected error: {e}", exc_info=True)
finally:
self.running = False
# Close write end of pipe to signal EOF
try:
if self.pipe_write is not None:
os.close(self.pipe_write)
self.pipe_write = None
except:
pass
def stop(self):
"""Stop the HTTP stream reader"""
logger.info("Stopping HTTP stream reader")
self.running = False
# Close response
if self.response:
try:
self.response.close()
except:
pass
# Close session
if self.session:
try:
self.session.close()
except:
pass
# Close write end of pipe
if self.pipe_write is not None:
try:
os.close(self.pipe_write)
self.pipe_write = None
except:
pass
# Wait for thread
if self.thread and self.thread.is_alive():
self.thread.join(timeout=2.0)

View file

@ -114,6 +114,9 @@ class StreamManager:
self.stderr_reader_thread = None
self.ffmpeg_input_phase = True # Track if we're still reading input info
# Add HTTP reader thread property
self.http_reader = None
def _create_session(self):
"""Create and configure requests session with optimal settings"""
session = requests.Session()
@ -740,9 +743,9 @@ class StreamManager:
def _establish_http_connection(self):
"""Establish a direct HTTP connection to the stream"""
"""Establish HTTP connection using thread-based reader (same as transcode path)"""
try:
logger.debug(f"Using TS Proxy to connect to stream: {self.url}")
logger.debug(f"Using HTTP streamer thread to connect to stream: {self.url}")
# Check if we already have active HTTP connections
if self.current_response or self.current_session:
@ -759,43 +762,39 @@ class StreamManager:
logger.debug(f"Closing existing transcode process before establishing HTTP connection for channel {self.channel_id}")
self._close_socket()
# Create new session for each connection attempt
session = self._create_session()
self.current_session = session
# Use HTTPStreamReader to fetch stream and pipe to a readable file descriptor
# This allows us to use the same fetch_chunk() path as transcode
from .http_streamer import HTTPStreamReader
# Stream the URL with proper timeout handling
# Use same chunk timeout as socket connections for consistency
chunk_timeout = ConfigHelper.chunk_timeout()
response = session.get(
self.url,
stream=True,
timeout=(5, 10) # 5s connect timeout, 10s read timeout
# Create and start the HTTP stream reader
self.http_reader = HTTPStreamReader(
url=self.url,
user_agent=self.user_agent,
chunk_size=self.chunk_size
)
self.current_response = response
if response.status_code == 200:
self.connected = True
self.healthy = True
logger.info(f"Successfully connected to stream source for channel {self.channel_id}")
# Start the reader thread and get the read end of the pipe
pipe_fd = self.http_reader.start()
# Store connection start time for stability tracking
self.connection_start_time = time.time()
# Wrap the file descriptor in a file object (same as transcode stdout)
import os
self.socket = os.fdopen(pipe_fd, 'rb', buffering=0)
self.connected = True
self.healthy = True
# Set channel state to waiting for clients
self._set_waiting_for_clients()
logger.info(f"Successfully started HTTP streamer thread for channel {self.channel_id}")
# Store connection start time for stability tracking
self.connection_start_time = time.time()
# Set channel state to waiting for clients
self._set_waiting_for_clients()
return True
return True
else:
logger.error(f"Failed to connect to stream for channel {self.channel_id}: HTTP {response.status_code}")
self._close_connection()
return False
except requests.exceptions.RequestException as e:
logger.error(f"HTTP request error: {e}")
self._close_connection()
return False
except Exception as e:
logger.error(f"Error establishing HTTP connection for channel {self.channel_id}: {e}", exc_info=True)
self._close_connection()
self._close_socket()
return False
def _update_bytes_processed(self, chunk_size):
@ -823,66 +822,19 @@ class StreamManager:
logger.error(f"Error updating bytes processed: {e}")
def _process_stream_data(self):
"""Process stream data until disconnect or error"""
"""Process stream data until disconnect or error - unified path for both transcode and HTTP"""
try:
if self.transcode:
# Handle transcoded stream data
while self.running and self.connected and not self.stop_requested and not self.needs_stream_switch:
if self.fetch_chunk():
self.last_data_time = time.time()
else:
if not self.running:
break
gevent.sleep(0.1) # REPLACE time.sleep(0.1)
else:
# Handle direct HTTP connection
chunk_count = 0
# Check if response is still valid before attempting to read
if not self.current_response:
logger.debug(f"Response object is None for channel {self.channel_id}, connection likely closed")
self.connected = False
return
try:
for chunk in self.current_response.iter_content(chunk_size=self.chunk_size):
# Check if we've been asked to stop
if self.stop_requested or self.url_switching or self.needs_stream_switch:
break
if chunk:
# Track chunk size before adding to buffer
chunk_size = len(chunk)
self._update_bytes_processed(chunk_size)
# Add chunk to buffer with TS packet alignment
success = self.buffer.add_chunk(chunk)
if success:
self.last_data_time = time.time()
chunk_count += 1
# Update last data timestamp in Redis
if hasattr(self.buffer, 'redis_client') and self.buffer.redis_client:
last_data_key = RedisKeys.last_data(self.buffer.channel_id)
self.buffer.redis_client.set(last_data_key, str(time.time()), ex=60)
except (requests.exceptions.ReadTimeout, ReadTimeoutError, requests.exceptions.ConnectionError) as e:
if self.stop_requested or self.url_switching:
logger.debug(f"Expected connection error during shutdown/URL switch for channel {self.channel_id}: {e}")
else:
# Handle timeout errors - log and close connection, let main loop handle retry
logger.warning(f"Stream read timeout for channel {self.channel_id}: {e}")
# Close the current connection
self._close_connection()
return # Exit this method, main loop will retry based on retry_count
except (AttributeError, ConnectionError) as e:
if self.stop_requested or self.url_switching:
logger.debug(f"Expected connection error during shutdown/URL switch for channel {self.channel_id}: {e}")
else:
logger.error(f"Unexpected stream error for channel {self.channel_id}: {e}")
raise
# Both transcode and HTTP now use the same subprocess/socket approach
# This gives us perfect control: check flags between chunks, timeout just returns False
while self.running and self.connected and not self.stop_requested and not self.needs_stream_switch:
if self.fetch_chunk():
self.last_data_time = time.time()
else:
# fetch_chunk() returned False - could be timeout, no data, or error
if not self.running:
break
# Brief sleep before retry to avoid tight loop
gevent.sleep(0.1)
except Exception as e:
logger.error(f"Error processing stream data for channel {self.channel_id}: {e}", exc_info=True)
@ -1206,6 +1158,15 @@ class StreamManager:
if self.current_response or self.current_session:
self._close_connection()
# Stop HTTP reader thread if it exists
if hasattr(self, 'http_reader') and self.http_reader:
try:
logger.debug(f"Stopping HTTP reader thread for channel {self.channel_id}")
self.http_reader.stop()
self.http_reader = None
except Exception as e:
logger.debug(f"Error stopping HTTP reader for channel {self.channel_id}: {e}")
# Otherwise handle socket and transcode resources
if self.socket:
try: