TrickleClient: improve error handling in `send_data_loop`

[eliteprox]

Investigate performance and error handling for send_data_loop:

pytrickle/pytrickle/client.py

Lines 295 to 327 in d5a9bbc

	async def _send_data_loop(self):
	"""Send data to the server every 333ms, batching all available items."""
	try:
	while not self.stop_event.is_set() and not self.error_event.is_set():
	# Wait for send_data_interval or until stop/error event is set
	if await self._wait_for_interval(self.send_data_interval):
	break # Stop or error event was set, exit loop
	# Pull all available items from the data_queue
	data_items = []
	while len(self.data_queue) > 0 and not self.stop_event.is_set() and not self.error_event.is_set():
	data = self.data_queue.popleft()
	if data is None:
	# Sentinel value to stop loop
	if data_items:
	# Send any remaining items before stopping
	break
	else:
	return # No items to send, just stop
	else:
	data_items.append(data)
	# Send all collected data items
	if len(data_items) > 0:
	try:
	data_str = json.dumps(data_items) + "\n"
	except Exception as e:
	logger.error(f"Error serializing data items: {e}")
	continue
	await self.protocol.publish_data(data_str)
	except Exception as e:
	logger.error(f"Error in data sending loop: {e}")

We could write additional tests to validate performance of the data publisher code path in client.py

I think we should wrap this in a try/except block as well, and should allow us to remove the check for the queue length.

Originally posted by @pschroedl in #25 (comment)

[linear]

COMFY-265