Implement new logic depending on Discord's rate limit information coded in headers instead of arbitrary sleep time

src/discord.py

@@ -129,16 +129,20 @@ async def send_to_discord_webhook_monitoring(data: DiscordMessage):
 			return 3
 
 
-async def send_to_discord_webhook(data: DiscordMessage, webhook_url: str):
+async def send_to_discord_webhook(data: DiscordMessage, webhook_url: str) -> tuple:
+	"""Sends a message to webhook
+
+	:return tuple(status code for request, rate limit info (None for can send more, string for amount of seconds to wait)"""
 	header = settings["header"]
 	header['Content-Type'] = 'application/json'
 	async with aiohttp.ClientSession(headers=header, timeout=aiohttp.ClientTimeout(5.0)) as session:
 		try:
 			result = await session.post("https://discord.com/api/webhooks/"+webhook_url, data=repr(data))
+			rate_limit = None if int(result.headers.get('x-ratelimit-limit')) > 0 else result.headers.get('x-ratelimit-reset-after')
 		except (aiohttp.ClientConnectionError, aiohttp.ServerConnectionError, TimeoutError):
 			logger.exception("Could not send the message to Discord")
-			return 3
+			return 3, None
-		return await handle_discord_http(result.status, repr(data), await result.text(), data)
+		return await handle_discord_http(result.status, repr(data), await result.text(), data), rate_limit
 
 
 async def handle_discord_http(code, formatted_embed, result, dmsg):

src/msgqueue.py

@@ -43,10 +43,12 @@ class MessageQueue:
 	async def send_msg_set(self, msg_set: tuple):
 		webhook_url, messages = msg_set
 		for msg in messages:
-			if await send_to_discord_webhook(msg, webhook_url) < 2:
+			status = await send_to_discord_webhook(msg, webhook_url)
+			if status[0] < 2:
 				logger.debug("Sending message succeeded")
 				self._queue.remove(msg)
-				await asyncio.sleep(1.9)
+				if status[1] is not None:
+					await asyncio.sleep(float(status[1]))
 			else:
 				logger.debug("Sending message failed")
 				break