File size: 8,265 Bytes
78efe79 440418c 302cdd6 7edf3cc dc80b35 6a13d8f 1899abe b4bdc7a 302cdd6 7edf3cc f2ef6a6 4f7204d 262fbfb 4f7204d 407a575 262fbfb 672b059 440418c 262fbfb 440418c 22dee1c 440418c 22dee1c 08baccf 262fbfb 70b82a0 dc80b35 262fbfb 829fd20 1e3100b 262fbfb 302cdd6 11fec5e 302cdd6 8892907 302cdd6 262fbfb 302cdd6 1899abe 302cdd6 262fbfb 8892907 262fbfb 302cdd6 2db254d 302cdd6 8892907 262fbfb 8892907 302cdd6 1899abe 302cdd6 1899abe 302cdd6 1899abe 302cdd6 1899abe 302cdd6 1899abe 302cdd6 829fd20 1e3100b 1899abe 302cdd6 304f275 302cdd6 8892907 302cdd6 1e3100b 1899abe 302cdd6 8892907 2db254d 302cdd6 1899abe bc99da9 302cdd6 bc99da9 1899abe b63eee6 11fec5e b63eee6 2db254d 8892907 1899abe 302cdd6 829fd20 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 |
import discord
import logging
import os
import re
import asyncio
import subprocess
import aiohttp
from huggingface_hub import InferenceClient
from googleapiclient.discovery import build
from youtube_transcript_api import YouTubeTranscriptApi
from youtube_transcript_api.formatters import TextFormatter
from dotenv import load_dotenv
# ํ๊ฒฝ ๋ณ์ ๋ก๋
load_dotenv()
# ๋ก๊น
์ค์
logging.basicConfig(level=logging.DEBUG, format='%(asctime)s:%(levelname)s:%(name)s:%(message)s', handlers=[logging.StreamHandler()])
# ์ธํ
ํธ ์ค์
intents = discord.Intents.default()
intents.message_content = True
intents.messages = True
intents.guilds = True
intents.guild_messages = True
# ์ถ๋ก API ํด๋ผ์ด์ธํธ ์ค์
hf_client = InferenceClient("CohereForAI/c4ai-command-r-plus-08-2024", token=os.getenv("HF_TOKEN"))
# YouTube API ์ค์
API_KEY = os.getenv("YOUTUBE_API_KEY")
youtube_service = build('youtube', 'v3', developerKey=API_KEY)
# ํน์ ์ฑ๋ ID
SPECIFIC_CHANNEL_ID = int(os.getenv("DISCORD_CHANNEL_ID"))
# ์นํ
URL ์ค์
WEBHOOK_URL = "https://connect.pabbly.com/workflow/sendwebhookdata/IjU3NjUwNTY1MDYzMjA0MzA1MjY4NTUzMDUxMzUi_pc"
# ์ ์ก ์คํจ ์ ์ฌ์๋ ํ์
MAX_RETRIES = 3
class MyClient(discord.Client):
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
self.is_processing = False
self.session = None
async def on_ready(self):
logging.info(f'{self.user}๋ก ๋ก๊ทธ์ธ๋์์ต๋๋ค!')
# web.py ํ์ผ ์คํ
subprocess.Popen(["python", "web.py"])
logging.info("Web.py ์๋ฒ๊ฐ ์์๋์์ต๋๋ค.")
# aiohttp ํด๋ผ์ด์ธํธ ์ธ์
์์ฑ
self.session = aiohttp.ClientSession()
# ๋ด์ด ์์๋ ๋ ์๋ด ๋ฉ์์ง๋ฅผ ์ ์ก
channel = self.get_channel(SPECIFIC_CHANNEL_ID)
if channel:
await channel.send("์ ํ๋ธ ๋น๋์ค URL์ ์
๋ ฅํ๋ฉด, ์๋ง๊ณผ ๋๊ธ์ ๊ธฐ๋ฐ์ผ๋ก ๋ต๊ธ์ ์์ฑํฉ๋๋ค.")
async def on_message(self, message):
if message.author == self.user:
return
if not self.is_message_in_specific_channel(message):
return
if self.is_processing:
return
self.is_processing = True
try:
video_id = extract_video_id(message.content)
if video_id:
transcript = await get_best_available_transcript(video_id)
comments = await get_video_comments(video_id)
if comments and transcript:
replies = await generate_replies(comments, transcript)
await create_thread_and_send_replies(message, video_id, comments, replies, self.session)
else:
await message.channel.send("์๋ง์ด๋ ๋๊ธ์ ๊ฐ์ ธ์ฌ ์ ์์ต๋๋ค.")
else:
await message.channel.send("์ ํจํ ์ ํ๋ธ ๋น๋์ค URL์ ์ ๊ณตํด ์ฃผ์ธ์.")
finally:
self.is_processing = False
def is_message_in_specific_channel(self, message):
return message.channel.id == SPECIFIC_CHANNEL_ID or (
isinstance(message.channel, discord.Thread) and message.channel.parent_id == SPECIFIC_CHANNEL_ID
)
async def close(self):
# aiohttp ํด๋ผ์ด์ธํธ ์ธ์
์ข
๋ฃ
if self.session:
await self.session.close()
await super().close()
def extract_video_id(url):
video_id = None
youtube_regex = (
r'(https?://)?(www\.)?'
'(youtube|youtu|youtube-nocookie)\.(com|be)/'
'(watch\?v=|embed/|v/|.+\?v=)?([^&=%\?]{11})')
match = re.match(youtube_regex, url)
if match:
video_id = match.group(6)
logging.debug(f'์ถ์ถ๋ ๋น๋์ค ID: {video_id}')
return video_id
async def get_best_available_transcript(video_id):
try:
transcript = YouTubeTranscriptApi.get_transcript(video_id, languages=['ko'])
except Exception as e:
logging.warning(f'ํ๊ตญ์ด ์๋ง ๊ฐ์ ธ์ค๊ธฐ ์ค๋ฅ: {e}')
try:
transcript = YouTubeTranscriptApi.get_transcript(video_id, languages=['en'])
except Exception as e:
logging.warning(f'์์ด ์๋ง ๊ฐ์ ธ์ค๊ธฐ ์ค๋ฅ: {e}')
try:
transcripts = YouTubeTranscriptApi.list_transcripts(video_id)
transcript = transcripts.find_manually_created_transcript().fetch()
except Exception as e:
logging.error(f'๋์ฒด ์๋ง ๊ฐ์ ธ์ค๊ธฐ ์ค๋ฅ: {e}')
return None
formatter = TextFormatter()
transcript_text = formatter.format_transcript(transcript)
logging.debug(f'๊ฐ์ ธ์จ ์๋ง: {transcript_text}')
return transcript_text
async def get_video_comments(video_id):
comments = []
response = youtube_service.commentThreads().list(
part='snippet',
videoId=video_id,
maxResults=100 # ์ต๋ 100๊ฐ์ ๋๊ธ ๊ฐ์ ธ์ค๊ธฐ
).execute()
for item in response.get('items', []):
comment = item['snippet']['topLevelComment']['snippet']['textOriginal']
comment_id = item['snippet']['topLevelComment']['id']
comments.append((comment, comment_id))
logging.debug(f'๊ฐ์ ธ์จ ๋๊ธ: {comments}')
return comments
async def generate_replies(comments, transcript):
replies = []
for comment, _ in comments:
messages = [
{"role": "system", "content": """๋์ ์ด๋ฆ์ OpenFreeAI์ด๋ค. ๋ต๊ธ ์์ฑํ ๊ฐ์ฅ ๋ง์ง๋ง์ ๋์ ์ด๋ฆ์ ๋ฐํ๊ณ ๊ณต์ํ๊ฒ ์ธ์ฌํ๋ผ. ๋น๋์ค ์๋ง: {transcript}"""},
{"role": "user", "content": comment}
]
loop = asyncio.get_event_loop()
response = await loop.run_in_executor(None, lambda: hf_client.chat_completion(
messages, max_tokens=250, temperature=0.7, top_p=0.85))
if response.choices and response.choices[0].message:
reply = response.choices[0].message['content'].strip()
else:
reply = "๋ต๊ธ์ ์์ฑํ ์ ์์ต๋๋ค."
replies.append(reply)
logging.debug(f'์์ฑ๋ ๋ต๊ธ: {replies}')
return replies
async def send_webhook_data(session, chunk_data, chunk_number):
for attempt in range(MAX_RETRIES):
try:
async with session.post(WEBHOOK_URL, json=chunk_data) as resp:
if resp.status == 200:
logging.info(f"์นํ
์ผ๋ก ๋ฐ์ดํฐ ์ ์ก ์ฑ๊ณต: {chunk_number} ๋ฒ์งธ ์๋")
return True # ์ฑ๊ณต ์ ์ข
๋ฃ
else:
logging.error(f"์นํ
์ผ๋ก ๋ฐ์ดํฐ ์ ์ก ์คํจ: {resp.status}, {chunk_number} ๋ฒ์งธ ์๋")
except aiohttp.ClientError as e:
logging.error(f"์นํ
์ ์ก ์ค ์ค๋ฅ ๋ฐ์: {e}, {chunk_number} ๋ฒ์งธ ์๋")
await asyncio.sleep(1) # ์ฌ์๋ ์ ์ ์ ์ ๋๊ธฐ
return False # ์ฌ์๋ ํ์ ์ด๊ณผ ์ ์คํจ๋ก ๊ฐ์ฃผ
async def create_thread_and_send_replies(message, video_id, comments, replies, session):
thread = await message.channel.create_thread(name=f"{message.author.name}์ ๋๊ธ ๋ต๊ธ", message=message)
webhook_data = {"video_id": video_id, "replies": []}
for (comment, comment_id), reply in zip(comments, replies):
embed = discord.Embed(description=f"**๋๊ธ**: {comment}\n**๋ต๊ธ**: {reply}")
await thread.send(embed=embed)
# ์นํ
๋ฐ์ดํฐ ์ค๋น (comment id ํฌํจ)
webhook_data["replies"].append({"comment": comment, "reply": reply, "comment_id": comment_id})
# ๋ฐ์ดํฐ๋ฅผ ์ฌ๋ฌ ๋ฒ ๋๋์ด ์ ์ก
chunk_size = 1 # ์ ์กํ ๋ฐ์ดํฐ์ ๊ฐ์๋ฅผ 1๋ก ์ค์ ํ์ฌ ๊ฐ ๋ฐ์ดํฐ๋ฅผ ๋ณ๋๋ก ์ ์ก
for i in range(0, len(webhook_data["replies"]), chunk_size):
chunk = webhook_data["replies"][i:i+chunk_size]
chunk_data = {"video_id": video_id, "replies": chunk}
success = await send_webhook_data(session, chunk_data, i // chunk_size + 1)
if not success:
logging.error(f"๋ฐ์ดํฐ ์ ์ก ์คํจ: {i // chunk_size + 1} ๋ฒ์งธ ์ฒญํฌ")
if __name__ == "__main__":
discord_client = MyClient(intents=intents)
discord_client.run(os.getenv('DISCORD_TOKEN'))
|