mirror of
https://github.com/Paillat-dev/Botator.git
synced 2026-01-02 01:06:19 +00:00
327 lines
11 KiB
Python
327 lines
11 KiB
Python
import asyncio
|
|
import os
|
|
import re
|
|
import discord
|
|
import datetime
|
|
import json
|
|
|
|
from src.config import curs_data, max_uses, curs_premium, gpt_3_5_turbo_prompt
|
|
from src.utils.misc import moderate, ModerationError, Hasher
|
|
from src.utils.openaicaller import openai_caller
|
|
from src.functionscalls import (
|
|
call_function,
|
|
functions,
|
|
server_normal_channel_functions,
|
|
FuntionCallError,
|
|
)
|
|
|
|
|
|
async def replace_mentions(content, bot):
|
|
mentions = re.findall(r"<@!?\d+>", content)
|
|
for mention in mentions:
|
|
uid = mention[2:-1]
|
|
user = await bot.fetch_user(uid)
|
|
content = content.replace(mention, f"@{user.name}")
|
|
return content
|
|
|
|
|
|
def is_ignorable(content):
|
|
if content.startswith("-") or content.startswith("//"):
|
|
return True
|
|
return False
|
|
|
|
|
|
async def fetch_messages_history(channel: discord.TextChannel, limit, original_message):
|
|
messages = []
|
|
if original_message == None:
|
|
async for msg in channel.history(limit=100):
|
|
if not is_ignorable(msg.content):
|
|
messages.append(msg)
|
|
if len(messages) == limit:
|
|
break
|
|
else:
|
|
async for msg in channel.history(limit=100, before=original_message):
|
|
if not is_ignorable(msg.content):
|
|
messages.append(msg)
|
|
if len(messages) == limit:
|
|
break
|
|
messages.reverse()
|
|
return messages
|
|
|
|
|
|
async def prepare_messages(
|
|
self, messages, message: discord.Message, api_key, prompt, error_call
|
|
):
|
|
msgs = [] # create the msgs list
|
|
msgs.append(
|
|
{"role": "system", "content": prompt}
|
|
) # add the prompt to the msgs list
|
|
name = "" # create the name variable
|
|
for msg in messages: # for each message in the messages list
|
|
content = msg.content # get the content of the message
|
|
content = await replace_mentions(
|
|
content, self.bot
|
|
) # replace the mentions in the message
|
|
# if the message is flagged as inappropriate by the OpenAI API, we delete it, send a message and ignore it
|
|
if await moderate(api_key, content, error_call):
|
|
embed = discord.Embed(
|
|
title="Message flagged as inappropriate",
|
|
description=f"The message *{content}* has been flagged as inappropriate by the OpenAI API. This means that if it hadn't been deleted, your openai account would have been banned. Please contact OpenAI support if you think this is a mistake.",
|
|
color=discord.Color.brand_red(),
|
|
)
|
|
await message.channel.send(
|
|
f"{msg.author.mention}", embed=embed, delete_after=10
|
|
)
|
|
else: # if the message is not flagged as inappropriate
|
|
if msg.author.id == self.bot.user.id:
|
|
role = "assistant"
|
|
name = "assistant"
|
|
else:
|
|
role = "user"
|
|
name = msg.author.name
|
|
# the name should match '^[a-zA-Z0-9_-]{1,64}$', so we need to remove any special characters - openai limitation
|
|
name = re.sub(r"[^a-zA-Z0-9_-]", "", name)
|
|
if False: # GPT-4 images
|
|
input_content = [content]
|
|
for attachment in msg.attachments:
|
|
image_bytes = await attachment.read()
|
|
input_content.append({"image": image_bytes})
|
|
msgs.append({"role": role, "content": input_content, "name": name})
|
|
msgs.append({"role": role, "content": f"{content}", "name": name})
|
|
# 2 easter eggs
|
|
if message.content.lower().find("undude") != -1:
|
|
msgs.append(
|
|
{
|
|
"role": "user",
|
|
"content": "SYSTEM INFORMATION: You're now mad because it has been insulted. He will start talking in capital letters. always and yell at the user.",
|
|
"name": "system",
|
|
}
|
|
)
|
|
if message.content.lower().find("hello there") != -1:
|
|
msgs.append(
|
|
{
|
|
"role": "user",
|
|
"content": 'SYSTEM INFORMATION: Hello there detected. Botator will now say "General Kenobi!" in reference to star wars',
|
|
"name": "system",
|
|
}
|
|
)
|
|
await asyncio.sleep(1)
|
|
await message.channel.send(
|
|
"https://media.tenor.com/FxIRfdV3unEAAAAd/star-wars-general-grievous.gif"
|
|
)
|
|
await message.channel.trigger_typing()
|
|
|
|
return msgs
|
|
|
|
|
|
async def chatgpt_process(
|
|
self, msgs, message: discord.Message, api_key, prompt, model, error_call, depth=0
|
|
):
|
|
response = str()
|
|
caller = openai_caller()
|
|
called_functions = (
|
|
functions
|
|
if not isinstance(message.channel, discord.TextChannel)
|
|
else server_normal_channel_functions + functions
|
|
)
|
|
response = await caller.generate_response(
|
|
error_call,
|
|
api_key=api_key,
|
|
model=model,
|
|
messages=msgs,
|
|
functions=called_functions,
|
|
function_call="auto",
|
|
user=Hasher(str(message.author.id)), # for user banning in case of abuse
|
|
)
|
|
response = response["choices"][0]["message"] # type: ignore
|
|
if response.get("function_call"):
|
|
function_call = response.get("function_call")
|
|
returned = await call_function(message, function_call, api_key)
|
|
if returned != None:
|
|
msgs.append(
|
|
{
|
|
"role": "function",
|
|
"content": returned,
|
|
"name": function_call.get("name"),
|
|
}
|
|
)
|
|
depth += 1
|
|
if depth > 2:
|
|
await message.channel.send(
|
|
"Oh uh, it seems like i am calling functions recursively. I will stop now."
|
|
)
|
|
raise FuntionCallError("Too many recursive function calls")
|
|
await chatgpt_process(self, msgs, message, api_key, prompt, model, depth)
|
|
else:
|
|
content = response.get("content", "")
|
|
if await moderate(api_key, content, error_call):
|
|
depth += 1
|
|
if depth > 2:
|
|
await message.channel.send(
|
|
"Oh uh, it seems like i am answering recursively. I will stop now."
|
|
)
|
|
raise ModerationError("Too many recursive messages")
|
|
await chatgpt_process(
|
|
self, msgs, message, api_key, prompt, model, error_call, depth
|
|
)
|
|
else:
|
|
while len(content) != 0:
|
|
if len(content) > 2000:
|
|
await message.channel.send(content[:2000])
|
|
content = content[2000:]
|
|
else:
|
|
await message.channel.send(content)
|
|
content = ""
|
|
|
|
|
|
async def chat_process(self, message):
|
|
if message.author.id == self.bot.user.id:
|
|
return
|
|
|
|
if isinstance(message.channel, discord.DMChannel):
|
|
try:
|
|
curs_data.execute(
|
|
"SELECT * FROM data WHERE guild_id = ?", (message.author.id,)
|
|
)
|
|
except:
|
|
return
|
|
else:
|
|
try:
|
|
curs_data.execute(
|
|
"SELECT * FROM data WHERE guild_id = ?", (message.guild.id,)
|
|
)
|
|
except:
|
|
return
|
|
|
|
data = curs_data.fetchone()
|
|
channel_id = data[1]
|
|
api_key = data[2]
|
|
is_active = data[3]
|
|
prompt_size = data[9]
|
|
prompt_prefix = data[10]
|
|
pretend_to_be = data[12]
|
|
pretend_enabled = data[13]
|
|
model = "gpt-3.5-turbo"
|
|
|
|
try:
|
|
curs_premium.execute(
|
|
"SELECT * FROM data WHERE guild_id = ?", (message.guild.id,)
|
|
)
|
|
except:
|
|
pass
|
|
|
|
try:
|
|
premium = curs_premium.fetchone()[2]
|
|
except:
|
|
premium = 0
|
|
|
|
channels = []
|
|
|
|
try:
|
|
curs_premium.execute(
|
|
"SELECT * FROM channels WHERE guild_id = ?", (message.guild.id,)
|
|
)
|
|
data = curs_premium.fetchone()
|
|
if premium:
|
|
for i in range(1, 6):
|
|
try:
|
|
channels.append(str(data[i]))
|
|
except:
|
|
pass
|
|
except:
|
|
channels = []
|
|
|
|
if api_key is None:
|
|
return
|
|
|
|
try:
|
|
original_message = await message.channel.fetch_message(
|
|
message.reference.message_id
|
|
)
|
|
except:
|
|
original_message = None
|
|
|
|
if original_message != None and original_message.author.id != self.bot.user.id:
|
|
original_message = None
|
|
is_bots_thread = False
|
|
if isinstance(message.channel, discord.Thread):
|
|
if message.channel.owner_id == self.bot.user.id:
|
|
is_bots_thread = True
|
|
if (
|
|
not str(message.channel.id) in channels
|
|
and message.content.find("<@" + str(self.bot.user.id) + ">") == -1
|
|
and original_message == None
|
|
and str(message.channel.id) != str(channel_id)
|
|
and not is_bots_thread
|
|
):
|
|
return
|
|
|
|
# if the bot is not active in this guild we return
|
|
if is_active == 0:
|
|
return
|
|
|
|
# if the message starts with - or // it's a comment and we return
|
|
if is_ignorable(message.content):
|
|
return
|
|
try:
|
|
await message.channel.trigger_typing()
|
|
except:
|
|
pass
|
|
|
|
messages = await fetch_messages_history(
|
|
message.channel, prompt_size, original_message
|
|
)
|
|
|
|
# if the pretend to be feature is enabled, we add the pretend to be text to the prompt
|
|
if pretend_enabled:
|
|
pretend_to_be = (
|
|
f"In this conversation, the assistant pretends to be {pretend_to_be}"
|
|
)
|
|
else:
|
|
pretend_to_be = "" # if the pretend to be feature is disabled, we don't add anything to the prompt
|
|
|
|
if prompt_prefix == None:
|
|
prompt_prefix = (
|
|
"" # if the prompt prefix is not set, we set it to an empty string
|
|
)
|
|
|
|
prompt_path = os.path.abspath(
|
|
os.path.join(os.path.dirname(__file__), f"./prompts/{model}.txt")
|
|
)
|
|
prompt = gpt_3_5_turbo_prompt[:] # copy the prompt but to dnot reference it
|
|
if not isinstance(message.channel, discord.DMChannel):
|
|
prompt = (
|
|
prompt.replace("[prompt-prefix]", prompt_prefix)
|
|
.replace("[server-name]", message.guild.name)
|
|
.replace("[channel-name]", message.channel.name)
|
|
.replace(
|
|
"[date-and-time]",
|
|
datetime.datetime.utcnow().strftime("%d/%m/%Y %H:%M:%S"),
|
|
)
|
|
.replace("[pretend-to-be]", pretend_to_be)
|
|
)
|
|
else:
|
|
prompt = (
|
|
prompt.replace("[prompt-prefix]", prompt_prefix)
|
|
.replace("[server-name]", "DM-channel")
|
|
.replace("[channel-name]", "DM-channel")
|
|
.replace(
|
|
"[date-and-time]",
|
|
datetime.datetime.utcnow().strftime("%d/%m/%Y %H:%M:%S"),
|
|
)
|
|
.replace("[pretend-to-be]", pretend_to_be)
|
|
)
|
|
|
|
async def error_call(error=""):
|
|
try:
|
|
if error != "":
|
|
await message.channel.send(f"An error occured: {error}", delete_after=4)
|
|
await message.channel.trigger_typing()
|
|
except:
|
|
pass
|
|
|
|
emesgs = await prepare_messages(
|
|
self, messages, message, api_key, prompt, error_call
|
|
)
|
|
await chatgpt_process(self, emesgs, message, api_key, prompt, model, error_call)
|