236 lines
8.2 KiB
Python
236 lines
8.2 KiB
Python
import asyncio
|
|
import logging
|
|
import os
|
|
import sqlite3
|
|
from datetime import datetime, timedelta
|
|
from pathlib import Path
|
|
from typing import Dict, Iterable
|
|
|
|
import discord
|
|
import pandas as pd
|
|
from discord.raw_models import RawReactionActionEvent
|
|
from dotenv import load_dotenv
|
|
|
|
LOGGER = logging.getLogger(__name__)
|
|
|
|
|
|
class MsgData:
|
|
msgs: pd.DataFrame
|
|
reactions: pd.DataFrame
|
|
lock: asyncio.Lock
|
|
|
|
@classmethod
|
|
async def create(cls, client: discord.Client, **kwargs):
|
|
self = MsgData()
|
|
self.msgs: pd.DataFrame = await message_df(client, **kwargs)
|
|
self.msgs = self.msgs.sort_values('created')
|
|
self.reactions: pd.DataFrame = await reaction_df(self.msgs['object'].tolist())
|
|
self.lock = asyncio.Lock()
|
|
return self
|
|
|
|
@classmethod
|
|
def from_sql(cls, db):
|
|
if isinstance(db, (str, Path)):
|
|
con = sqlite3.connect(db)
|
|
elif isinstance(db, sqlite3.Connection):
|
|
con = db
|
|
|
|
self = MsgData()
|
|
self.msgs: pd.DataFrame = pd.read_sql('select * from msgs', con=con, index_col='id')
|
|
self.msgs['created'] = self.msgs['created'].apply(pd.to_datetime, utc=True)
|
|
self.reactions: pd.DataFrame = pd.read_sql('select * from reactions', con).set_index(['msg id', 'emoji'])
|
|
return self
|
|
|
|
def to_sql(self, db):
|
|
if isinstance(db, (str, Path)):
|
|
con = sqlite3.connect(db)
|
|
elif isinstance(db, sqlite3.Connection):
|
|
con = db
|
|
|
|
self.msgs.drop('object', axis=1).to_sql(
|
|
name='msgs',
|
|
con=con,
|
|
if_exists='replace',
|
|
index=True,
|
|
index_label=self.msgs.index.name
|
|
)
|
|
self.reactions.to_sql(
|
|
name='reactions',
|
|
con=con,
|
|
if_exists='replace',
|
|
index=True,
|
|
index_label=self.reactions.index.name
|
|
)
|
|
|
|
def __str__(self):
|
|
return str(self.msgs) + '\n\n' + str(self.reactions)
|
|
|
|
async def add_msg(self, message: discord.Message):
|
|
async with self.lock:
|
|
mdict = message_dict(message)
|
|
mdict.pop('id')
|
|
self.msgs.loc[message.id] = pd.Series(mdict)
|
|
LOGGER.info(f'Added message id {message.id} from {message.author}: {message.content}')
|
|
|
|
async def update_reaction(self, client: discord.Client, payload: RawReactionActionEvent):
|
|
if isinstance(payload.emoji, discord.PartialEmoji):
|
|
chan: discord.TextChannel = await client.fetch_channel(channel_id=payload.channel_id)
|
|
msg: discord.Message = await chan.fetch_message(id=payload.message_id)
|
|
idx = (msg.id, payload.emoji.name)
|
|
|
|
for reaction in msg.reactions:
|
|
if isinstance(reaction.emoji, discord.Emoji) and reaction.emoji.name == payload.emoji.name:
|
|
reactions = pd.Series(await reaction_dict(reaction))
|
|
async with self.lock:
|
|
self.reactions.loc[pd.IndexSlice[idx], :] = reactions
|
|
LOGGER.info(f'Added {str(idx)}, {int(self.reactions.loc[pd.IndexSlice[idx], "count"])} total')
|
|
break
|
|
else:
|
|
# only reaches here if the remove action was to take off the last reaction of that type
|
|
if payload.event_type == 'REACTION_REMOVE':
|
|
try:
|
|
async with self.lock:
|
|
self.reactions = self.reactions.drop(idx, axis=0)
|
|
except KeyError as e:
|
|
LOGGER.info(f'{idx} not in index')
|
|
else:
|
|
LOGGER.info(f'Dropped {idx}')
|
|
|
|
def cancellations(self, days: int = 14):
|
|
return cancellations(msg_df=self.msgs, react_df=self.reactions, days=days)
|
|
|
|
def cancellation_totals(self, days):
|
|
return cancelled_totals(cdf=self.cancellations(days=days))
|
|
|
|
def cancellation_leaderboard(self, days, top: int = None):
|
|
df = self.cancellation_totals(days)
|
|
if top is not None:
|
|
df = df.iloc[:top]
|
|
width = max(list(map(lambda s: len(str(s)), df.index.values)))
|
|
res = f'Cancellation totals, past {days} days\n'
|
|
res += '\n'.join(
|
|
f"`{name.ljust(width + 1)}with {row['total']:<2.0f} total`"
|
|
for name, row in df.iterrows()
|
|
)
|
|
return res
|
|
|
|
def worst_offsenses(self, user: str, days: int):
|
|
cdf = self.cancellations(days=days)
|
|
cdf = cdf[cdf['display_name'].str.contains(user, case=False)]
|
|
|
|
if cdf.shape[0] > 0:
|
|
res = f'{user}\'s top 5 cancellations in the last {days} days:\n'
|
|
res += f'\n'.join(
|
|
f'`{row["count"]:<2.0f}cancellations`\n{row["link"]}' for idx, row in cdf.iloc[:5].iterrows())
|
|
else:
|
|
res = f'No cancellations for {user} in the past {days} days'
|
|
|
|
return res
|
|
|
|
|
|
async def message_df(client: discord.Client, **kwargs):
|
|
return pd.DataFrame(
|
|
[message_dict(m) async for m in message_gen(client, **kwargs)]
|
|
).set_index('id').sort_values('created', ascending=False)
|
|
|
|
|
|
async def message_gen(client: discord.Client, limit=20, days: int = 90, **kwargs):
|
|
channels = client.get_all_channels()
|
|
channels = filter(lambda c: isinstance(c, discord.TextChannel), channels)
|
|
channels = filter(lambda c: c.category.name != 'Archive', channels)
|
|
channels = sorted(channels, key=lambda c: (c.category.name, c.name))
|
|
for channel in channels:
|
|
LOGGER.info(f'{channel.category.name} #{channel.name}')
|
|
if 'after' not in kwargs:
|
|
kwargs['after'] = (datetime.today() - timedelta(days=days))
|
|
elif isinstance((after := kwargs.get('after', None)), datetime):
|
|
kwargs['after'] = after.replace(tzinfo=None)
|
|
async for msg in channel.history(limit=limit, **kwargs):
|
|
yield msg
|
|
|
|
|
|
def message_dict(m: discord.Message) -> Dict:
|
|
return {
|
|
'object': m,
|
|
'id': m.id,
|
|
'created': m.created_at.astimezone(),
|
|
'display_name': m.author.display_name,
|
|
'user id': m.author.id,
|
|
'message': m.content,
|
|
'channel': m.channel.name,
|
|
'channel link': f'<#{m.channel.id}>',
|
|
'link': m.jump_url,
|
|
}
|
|
|
|
|
|
async def reaction_df(msgs: Iterable[discord.Message]):
|
|
return pd.concat([await reaction_series(msg) for msg in msgs]).set_index(['msg id', 'emoji'])
|
|
|
|
|
|
async def reaction_series(msg: discord.Message):
|
|
return pd.DataFrame([
|
|
await reaction_dict(r)
|
|
for r in msg.reactions
|
|
if isinstance(r.emoji, discord.Emoji)
|
|
])
|
|
|
|
|
|
async def reaction_dict(r: discord.Reaction) -> Dict:
|
|
return {
|
|
'msg id': r.message.id,
|
|
'emoji': r.emoji.name,
|
|
'emoji id': r.emoji.id,
|
|
'count': int(r.count),
|
|
# 'users': str(list(map(lambda u: u.display_name, (u for u in await r.users().flatten())))),
|
|
}
|
|
|
|
|
|
def cancellations(msg_df, react_df, days: int = 10) -> pd.DataFrame:
|
|
# get reactions with a cancellation emoji
|
|
cancel_reactions = react_df.loc[pd.IndexSlice[:, 'cancelled'], :]
|
|
cancel_msgs = msg_df.loc[cancel_reactions.index.get_level_values(0).to_list()]
|
|
|
|
cancel_msgs['count'] = cancel_msgs.index.to_series().apply(
|
|
lambda idx: cancel_reactions.loc[pd.IndexSlice[idx, 'cancelled'], 'count'])
|
|
|
|
# filter outdated messages
|
|
cancel_msgs = cancel_msgs[cancel_msgs['created'] >= (datetime.today() - timedelta(days=days)).astimezone()]
|
|
|
|
cancel_msgs = cancel_msgs.sort_values('count', ascending=False)
|
|
|
|
return cancel_msgs
|
|
|
|
return cancelled_msgs
|
|
|
|
|
|
def cancelled_totals(cdf: pd.DataFrame) -> pd.DataFrame:
|
|
totals = cdf.groupby('display_name').sum()['count'].sort_values(ascending=False)
|
|
max_channels = (
|
|
cdf
|
|
.groupby(['display_name', 'channel'])
|
|
.sum()['count']
|
|
.sort_values(ascending=False)
|
|
.groupby(level=0)
|
|
.apply(lambda gdf: gdf.idxmax()[1])
|
|
)
|
|
return pd.DataFrame({
|
|
'total': totals,
|
|
'max channel': max_channels,
|
|
# 'worst': cdf.groupby('display_name').max()['link']
|
|
}).sort_values('total', ascending=False)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
client = discord.Client()
|
|
|
|
logging.basicConfig(level=logging.INFO)
|
|
|
|
|
|
@client.event
|
|
async def on_ready():
|
|
print(f'{client.user} has connected to Discord!')
|
|
|
|
|
|
load_dotenv()
|
|
client.run(os.getenv('DISCORD_TOKEN'))
|