Basic Pycord Help

[sage tendon]

does it basically act like a record in java?

(if you know java)

[frail basin]

let me write a quick demo with timeit

This is chatgpt json, not real client data / schema but you get the point
https://paste.sparked.host/obunifinef.rb

[sage tendon]

hm ok

but can you change where that decorator would go

[frail basin]

wdym

[sage tendon]

like this decorator you mean
I've never seen that
can you show that in your code

[frail basin]

@attrs.define?

[sage tendon]

yes

[frail basin]

Oh thats the exact same as @frozen in the test, just with with one of the params set to True

[sage tendon]

yes but can you show it in your code lol

[frail basin]

The paste server has the code

[sage tendon]

oh i didnt see the paste link as that lol

thought it was just yet another image hoster link

[frail basin]

Just a fork of https://github.com/toptal/haste-server

[sage tendon]

hm interesting, basically a bit like a java record class
and i guess it also provides getters and setters? or is that decorator really only to simplify the init?

[frail basin]

It provides extra functionality

Basically an improved version of the builtin dataclasses.define

[sage tendon]

but yea, i guess python shows its ugly side there regarding performance

[frail basin]

I might also try to compile it using CPython, curious how much that improves

[sage tendon]

or use mojo

[frail basin]

Havent heard of that one before

[sage tendon]

i think most of your latency will be unavoidable though

[frail basin]

Yeah, hopefully localhost API calls + DC ping will improve on that though

Would be nice if discord disclosed the locations of their API gateways

Never going to happen, ik

[sage tendon]

pointless info

[frail basin]

Not really

[sage tendon]

very

[frail basin]

Being in the same DC would improve ping

Maybe not that significantly

But would improve nevertheless

[sage tendon]

what

as long as youre at least on the same continent as the gateway, any difference in location is negligible

and as you can see, a single line in your code can add more latency than going half across the world, literally

[haughty grove]

is it possible to get a guild specific banner?

fetching the user returns the normal banner and member.banner is always None

[sage tendon]

not yet, but 2.7 i think

or wait, are you on 2.6.1?

[haughty grove]

yep

[sage tendon]

yea then its in 2.7 when it releases

or wait, user app?

[haughty grove]

nope

[sage tendon]

ah ok

[haughty grove]

okay, thanks

[marsh nymph]

is there a way if a bot joins in a guild with manage roles perms, within the event on_guild_join to edit the self_role ?

[sage tendon]

you cant edit your highest role

[marsh nymph]

i thought maybe theres another way to edit the role of the bot like the color or name

[sage tendon]

no

not sure if bots can edit their role if they have a higher role, might work then

but thats usually not the case

[tiny fractal]

        banr = self.bot.get_guild(BANR_ID)
        await ctx.send(f'{banr.members}')

am I high? this is only returning two members myself and a bot, did something change. this worked for ages

[little cobalt]

.tag get_x

[sly karma]

Any function that starts with get_ in Py-cord is retrieving the related object from your bots cache. If the object is not in the bots cache the get_ method will return None. Because of this behavior you should check if the get_x method is None and if it is use the fetch_x method.

Why Is Using fetch_ Without Using get_ First Bad?
The fetch_ method makes a call to the discord API. This API call is unneeded if you already have the information. It will also make your command take longer because it will have to send and than wait for a response from the discord API. It will also contribute to the discord APIs global rate limit of 50 requests per second.

What Is Cache?
The cache is a temporary storage inside your bot. It holds many objects from members to messages. When you restart your bot the cache will be empty. When the cache is full it will delete older objects to make space for the new objects.

[tiny fractal]

yea its not cache

i've tried both

with fetch it returns zero members

[errant trout]

Do you have members intents

[tiny fractal]

i've got it set to all

actually... you know what.. im an idiot

i forgot the part in bot = discord.Bot(intents=intents) where you pass intents

[sage tendon]

a good IDE should've warned you about that

[tiny fractal]

yea, vscode normally does. but i think its because im remoting to a linux server to do work

[opal swan]

slash_command in logic don't need an help ? bcs all cmds are in slash

[lapis dock]

Please rephrase your question.

[opal swan]

a bot with somes slash commands don't need an help

?

(bcs all commands are show in slash)

[lapis dock]

It is your preference. But I do not use a help command. I just make sure to give each slash command a description

[opal swan]

ye me too

[mossy star]

is it possible to have a custom embed and a link embed (an embed auto-generated by discord from a link) in the same message?

[sage tendon]

yea

i think a message can have 10 embeds

[mossy star]

because when i try, the link embed doesn't show up when i have a custom embed

[sage tendon]

show how it looks like in discord

[mossy star]

this is the result of this command:
r = await c.send(f"https://imgur.com/jTduCVL",embed=discord.Embed(description='custom embed'))

the imgur will load if i remove the other embed, but doesn't if there is

[sage tendon]

ah yea i see the issue

it should work if you send an empty message and edit the link in later

the issue is that you are passing embed= which basically tells discord to only show that embed

[lapis dock]

I am pretty sure that a message has to only have the link for it to render

https://google.com/ extra chars

[sage tendon]

no

[lapis dock]

Nevermind Im dumb

[sage tendon]

what you're thinking of is that it hides image links if the message consists only of the link

[lapis dock]

yeah your right

[mossy star]

time.sleep(1)
y = await c.fetch_message(r.id)
await y.edit(content='https://imgur.com/jTduCVL')```

[sage tendon]

b
https://imgur.com/jTduCVL

probably wont work but can you try doing embeds=[your embed] instead of embed=

you need the brackets btw

[mossy star]

just tried it, no luck

[sage tendon]

then i think you're out of luck
it might work differently because its a media embed and not like, a normal embed
i think it should work with e.g. google.com if you put that in there

can you try that rq

[mossy star]

still using the embeds=[] method

well, it's not too big of a deal if i have to send two messages, but a little annoying

[lapis dock]

If that does not work you can put the link in its own embed by using
Embed.set_image()

[mossy star]

well, the reason i ended up down this rabbit hole was to handle video links, which that can't handle

[sage tendon]

yea i think video embeds are reserved for some sites

[lapis dock]

Ah, ok

[sage tendon]

or at least cant be manually made afaik

[mossy star]

i also tried doing the embed first and then editing a second embed in later, which kinda works but screws up the link embed and strips out videos

[lapis dock]

Yo ucould try this the other way. Edit in your custom embed

[mossy star]

there is technically an embed, but it ends up small and weird with no video

time.sleep(1)
y = await c.fetch_message(r.id)
await y.edit(embeds=[discord.Embed(description='custom embed'), *y.embeds])```

[sage tendon]

this might be very forceful but you MIGHT be able to make videos work with

embed = discord.Embed()
embed.video = discord.EmbedMedia(video_link)

[mossy star]

from my testing, pycord/discord will ignore any video data you try to send in an embed, i've tried

[sage tendon]

f

[errant trout]

users/bots cannot send videos in embeds

[mossy star]

yeah, which is why i'm trying workarounds lmao

even just editing a message with a functional link/video embed will break it

[errant trout]

well.. the only "workarounds" are using gifs or sending a site that embeds videos itself

websites are able to configure video embeds using opengraph

[sage tendon]

proxy it through your own VPS and abuse it as CDN

lmao

that would work but itd probably be unreasonable effort

[mossy star]

basically, the only thing that works in my testing is a regular text message with a link/video embed generated by discord. which means i can't use embed decorators for my text which is annoying, so that's why i figured i'd ask

anyways, thanks

[lament totem]

How does autosharding work? Does it spawn multiple processes and handle ipc or does it still use a single process but use multiple gateway connections? This is a bit confusing

[little cobalt]

Do you really need it?

[lament totem]

I have a bot with 2300~ servers, I feel like I should be thinking about sharding soon

[lament totem]

Would it be wrong to ask without needing it?

[lofty parcel]

It is multiple gateway connections.

At least what it says in the source code.

"Instead of a single websocket, we have multiples. The key is the shard_id"

[little cobalt]

nothing is wrong about asking ^^

[lapis dock]

We have some people who come here wanting to start sharding at 150 servers so we want to make sure 🙃

[sage tendon]

i think its forced at 2500 so yes you should do it

[lament totem]

I see, that's understandable 😛

I'm just trying to understand this, so for example if I do autoshardedbot.guilds will I get all the guilds where the bot is installed? or all the guilds corresponding to the shards in the current process being run?

[sage tendon]

it will show all the bot's guilds

the only real difference is that not all guilds run on the same shard / websocket connection

[round heart]

Mine is in two servers. How can I spawn 10 shards?

[lapis dock]

1. add your bot to about 10k more servers
2. Use autosharded bot
   easy :)

[lament totem]

Even if for example I run two processes each with its own subset of shards?

[sage tendon]

you can do that but theres little point if its on the same machine

[lofty parcel]

Why would you run two processes tho

The point of autoshardedbot is to avoid you the frustration of handling IPC

[lament totem]

My bot is not shared rn and sits at 30% single core utilization all the time. I'm thinking I'll need more processes to avoid single process constraints.

Later on ofc, assuming the bot gets installed in more and more servers

If I understand, it's still a single process but receiving guild events through multiple gateway sockets, no?

[sage tendon]

30% is nothing

[lament totem]

Doesn't 30% mean that I can only handle 3~ times the current bot usage/activity on a single process at most?

[sage tendon]

No

And you arent suddenly getting 3x as many servers to handle

[lament totem]

Agreed, I'm just thinking about the future

[lament totem]

so every shard gets a new process? I feel like there's something I'm missing

[sage tendon]

no lol

its just a new gateway connection

[lament totem]

Then why is this not true? I'm sorry I just dont understant lol, thank you for your patience with me

[sage tendon]

not 100% of your CPU usage is variable

there'll be some % that remain even if you have 0 servers in your bot

[lament totem]

I think all of it is related to event handling in my case, when on my testing environment it stays at 0% on avg.

[sage tendon]

yea okay then i overestimated it

but still, 30% is nothing, you'll have long until you reach high percentages still

[lament totem]

I agree, I've done a lot to keep it as low as possible but definitely want to avoid it getting high so that it can stay responsive

but I think I understand a bit better now, I guess I should give the source code a good look sometime soon

[sage tendon]

it wont become unresponsive until it reaches literally 100%

[lament totem]

I guess what I meant was, like being late to defer or respond to interactions (buttons / slash commands, etc) not really unresponsive in the strict sense

[sage tendon]

same thing there

its not slower at 95% CPU usage than it is at 5%

well, it is, but thats about a few nanoseconds at best

[lament totem]

If that's the case then I must be doing something wrong 👀

[sage tendon]

why?

[lament totem]

well if the bot is being late here and there on defering stuff then something must be blocking for longer than 3 secs or its just too slow to do the stuff ahead in the queue

[sage tendon]

yea, but thats unrelated to CPU usage

thats just because you have blocking code that adds up

may or may not be unavoidable

[agile oar]

Is there no way to handle a ratelimit on channel edits? I know the limit it 2 edits per 10 minutes, but im not getting any errors raised on a try except.

How am I supposed to know if the bot is ratelimited on editing a channel before trying an edit if it does not return anything?

[lapis dock]

In general the library will handle rate limits like that. With how discord is set up you cannot check if you are going to hit a rate limit. You have to hit it and then wait the timeout.
When rate limited the library will queue your request to be sent when unratelimiter.

[stuck oar]

I understand that to add tags after creating a thread in a forum channel, I should use ForumChannel#edit(applied_tags=[]), but I don't know what values to put in discord.ForumTag. I tried using discord.ForumTag(name="Plugin", emoji="🌙"), but it doesn't work (even though it's an actual tag).

[errant trout]

you need to get the tag by ID; either check through ForumChannel.available_tags, or use ForumChannel.get_tag if you know the id

[stuck oar]

oh Thanks!!

[agile oar]

Hm okay so best way is to just count channel edits per channel id then

[frail basin]

Is there a way to get the message that a view is attached to, from within the view's code? self.message is not None, but raises an Unknown message error when I try editing it, even though I can still clearly see it on my client and even use the view's buttons.

[little cobalt]

which intents does your bot have?

[frail basin]

all

[little cobalt]

Can I see it? ;3

[fresh sierra]

Defer is the original response

As far as I know

Try again without the defer

[frail basin]

did not work

[sage tendon]

is your view in an ephemeral message?

[frail basin]

Yes

[sage tendon]

that is the problem

[frail basin]

Any workarounds?

[sage tendon]

dont think so
just do an ephemeral followup instead of editing it

[frail basin]

That wont prevent the original issue, of the user pressing the buttons multiple times

[sage tendon]

then disable the buttons once clicked

[frail basin]

And since I dont have the original message, I cant disable the button

[sage tendon]

self.disable_all_items()

[frail basin]

Dont I have to edit the message with view=self for that?

[sage tendon]

not sure, p sure if you disable all items itll still reject any interactions sent to them

let me check my code

[frail basin]

Well yea it will reject them, causing an "interaction failed" thing

iirc to disable the buttons you have to edit the message

[sage tendon]

yea, i dont think you can properly disable the buttons in an ephemeral message

[frail basin]

Well thats dumb

[sage tendon]

i mean, discord limitation
ephemeral messages are really barely messages at all lol

[frail basin]

I am aware that this is discord's bs

[sage tendon]

actually lemme check smth

okay forget what i said, wtf
I just remembered a bot i use has ephemeral messages that can be edited

sorry lol

but i have no idea how that's done

have you tried editing both self.message (by passing it from the original view) and self.original_response() with ephemeral=True passed?

[frail basin]

I fixed it

not sure how lol

I am not touching this

seems like its actually editing it though

[sage tendon]

hm

using the top or bottom path

[frail basin]

top

luckily

[sage tendon]

but didnt you do exactly that before

[frail basin]

Yes.. Yes I did

Oh

You know what I changed? Removed ephemeral=True from the defer

But the message is still somehow ephemeral

So I dont know wtf is going on

[echo wraith]

wat

[frail basin]

Exactly 😂

[sage tendon]

thats not even possible

if you respond non-ephemerally the response cant be ephemeral

[frail basin]

Oh I know why,

Slashcommand -> ephemeral view -> non ephemeral defer -> ephemeral edit_original_response

[sage tendon]

wow

[frail basin]

And it stays ephemeral all the way

[errant trout]

you can

[sage tendon]

yea i had a brain lag

we figured it out now, somewhat

[errant trout]

ephemeral messages are full messages (to the point they appear on different clients)

[sage tendon]

but explicitly passing ephemeral=True to the defer shouldnt break it lol, since its all forced ephemeral anyway apparently

[frail basin]

Not sure why it did that. I bet if I passed it now, it wouldn't break.

[errant trout]

this should have been interaction.response.edit_message

[frail basin]

Whats the difference between that and edit_original_response?

[sage tendon]

this is why i hate views, honestly

way too much shit that sounds samey

[errant trout]

response.edit_message is for the initial interaction

[frail basin]

so thats basically interaction.message.edit()?

[errant trout]

edit_original_response is for editing the response you already made later

(and again, we implemented interaction.edit to simplify it)

more or less yeah

[frail basin]

How would I make the 3rd message replace the 2nd one then? The 2nd message's "send" button triggers the 3rd message.

I tried:

• interaction.response.edit_message -> raises InteractionResponded (I'd assume cause the defer?)

• interaction.edit -> makes a 3rd ephemeral message

• interaction.edit_original_response -> same as above

• interaction.message.edit -> Unknown Message

• interaction.followup.edit_message(message_id=interaction.message.id, ...) -> Nothing happens

Here is the relevant code: https://paste.sparked.host/owujovapit.rust

[errant trout]

defer with invisible=True (which is default....) -> interaction.edit, you've overcomplicated it

if you defer with invisble=False, discord explicitly creates a new message

[frail basin]

ah

That was the missing piece of info here

Thanks

[errant trout]

allgood

[round heart]

wait, what? What takes an invisible kwarg?

[sage tendon]

defer of a button interaction

doesnt show the thinky dots

[round heart]

Ahhhhh. Do the dots automatically disable the control (button), though? That could be helpful.

[sage tendon]

wdym?

[round heart]

You say it's used when receiving a button press. The discussion above was about disabling the button once pressed (which it doesn't generally do). So I'm wondering if not having it 'invisible' brings that behavior with it.

[sage tendon]

no clue

[frail basin]

To my understanding, invisible is True, you cannot interaction.edit() (disable the button), because it forces a new ephemeral message

[errant trout]

opposite, but yeah

invisible was introduced so you could respond to components without having to make a new message

i.e. before that, you ALWAYS had to respond to them with a new message (but this was very early on)

on discord's end, this is referred to as DEFERRED_UPDATE_MESSAGE - the library defaults to this behavior because, for the vast majority of use cases, it's more useful

[full ore]

Hello,

I often have issues with Discord rate limits when programming a bot. I have to restart it frequently to apply changes, and I probably restart it five times or more within 10 minutes.

Here’s my GitHub link to the bot:
https://github.com/DerTarzan/AuraCity

Can someone explain to me why I keep getting rate limited? Is there a way to prevent this, as it's nearly impossible to program a bot under these conditions?

[sage tendon]

you still didnt send your ratelimit message

[full ore]

I have no one but my commands dont sync...

in the morning yestady i have a message like this:

2024-09-28 15:05:48,363 - AuraCityBot - ERROR - ❌ Failed to sync commands: 429 Too Many Requests (error code: 30034): Max number of daily application command creates has been reached (200)

[sage tendon]

yea, dont edit your command metadata so often

names, descriptions, options

[full ore]

explain please i dont unterstand

[sage tendon]

dont edit this of your commands so often, thats it

the limit is 200 commands per day, impressive that you managed to reach that lol

[full ore]

But i dont have 200 Commands i have only 5 or 7 idk

[sage tendon]

yea, but it adds up whenever you edit their data

[full ore]

could you send or explain me a example i dont unterstandt this

[sage tendon]

just dont edit the names, options and descriptions of your commands so often

thats all there is to it

[full ore]

ahh okay

But i have to look if my changes work...

yk ?

[sage tendon]

the changes of the command functions doesn't count, i never mentioned that

[full ore]

    @slash_command(name="abmeldung", description="/abmeldung - Melde dich ab von AuraCity-Team.")
    async def deregistration(self, ctx: discord.ApplicationContext):
        user_id = ctx.author.id

        # Fetch the last deregistration timestamp from the database
        last_deregistration = await self.database.get_last_deregistration(user_id)

        # Ensure we get a datetime object from the query result
        if last_deregistration:
            # Since the result is a tuple, extract the first element (the datetime)
            last_deregistration_time = last_deregistration[0] if isinstance(last_deregistration, tuple) else last_deregistration

            # Check if last_deregistration_time is a string and convert it to datetime if necessary
            if isinstance(last_deregistration_time, str):
                last_deregistration_time = datetime.strptime(last_deregistration_time, "%Y-%m-%d %H:%M:%S.%f")

            # Ensure last_deregistration_time is timezone-aware
            if last_deregistration_time.tzinfo is None:
                last_deregistration_time = last_deregistration_time.replace(tzinfo=timezone.utc)

            # Check if the user has deregistered within the last 48 hours
            time_since_last = (discord.utils.utcnow() - last_deregistration_time).total_seconds()
            if time_since_last < 48 * 3600:  # Allowing deregistration every 48 hours
                await ctx.respond("Du kannst dich nur alle 48 Stunden abmelden.", ephemeral=True)
                return

        # Update the last deregistration timestamp in the database
        await self.database.add_deregistration(user_id, discord.utils.utcnow())
        await ctx.respond("Du hast dich erfolgreich abgemeldet.", ephemeral=True)

[full ore]

This is my command to derigster... i have to show if it still work

[sage tendon]

again, code changes of the function itself don't count

[full ore]

Sorry i know but i try to understand this but i dont unterstand...

[sage tendon]

and if you dont somehow literally edit the options, names or descriptions of your commands 200 times before hitting the ratelimit you are doing something weird
show your main bot file

[full ore]

yeah wait i have to edit...

[sage tendon]

yea, this is the issue

dont do that

or this in your on ready

[full ore]

but whyyy when i dont do then dont sync my command -> my changes dont work....

[sage tendon]

and you cannot load cogs in your on_ready, it doesn't do anythig unless you have prefix commands

[sage tendon]

syncing your commands has nothing to do with seeing changes in your code. if you mean commands not showing up at all etc, you need to restart your discord

[full ore]

where should I load them then

[sage tendon]

before on ready

lol

before your bot start

[full ore]

yeahhh this is my problem they dont showing up

work ctrl + r in my case ?

[sage tendon]

yea

[full ore]

HAHHA sorry i still try to unterstand don't take it badly

[sage tendon]

where did you get that code from btw

thats most certainly not from the pycord guide

[full ore]

which code do you mean

[sage tendon]

all of it

[full ore]

I have write this...

this my bot im currently working on

[sage tendon]

yea but you must've gotten the general concept of how to load cogs etc from somewhere

[full ore]

okay wait i read this

cogs_list = [
    'greetings',
    'moderation',
    'fun',
    'owner'
]

for cog in cogs_list:
    bot.load_extension(f'cogs.{cog}')

you mean this ?

[sage tendon]

you have to do that before running bot.run, yes

[full ore]

def load_cogs(self, directory: str, is_root: bool = True) -> None:
        """Loads all cogs from the specified directory."""
        if is_root:
            logger.info(":package: Loading Cogs...")

        for filename in os.listdir(directory):
            if os.path.isdir(f'{directory}/{filename}'):
                self.load_cogs(f'{directory}/{filename}', is_root=False)
            elif filename.endswith('.py'):
                try:
                    self.load_extension(f'{directory.replace("/", ".")}.{filename[:-3]}')
                    logger.info(f' - :white_check_mark: Loaded Cog: {directory}/{filename}')
                except Exception as e:
                    logger.error(f':x: Failed to load cog {filename}: {e}')
        if is_root:
            logger.info(":tada: All Cogs Loaded Successfully.")

But this is the same just more complicated

[sage tendon]

yea but also the command syncing stuff etc
thats not in the guide and you dont typically use that

[full ore]

Project Structure:

base -> cogs -> events.py

I also have:

base -> cogs -> TicketSystem -> ticket.py

yk ??

[full ore]

okay my bad thank you very much...

[sage tendon]

i still dont know where you got all that stuff from lol

[full ore]

my braiiinnnn

my brain tell me write this... and i sayd yes it makes sence...

[sage tendon]

you didnt just randomly think all of that up.... where did you get it from, the guide, the docs, stackoverflow?

[full ore]

idk i programm since 4 years that all form the years i programm.. yk

sometimes i use ChatGPT or i read the py-cord guide only for idk

see what discord.Member have for attributes...

[sage tendon]

alright

just remove any manual command syncing

its unnecessary

[full ore]

its only for global bots like mee6 or rythm ?

[sage tendon]

i dont see why those would inherently need it either

and with how many commands they have, they probably couldn't

[full ore]

okay wait i make my changes i hope im not currently ratelimitet...

Today i have to wait for 5 hours

yeah my commands dont show up i have restart my discord...

Bot says cogs been loaded outsite the on_ready func...

[sage tendon]

show your main file and also one of your cogs

[full ore]

okay wait

from dis import disco

import discord

from base.bot import AuraCityBot
from base.logger import AuraCityLogger

logger = AuraCityLogger("AuraCityMain").get_logger()

class AuraCity(AuraCityBot):
    def __init__(self):
        super().__init__()
        self.config.DEV_MODE = True

    @staticmethod
    def setup():
        if os.name == "nt":
            pass
        elif os.name == "posix":
            pass

    def start_bot(self):
        try:
            self.load_cogs("base/cogs")
            self.run(self.config.TOKEN)
        except discord.LoginFailure:
            token = self.config.TOKEN
            if len(token) != 59:
                logger.error("Ungültiger Token. Bitte überprüfe den Token in der Konfiguration.")
                return
            logger.error("Fehler beim Einloggen. Bitte überprüfe den Token in der Konfiguration.")

if __name__ == "__main__":
    AuraCity.setup()
    aura_city = AuraCity()
    aura_city.start_bot()

[sage tendon]

how do you even have so much command code without ever being able to test it lol

[full ore]

what do you mean

class Moderation(commands.Cog):
    def __init__(self, bot):
        self.bot = bot
        self.config: AuraCityBotConfig = bot.config
        self.embeds = AuraCityEmbeds()

    @has_permissions(administrator=True)
    @slash_command(name="clear", description="Lösche eine bestimmte Anzahl von Nachrichten.")
    async def clear(self, ctx: discord.ApplicationContext, amount: int = 1):
        await ctx.channel.purge(limit=amount + 1, check=lambda m: not m.pinned)
        embed = self.embeds.clear_embed(amount, ctx.guild.icon.url)
        await ctx.respond(embed=embed, delete_after=5)

    @has_permissions(administrator=True)
    @slash_command(name="sync", description="Synchronisiere die Befehle mit Discord.")
    async def sync(self, ctx: discord.ApplicationContext):
        await ctx.defer()
        await self.bot.sync_commands()
        embed = self.embeds.sync_embed(ctx.guild.icon.url, [c.name for c in self.bot.commands])
        await ctx.respond(embed=embed, delete_after=5)

    @clear.error
    async def clear_error(self, ctx: discord.ApplicationContext, error: commands.CommandError):
        channel = self.bot.get_channel(self.config.ERROR_LOGS_CHANNEL_ID)
        embed = self.embeds.clear_error_embed(ctx.guild.icon.url, str(error))
        await ctx.respond(embed=embed)
        await channel.send(f"Failed to clear messages. [{error}]")

    @sync.error
    async def sync_error(self, ctx: discord.ApplicationContext, error: commands.CommandError):
        channel = self.bot.get_channel(self.config.ERROR_LOGS_CHANNEL_ID)
        embed = self.embeds.sync_error_embed(ctx.guild.icon.url, str(error))
        await ctx.respond(embed=embed)
        await channel.send(f"Failed to sync commands. [{error}]")

def setup(bot):
    bot.add_cog(Moderation(bot))`

[sage tendon]

did your commands ever show up?

[full ore]

yeah when i have sync them#

[little cobalt]

why is the permission at the top of the command?

[sage tendon]

has_permissions shouldnt be used for slash commands anyway but thats not the issue rn

[full ore]

why not... i this a problem for py-cord `?

[sage tendon]

it doesnt work.

[full ore]

but it still work

i have test it

wait

ahh i cant show you my commands dont show up HHAA i forget it

[sage tendon]

and i mean, you are out of luck for today anyway, given this message
try again tomorrow

[full ore]

So they worked 2 hours ago after I got the error

idk why

so my commands were visible

[sage tendon]

because a syncing error doesnt magically unsync your commands
im kinda confused why they dont show up at all now but

[full ore]

idk i cant tell you

everything else goes

like on_message event

[sage tendon]

https://github.com/Pycord-Development/pycord/blob/master/examples/app_commands/slash_perms.py also, look at this for proper slash command permission setting

[full ore]

thank you

omg my commands show up ahhhhhhh

thank you, you are really good guy...

[solid sapphire]

hey,
I've discovered that if a bot sends a message with a link, the link gets a preview
but if it sends exactly the same message but with an embed attached, the link doesn't get a preview
is there a way to send a message with a link with both link preview and embed?

[sage tendon]

dont think you can change that

[solid sapphire]

okay, thanks

[civic moon]

I'm have trouble getting an event to fire when someone leaves the guild. Is there something I'm missing? This code is not executing:

async def on_raw_member_remove(payload):
    print(payload.user.name + " left: on_raw_member_remove")

async def on_member_remove(member):
    print(member.name + " left: on_member_remove")

My intents are active (on_member_update executes fine). I am seeing another bot (ProBot) activate when a member leaves the guild. No way it would absorb the event right? Btw I may have handled the payload incorrectly, I don't have any errors though.

[sage tendon]

what are your intents?

[civic moon]

intents=discord.Intents.all()
with all intents active on the developer portal

[lapis dock]

You do not have the listener decorator

@bot.listen()

[civic moon]

Omg you are right thank you 🤦‍♂️

[lapis dock]

:)

[civic moon]

Do you know if that payload is handled correctly for the on_raw_member_remove?

[errant trout]

looks fine

[lapis dock]

This is the docs for they payload for that event.
https://docs.pycord.dev/en/stable/api/models.html#discord.RawMemberRemoveEvent

[red mist]

Did I get pin

*ping?

[twilit socket]

ban in Member.ban is maximum of 7 days of message deletion? is this an API call for all messages? I want to delete a large history from years ago, do I really need to fetch each message with API calls?

[lofty parcel]

The max being 7 days is an api limitation

And I wouldn't recommend you fetching and deleting the messages one by one, you'll get ratelimited very quickly if they're actually too many

[twilit socket]

I need to wipe my own messages from my own server

what do you recommend?

I can't delete the channels

[lapis dock]

I dont think that is a valid usecase of the API
You can request that discord deletes all of your data (including your account and messages on all servers) but i doubt this is what you want

[twilit socket]

I would want that, but I don't want to lose my bots/apps

I want to get off discord

but my bots needs to be kept running

[lapis dock]

You can transfer the bots ownership to another user

If they are verified you will have to contact support to do this

[twilit socket]

there's an option on my account to just disable it, not delete

disabling it would hide my messages or something?

[lapis dock]

You can disable your account but I dont think this will delete messages

[twilit socket]

what a tricky situation

[lapis dock]

It does not make a whole lot of sense why you want to get rid of your account but still have bots

[twilit socket]

it has been said that i'd be rate limited, but what are the numbers? I've downloaded messages before in the millions, and it's been pretty fast, but not deleted it

deleting it is an api call by itself

[twilit socket]

dont want to be on the internet anymore, personal reasons. My bot helps a lot of people and I don't want to just kill it

[lapis dock]

Discord does not post what numbers cause rate limits. It depends on a case by case basis. In generally the library should handle the rate limits for you but I am unsure if it would be able to with the quantity of requests. There is a hard limit of 50 requests per second though.

My recommendation is to transfer the bots to a trusted person or a account that you will not use. And then request discord deletes your data.

In my opinion deleting your messages is not worth it. You have already sent them so deleting them will not make much of a difference.

[twilit socket]

so I can do it slowly

does'nt need to be in seconds

I can wait all day

I'll back it up anyways

I'll first download it

thanks for all the help

[errant trout]

what country do you live in

i'd just go with what wolfy said, transferring your bot to a team under another account and then requesting discord to wipe and anonymize your data would be significantly easier

[echo wraith]

https://youtu.be/g5FbRfwMEuo
If it's your own messages

[lofty parcel]

The clickbait 💀

[sage tendon]

NTTS is pure misinformation in that disregard

its high level clickbait to lure in kids

[echo wraith]

I agree 100% but that video was actually worth it i wouldn't have shared it else

removed the embed

[deft kestrel]

discord.http.Route.BASE = ""
client = discord.Client()

This works on pycord?

[errant trout]

well.. what exactly are you trying to do?

i can see what it's doing in dpy... but i fail to understand why you'd ever want to

if you mean to change the api version then you modify discord.http.API_VERSION, which is an int, but note that if you change the version from the default then it may be harder to debug certain issues
most of the time people only want this to force v6 for message content... but i don't really think it's worth it considering how old v6 is

[edgy nest]

doesnt message content still not work anyways if you dont have the intent

[delicate wedge]

I am trying now and I only did slash cmds so far but no prefix and I can't get the to work.

@bot.command()
async def test(ctx):
    await ctx.send("Returned Test")

bot = discord.Bot(
    intents=intents,
    debug_guilds=[idthatIwontshow],
    status=status,
    activity=activity,
    command_prefix="."
    )

intents = discord.Intents.default()
intents.members = True

import discord
import os
from dotenv import load_dotenv
from ezcord import commands
```![criii](https://cdn.discordapp.com/emojis/1217764089078611999.webp?size=128 "criii")

[little cobalt]

stay in one channel pls

[deft kestrel]

Change API backend

[errant trout]

i think v6 bypassed verification for it or something, i forget

[round heart]

Btw load_extension has recursive=True so you don’t have to explicitly set each name or do your own iteration

https://docs.pycord.dev/en/stable/api/clients.html#discord.Bot.load_extension

[loud kayak]

Why is it that channel.send (sometimes) sends multiple messages when only being fired once? Is it just an API bug?

It’s usually in button callbacks or timers for me

[proven valley]

is there some guide on testing, so I can auto test commands and views?

[sage tendon]

Not that i know of
typically discord bots arent so complex that you cant quickly test a command yourself in 2 minutes

[full ore]

Does it make a difference if I use recursive=True or if I do it the way I see it picture:

[sage tendon]

i just dont get why people do it like that

loading cogs is a single line if you do it the normal way

[round heart]

I mean.. I guess not, but you're saving yourself a bunch of code duplication by using the built-in method.

[proven valley]

true but im lazy and the specific thing that needs testing is something that sends a view to 4 peoples dms and they have to interact with it so that I can see if the correct information and everything was sent.

[sage tendon]

i mean just send / duplicate all those messages to a channel instead ig

or your own dms

[lapis dock]

There was a pycord unit test project at one point but I think it kinda died out

[sage tendon]

it really isnt hugely feasible given you entirely rely on a foreign API

[torn barn]

You dont seem to have message content intent enabled

[lapis dock]

They are also using discord.Bot instead of ext.commands.Bot so prefix commands will not work anyways

[sage tendon]

already solved in a thread made by them later

[delicate wedge]

Yes

[delicate wedge]

Im a different server

[sage tendon]

here.

so that means you made minimum three threads for one issue reading the guide would've prevented

[loud kayak]

Why is it that channel.send (sometimes) sends multiple messages when only being fired once? Is it just an API bug? It’s usually in button callbacks or timers for me

[fresh sierra]

is there a built in way to do that ?

    async def flip(self, ctx: LumabotContext, bet: int | "all"):

[frail basin]

Do what exactly?

[fresh sierra]

bet: int | "all"):

set the type as int or a specific text

[frail basin]

If you are talking about slash command options, I doubt you can.

[fresh sierra]

dont need to block the user from entering but if it does it raise an error bad argument like in prefix command

[frail basin]

I dont think it will allow you to mix int and string

[fresh sierra]

its ok then i will make a custom converter

thanks for the help

[sage tendon]

tf are you trying

[fresh sierra]

I mean u can see it

For the user has to put either an int either all in str

[sage tendon]

so take a string, check if it says "all", if no, cast it to int

[fresh sierra]

Yeah but then I have to make at least 4 lignes each time

So u just made a custom converter so its easier

[sage tendon]

which is how many lines?

you keep overcomplicating absolutely everything you do once you hit anything you can't solve with the most simple things
and from what I saw it makes your code extremely overcomplicated and hard to maintain

[fresh sierra]

How a custom converter is hard to maintain 🤣

[sage tendon]

It's the sum of everything

the fuck do you need a custom converter for? it's literally 2 lines if you do it compactly

[fresh sierra]

Because I will not put try: int except raise for 10 commandes

Im just doing the type hint

[sage tendon]

wtf do you need to raise an exception for

[fresh sierra]

For a bad argument

[sage tendon]

take str
check if Str says all
if no, cast to int

[fresh sierra]

And if the user put neither ?

[sage tendon]

then for all I know the cast will be 0

[fresh sierra]

So I would have to handle that

[sage tendon]

you should already check for 0 being entered

[fresh sierra]

And to say the user he should put either all either a number

Or I use my custom converter

That does everything for me

[sage tendon]

mess up your codebase piece by piece, sure

why did you even ask if you already have the solution?

[fresh sierra]

True that does
bet: BetConverter

Its soooo messy

Thats sad

[frail basin]

Why not just use -1 instead of "all" ? 😂

[sage tendon]

no that's not really intuitive at all

all betting bots I know let you enter all

[sage tendon]

youve never programmed a bigger program in something like java, right

[fresh sierra]

Is that messy ?

What a messy way to handle the typehint of a command

It’s so more logic to add the same thing at the beginning of all command

[sage tendon]

and it makes your code less clear, is way longer, and introduces additional complexity as opposed to doing it exactly where you need it or even just moving it into a seperate function if youre lazy

[fresh sierra]

How is that less clear ?

[sage tendon]

again

[fresh sierra]

So ur saying adding the same function at the beginning of each command is more clear and shorter that using a custom converter ?

[fresh sierra]

Tbh I don’t see any link with the situation

[sage tendon]

That is unfortunate

good luck

[fresh sierra]

I mean I prefer to listen to danny than to listen to make making the same function again and again inside each command

[sage tendon]

and instead you're doing the same but in an unexpected code without even documenting it meaning any outside person reading your code can't intuitively know what the fuck that even does

[fresh sierra]

Why the fuck would I care about reading my code ?

[sage tendon]

I feel like sometimes you're compensating for lack of something with your overly complicated code for literally the most simple things ever

[frail basin]

You are frequently asking questions here, thats why. lol

[fresh sierra]

And I had been always to do everything that I need

Also if I want to make my converter more clear to the user I can use comment

I don’t see how a function inside the command that will do the same thing is more clear

[sage tendon]

Because then you have a properly named function that ideally already describes what it does in the name, you leave the typehint proper, etc

[fresh sierra]

Bet converter is not a good name ?

I mean it convert the bet option so

[sage tendon]

That is a substantive for a name, functions are named by actions using verbs.

[fresh sierra]

Yeah but that’s not a function but a class

So I understand ur point

I understand that ur point work if u are many user using this code

But since im the only one I don’t teallly need to put comment explaining what it does since I know what I just code

But using a custom converter is still more appropriate that make a fonction after since you are converting the arg

Also what danny make a guide using custom converter rather thzn runction inside the command

[errant trout]

if you're working with converters, i highly recommend reading through https://docs.pycord.dev/en/stable/ext/commands/commands.html#converters since it has a lot of interesting stuff (not all of this works with slash, but the important stuff should be fine)

[fresh sierra]

Yep I did use that to make my converter

[errant trout]

nice

[fresh sierra]

Yep here it is

But if you have advice tell me ;)

[errant trout]

nah looks good to me

[fresh sierra]

Ok thanks

[wheat field]

how do i stop a await ctx.respond if a if statment comes back true?

if condition?

    async def warn(self, ctx, user: Option(discord.Member, "User to warn"), reason: Option(str, "Reason for the warning")):
        # Check if the bot has permission to manage messages
        if not ctx.guild.me.guild_permissions.manage_messages:
            timeout_on_bot = discord.Embed(title="Error ", description=f"I don't have permission to warn members.", color=discord.Color.blue())
            await ctx.respond(embed=timeout_on_bot, ephemeral=True)
            return

        # Check if the user has permission to manage messages
        if not ctx.author.guild_permissions.manage_messages:
            timeout_on_bot = discord.Embed(title="Error ", description=f"You don't have permission to warn members.", color=discord.Color.blue())
            await ctx.respond(embed=timeout_on_bot, ephemeral=True)
            return
        
        if user == ctx.bot.user:
            embed = discord.Embed(title="Error", description="You cannot warn the bot")
            await ctx.respond(embed=embed)
            return
        ```

[little cobalt]

why?

[wheat field]

this my maint warn command but i wanna check if the user is == the bots user name and if it is send embed saying it is but i need to stop the the bot from editint database info

        cursor.execute("SELECT warnings FROM warnings WHERE user_id = %s", (user.id,))
        result = cursor.fetchone()
        if result is None:
            await self.create_warning(user)
        else:
            warnings = result[0]
            warnings += 1
            cursor.execute("UPDATE warnings SET warnings = %s WHERE user_id = %s", (warnings, user.id))
            self.cnx.commit()
            if warnings > 20:
                await user.kick(reason="Exceeded warning limit")
                embed = discord.Embed(title="User Kicked", description=f"{user.mention} has been kicked for exceeding the warning limit", color=discord.Color.blue())
                await ctx.respond(embed=embed)
            else:
                embed = discord.Embed(title="Warning", description=f"{user.mention} has been warned for {reason}", color=discord.Color.blue())
                await ctx.respond(embed=embed)```

and stop it from saying warned user wtv

i swear i did it on another bot i just cant rember 😭

could i use difer?

[little cobalt]

for what?

[wheat field]

            embed = discord.Embed(title="Error", description="You cannot warn the bot")
            await ctx.respond(embed=embed)
            return``` this 

```else:
                embed = discord.Embed(title="Warning", description=f"{user.mention} has been warned for {reason}", color=discord.Color.blue())
                await ctx.respond(embed=embed)```

the if statment is checking if user is the bot name and if it is ill send the embed but how do i stop the bot from adding a warn

i have alr have return

heres little better picture

[wheat field]

to the db

i also wanna comepletly that stop the respond

[frail basin]

I dont see what the issue is here?

[wheat field]

no issue

permison checks n shit

when i try to warn my bot right

it will send the failed

and it will send the warn added

[frail basin]

If you return after an if check is true, it wont execute the db stuff

[wheat field]

not true

works perfectly fine for everyother comamnd

[frail basin]

The code itself looks fine

[wheat field]

yes ik the code is fine

[frail basin]

Are you sure that the bot is running this code?

[wheat field]

theres no visibal errors

but i wanna add error handling as embeds

[little cobalt]

Keep in mind if you do a if if if its gonna do all the stuff

[wheat field]

oh shit

i didnt even know that

[hazy turret]

Is there an easy way to find out which user has invited another user to the server?

[wheat field]

async def on_member_join(member):
    invites_before_join = await member.guild.invites()
    invites_after_join = await member.guild.invites()
    for invite in invites_before_join:
        if invite.uses < [i.uses for i in invites_after_join if i.code == invite.code][0]:
            inviter = invite.inviter
            await member.send(f"You were invited by {inviter.mention}")
            break```

?

@hazy turret like this?

[fresh sierra]

how that supposed to work ?

since ur comparing 2 times the same thing

but you can use a database to save each time the invite and they usage, then you can compare

[wheat field]

gonna need a db like he said but u can use mysql

[wheat field]

from discord.ext import commands
import mysql.connector

class InviteTracker(commands.Cog):
    def __init__(self, bot):
        self.bot = bot
        self.cnx = mysql.connector.connect(
            user=
            password=
            host=
            port=
            database=
        )

    async def create_invite_event(self, invite, invitee):
        try:
            cursor = self.cnx.cursor()
            cursor.execute("INSERT INTO invite_events (invite_code, inviter_id, invitee_id, invite_date) VALUES (%s, %s, %s, %s)", (invite.code, invite.inviter.id, invitee.id, invitee.joined_at))
            self.cnx.commit()
        except mysql.connector.Error as err:
            print(f"Error: {err}")


            
    async def create_invite_on_duplicate(self, invite):
        try:
            cursor = self.cnx.cursor()
            cursor.execute("", (invite.code, invite.inviter.id))
            self.cnx.commit()
        except mysql.connector.Error as err:
            print(f"Error: {err}")
            
            
    @commands.Cog.listener()
    async def on_member_join(self, member):
        try:
            invites = await member.guild.invites()
            for invite in invites:
                if invite.uses > 0:
                    inviter = invite.inviter
                    await self.create_invite_event(invite, member)
                    await member.send(f"You were invited by {inviter.mention}")
                    embed = discord.Embed(title="Invite Tracker", description=f"{member.mention} was invited by {inviter.mention}", color=discord.Color.blue())
                    await self.bot.get_channel(1271983219230900224).send(embed=embed)
                    break
        except discord.Forbidden:
            print("Error: Forbidden")
        except discord.HTTPException as err:
            print(f"Error: {err}")

def setup(bot):
    bot.add_cog(InviteTracker(bot))```

i tried

[elfin inlet]

if I have integration_types = [both], then how do I check if the command is invoked in a user install?

the reason is since bots cant react to user installs, which is what its trying to do, causing an error

can I do ctx.integration_type or smth?

[little cobalt]

your db is not even asyncio?

[wheat field]

why

[little cobalt]

what do you mean with why?

[little cobalt]

your db is not asyncio

[wheat field]

fixing rn

thanks

[elfin inlet]

@little cobalt can u help maybe

[little cobalt]

I never used user install commands

@sage tendon could help with that

[wheat field]

bruh

that explained

it a;;

why my events wouldnt work 😭

[sage tendon]

lol

oh i thought you were asking why = [both] doesnt work lmao

i think you need this

and then check whether guild is none, that's a user install

[elfin inlet]

ohhh that does make sense ty

[sage tendon]

well, it could also still be a DM command invocation i think

[elfin inlet]

ik you need discord.IntegrationType.user_install etc

[elfin inlet]

dms are in ctx.channel not ctx.guild

im p sure

[sage tendon]

I mean if guild is None here it can still be from a DM interaction, as it says there

https://docs.pycord.dev/en/stable/api/enums.html#discord.InteractionContextType this lets you specifically check tho, if you need it

[elfin inlet]

oh wait we're going in circles

lol

the issue is not checking if its in a dm, but if its a user install invoked in a server that doesnt have the bot

cause u can do that

[sage tendon]

i know, and that's done using this lol

[elfin inlet]

oki

ty

so its if discord.InteractionContextType?

[sage tendon]

No, the context tells you where an interaction happened

you need to check if guild is none in ctx.interaction.authorizing..thingy.guild

[elfin inlet]

well yeah I mean if discord.InteractionContextType == discord.IntegrationTypes.user_install

[elfin inlet]

?????

[sage tendon]

thats literally one of the first things i said lol

[sage tendon]

No.

if ctx.interaction.authorizing_integration_owners.guild is None

[elfin inlet]

ohhhhh

ok ty

im dumb lmao

[deft kestrel]

Any problems with bots?

WARNING: Shard ID None heartbeat blocked for more than 10 seconds.

[sage tendon]

do you have a non-async db or a lot of heavy blocking code?

[deft kestrel]

yeah but i use ctx.defer()

but i got nwo have this problem

[sage tendon]

that doesnt stop code from being blocking lol

[deft kestrel]

no code changes

[elfin inlet]

ctx.defer() just gives your bot more time to respond

you can still write blocking code

[sage tendon]

if your DB isnt async you should just switch over to the async version of whatever db you use

typically its just changing the import and adding await

[elfin inlet]

ctx.defer()
time.sleep(0.5)
``` will still be blocking

[little cobalt]

I still see people doing that and using something like requests at there bots

[sage tendon]

If you're new to python its an understandable mistake

[deft kestrel]

nevermind, it's an third party service that cause this error

[fresh sierra]

.tag blocking

[sly karma]

In asynchronous programming a blocking call is essentially all the parts of the function that are not await. Do not despair however, because not all forms of blocking are bad! Using blocking calls is inevitable, but you must work to make sure that you don’t excessively block functions. Remember, if you block for too long then your bot will freeze since it has not stopped the function’s execution at that point to do other things.

A common source of blocking for too long is something like time.sleep(). Don’t do that. Use asyncio.sleep() instead. Similar to this example:

# bad
time.sleep(10)

# good
await asyncio.sleep(10)

Read more

[meager token]

ctx.author.get_role is allways returning none, how do I find a users role with ApplicationContext from a slash command??

ctx.author.roles is none too

ctx.user.* is also allways none

It works just fine when I use the undocumented *._roles but not *.roles, this is in the first 2 lines and its the copy paste from example??

why does .roles even exist when its just a broken frontend of ._roles

Why does .roles exist when ._roles exists and is faster, less glitchy https://github.com/Pycord-Development/pycord/blob/master/discord/member.py#L556

[deft kestrel]

What are the best ways to make a temp role command?

[shell radish]

one is an array of role IDs, but the other is an array of roles. It is meant for internal use only

[deft kestrel]

Task loop to check when the role expires?

[shell radish]

what is a temp role in your usecase?

[keen relic]

Traceback (most recent call last):
  File "/home/container/.local/lib/python3.12/site-packages/discord/ext/commands/core.py", line 180, in wrapped
    ret = await coro(*args, **kwargs)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/container/banking/deposit.py", line 36, in deposit
    try:
  File "/home/container/.local/lib/python3.12/site-packages/discord/abc.py", line 1666, in send
    data = await state.http.send_message(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/container/.local/lib/python3.12/site-packages/discord/http.py", line 368, in request
    raise Forbidden(response, data)
discord.errors.Forbidden: 403 Forbidden (error code: 50007): Cannot send messages to this user
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
  File "/home/container/.local/lib/python3.12/site-packages/discord/ext/bridge/bot.py", line 144, in invoke
    await ctx.command.invoke(ctx)
  File "/home/container/.local/lib/python3.12/site-packages/discord/ext/commands/core.py", line 959, in invoke
    await injected(*ctx.args, **ctx.kwargs)
  File "/home/container/.local/lib/python3.12/site-packages/discord/ext/commands/core.py", line 189, in wrapped
    raise CommandInvokeError(exc) from exc
discord.ext.commands.errors.CommandInvokeError: Command raised an exception: Forbidden: 403 Forbidden (error code: 50007): Cannot send messages to this user```

dm is open

of the user

[fresh sierra]

does the bot is in the same server as the user ?

[keen relic]

yeap

[fresh sierra]

can the user dm the bot ?

[keen relic]

idk

[fresh sierra]

try

[keen relic]

user isn't online

[fresh sierra]

and you can you dm him / recevied message

[keen relic]

yeap

[fresh sierra]

i mean the message tells you the issue

[rain dune]

ctx.guild.get_member(config.bot_id).display_avatar

does this thing have any issue? it seems that it dont return anything

[little cobalt]

.tag get_x

[sly karma]

Any function that starts with get_ in Py-cord is retrieving the related object from your bots cache. If the object is not in the bots cache the get_ method will return None. Because of this behavior you should check if the get_x method is None and if it is use the fetch_x method.

Why Is Using fetch_ Without Using get_ First Bad?
The fetch_ method makes a call to the discord API. This API call is unneeded if you already have the information. It will also make your command take longer because it will have to send and than wait for a response from the discord API. It will also contribute to the discord APIs global rate limit of 50 requests per second.

What Is Cache?
The cache is a temporary storage inside your bot. It holds many objects from members to messages. When you restart your bot the cache will be empty. When the cache is full it will delete older objects to make space for the new objects.

[lofty parcel]

If you want the bot as a member just use ctx.guild.me

[jolly shuttle]

Can I have f"<@{user_id}>" inside of an embed?

[shell radish]

not in all fields

[jolly shuttle]

Not at all in fields (i.e. add_field), but maybe in description, author, footer etc?

[shell radish]

seems to work in description and field values

[jolly shuttle]

Im getting no luck with

embed.add_field(name=f"<@{user_id}>", value=f"{val*100:.2f}%", inline=True)

i even made sure to try user_id as an int

[shell radish]

it will only render in the value section

[jolly shuttle]

I just guessed that might be it.. thanks!

[shell radish]

maybe you can try using description instead?

[jolly shuttle]

Nah value should be fine

Thank you

[silk spire]

https://i.imgur.com/zIJT3CF.png for future reference

[hexed herald]

hey can someone help me getting my bot to really stop xD

bot = Bot()

@bot.listen("on_stop")
async def on_stop():
    await bot.api.close()
    await bot.db.close()
    await bot.close()

if __name__ == "__main__":
    try:
        loop = asyncio.get_event_loop()
        print(os.getpid())
        loop.run_until_complete(bot.start(DISCORD_API_KEY))
    except KeyboardInterrupt:
        bot.logger.critical("Shutting down...")
        print("I AM NOT DEAD")
        print("yet")
        bot.dispatch("stop")
        exit_()
        loop.run_until_complete(bot.close())

        # cancel all tasks lingering
    finally:
        loop.close()

I've got this piece of code to start the bot and later stop the bot on KeyboardInterrupt and in the stop dispatch every job with some async connection get's closed (e.g. aiosqlite connections and aiohttp sessions) but the process does not stop I am also closing the bot in the on_stop event even through when I let loop.run_until_complete() it's not completely shutting down this is why I use os.exit() (importet as exit_()) to exit the process. but it's not stopping

it' s remaining unresponsive since many minuets now and gave me this traceback

Traceback (most recent call last):
  File "E:\Programming\PyCharm\Dragons-BotV2\main.py", line 90, in <module>
    loop.run_until_complete(bot.start(DISCORD_API_KEY))
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\asyncio\base_events.py", line 640, in run_until_complete
    self.run_forever()
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\asyncio\windows_events.py", line 321, in run_forever
    super().run_forever()
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\asyncio\base_events.py", line 607, in run_forever
    self._run_once()
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\asyncio\base_events.py", line 1884, in _run_once
    event_list = self._selector.select(timeout)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\asyncio\windows_events.py", line 444, in select
    self._poll(timeout)
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\asyncio\windows_events.py", line 817, in _poll
    status = _overlapped.GetQueuedCompletionStatus(self._iocp, ms)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
KeyboardInterrupt
Exception in thread Thread-1:
Traceback (most recent call last):
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\threading.py", line 1038, in _bootstrap_inner
    self.run()
  File "E:\Programming\PyCharm\Dragons-BotV2\venv\Lib\site-packages\discord\gateway.py", line 166, in run
    f = asyncio.run_coroutine_threadsafe(coro, loop=self.ws.loop)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\asyncio\tasks.py", line 926, in run_coroutine_threadsafe
    loop.call_soon_threadsafe(callback)
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\asyncio\base_events.py", line 806, in call_soon_threadsafe
    self._check_closed()
  File "C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\asyncio\base_events.py", line 519, in _check_closed
    raise RuntimeError('Event loop is closed')
RuntimeError: Event loop is closed
C:\Users\User\AppData\Local\Programs\Python\Python311\Lib\threading.py:1040: RuntimeWarning: coroutine 'DiscordWebSocket.send_heartbeat' was never awaited
  self._invoke_excepthook(self)
RuntimeWarning: Enable tracemalloc to get the object allocation traceback

please ping me if someone responds to this

[little cobalt]

why do you run the bot like that?

[sage tendon]

on_stop isnt even a thing in pycord

Or at least it's not documented if it exists

so i dont think thats pycord

[little cobalt]

its not even at discord.py

[sage tendon]

lol

what is bro running

[little cobalt]

same for nextcord

[sage tendon]

@hexed herald

[lofty parcel]

He's dispatching his own event

bot.dispath("stop")

[sage tendon]

oh i see
but why

literally all of that code could be 2 lines

[lofty parcel]

Because they're coros

Just kill the process tbh

[sage tendon]

bot = discord.Bot(...)
bot.run(...)
simple as that

also accounts for keyboard interrupts by the fact that keyboardinterrupts kill programs by default

[hexed herald]

hi again, I am dispatching my own event because I've external connections which need to be closed before the bot shuts down like normal.

class Bot(multicogBot, commands.Bot):
    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)
        self.client_version = VersionInfo(1, 3, 3, "")
        self.api: aiohttp.ClientSession = None  # type: ignore
        self.boot_time = datetime.now()  # Ignoring because it's dynamically allocated
        self.db: ContentDB = None  # type: ignore
        self.sts: ShortTermStorage = None  # type: ignore
        self.ipc = ipc.Server(self, secret_key=IPC_SECRET)
        self.logger = CustomLogger(name="core", start_stamp=self.boot_time)

And since I've attached the db like that, to be able to access it in cogs, the bot does not shut down like a "simple" instance
(the None from the db is later replaced while the on_ready event)

[sage tendon]

but if the bot process dies, so do all the connections

[hexed herald]

yeah, the process is not dying, it's living as a kind of zombie ;-;

the bot isn't online but the process is not dead

[sage tendon]

if you ctrl+c?

also, what is this

[hexed herald]

legacy code, good you mention it xD

does not change the result if I remove it

[sage tendon]

yea i mean i get that but i dont get the overcomplicated start and stop of your bot

if you stop it with ctrl+c the process will die completely

and thats also the issue, something during your startup (i think)

[hexed herald]

but if I don't do it the complicated way it's taking ages until it stops being online and even longer until the process stops

with the complicated way it at least stops being online instantly even through the process is not really dying

[sage tendon]

why? you're starting it the exact same way in both cases

[hexed herald]

I have no clue, that's how I ended up here 😅

[sage tendon]

remove all this and replace it with a simple bot.run

[sage tendon]

this i mean

[hexed herald]

I did, the process is not dying since ~4 minuets

[sage tendon]

did you ctrl+c?

[hexed herald]

yes

[sage tendon]

show

[hexed herald]

spamming control + c at around 8 secs + 10 secs

[sage tendon]

i dont see any ctrl+c input

because it should print to the console instantly

[hexed herald]

(the cursor stops blinking)

[sage tendon]

also to test it just put all your closing stuff after bot.run

[hexed herald]

current code + console

    print(os.getpid())
    bot.run(DISCORD_API_KEY)
    print("I have stopped")

but I still can't reuse the terminal

and the process is still listed in the taskmanager

[sage tendon]

i think your cmd is borked lmfao

[hexed herald]

should I try with powershell xDDD

[hexed herald]

I had an Idea that worked o.o

    pid = os.getpid()
    bot.run(DISCORD_API_KEY)
    bot.logger.critical("Exiting")
    psutil.Process(pid).terminate()

Just killing the process after the bot finished running did the job

[tiny fractal]

is there a reason why my bot doesn't generate auditlog entries?

[sage tendon]

for what action

[tiny fractal]

specifically roles

it used to make audit log entries when it granted and removed roles. now it doesn't

but they generate when i do it my self

[sage tendon]

something in the very far back of my head tells me I had something similar too but I don't really remember

not sure, lemme test with my own bot in a bit
if anything it's a discord change or issue tho, nothing to do with your bot or pycord

[tiny fractal]

bummer

yea, weird. it shows audit entries when the bot creates and deletes channels and sets overrirdes

[keen relic]

can we set multiple value of command option

[sage tendon]

an option can only have one value passed to it

[keen relic]

ok

[stray pasture]

you are right just tested as well

my bot was relying on those audit log 😄

[tiny fractal]

same lol

[stray pasture]

what is happening 😄

no anoncement about it either

not even a warning

is there a way to report it to discord ?

[sage tendon]

they actively ignore bug reports, so no

don't waste time trying, i've done so numerous times

[stray pasture]

so I can only hope they notice the issue and fix it 😄

[sage tendon]

yip

[stray pasture]

[sage tendon]

i've easily reported 10+ minor-to-severe bugs by now, not a single one has been fixed

some are over 2y old

[stray pasture]

damn, my bot will be broken because of this to 😄

[edgy nest]

ask in discord.gg/discord-developers or make an issue in https://github.com/discord/discord-api-docs

since it's an api problem

(most likely)

[sage tendon]

good luck

[stray pasture]

Thank you ! I will need luck ahaha yes

[lapis dock]

I feel like the API team does better than the client team

[round heart]

No. I have to hit CTRL-C twice in order for my bot to completely exit. It's certainly my own issue, but I don't know why or how to correct it.

[errant trout]

are you intercepting keyboardinterrupt or something

[fossil valve]

hey, is there a way to catch autocomplete errors (e.g an event like we have for slash commands)?

[fresh sierra]

Hi, my bot rate limit a lot about message, but i dont have any fetch_message so what can be the cause of that ?

[echo wraith]

Maybe history or prune ?

[fresh sierra]

yeah but should not explain why i got that much

but i will wait

i think it might be because i just switch version

and so the old one already rate limit a lot

[sage tendon]

yea I was also gonna say history and big prunes

[fresh sierra]

might be big prune then

thanks

[heavy inlet]

How do I add a button to a message ONLY when a certain condition is True?

[little cobalt]

Check and edit the Message I would say

[sage tendon]

like what condition

[heavy inlet]

boolean, either True or False

[sage tendon]

Yea thanks I know what booleans are

I mean what condition in particular you're checking for

[errant trout]

not really... you can try-except, but your autocomplete ideally should never error

[deft kestrel]

for how much time discord caches an image (webp)?

[sage tendon]

wdym?

[deft kestrel]

like i send you an image link (bot embed thumbnail) and i change it

[sage tendon]

image links outside of discord are valid for 24h

[deft kestrel]

no i mean 3p discord image cache

[sage tendon]

inside of discord they're fetched even if the link is over 24h old

[sage tendon]

what

[deft kestrel]

the right logo (thumbnail) it's not updated, but if i open it i see the new bot logo

[sage tendon]

restart your discord ig

[deft kestrel]

nope

[sage tendon]

and wdym when you open it

[deft kestrel]

right click => open image

via browser

[sage tendon]

oh you mean you changed what image the link points to?

[deft kestrel]

i just changed the file with the same image link

yep

image replace

[sage tendon]

No clue then, the client caches stuff for relatively long

really you should just edit the embed if you plan on changing what the image is more often, and create new links

[deft kestrel]

deam

so it's only the client cache?

not server-side cache right

[sage tendon]

hm no idea actually, try clearing your local cache

[fresh sierra]

my decorator doesnt work and just return true to the command, am i the only one ?

(mean everyone can use every single command

like it does not even get print

    @staticmethod
    def has_perm(perm_level: str | None = None):
        async def wrapper(ctx: LumabotContext):
            print(ctx.command)
            perm = perm_level
        return commands.check(wrapper)

    @Lumabot.group("bot")
    @Lumabot.command()
    @Lumabot.has_perm()
    async def ping(self, ctx: LumabotContext):

[sage tendon]

show the entire decorator code

[fresh sierra]

i added the missing line

like it does not even print the first line

[sage tendon]

why do you not use the default @commands.check decorator

why do you do custom copies of literally everything the library offers

[fresh sierra]

its just a custom decorator man

when im doing the help, it does trigger the print of the decorator

well the slash trigger the print

[sage tendon]

so when does it not work

[fresh sierra]

using prefix

[sage tendon]

bridge group?

[fresh sierra]

yep

[sage tendon]

oh boy

no idea about those

[fresh sierra]

i will check how does it invoke it

i thing its a library issue

[sage tendon]

bridge is messy

[fresh sierra]

i did implement my thing

cos bridge cant work rn

but its still relying on bridge in some part

[sage tendon]

then use normal bridge first to check if it isnt an issue with your code

because i think, depending on how custom you made it, you are invoking the command incorrectly and bypass the checks

[fresh sierra]

i didnt change the invoke part

and i will check which invoke part is use

if its prefix one

or another one

[sage tendon]

just saying because of this

[fresh sierra]

yeah this is handle before by the

if await self.can_run(ctx, call_once=True):

and in my case this return True

but it doesnt trigger the decorator

which is the issue

[sage tendon]

im not sure if it works that way but can you check the checks attribute of the command?

cuz this

[fresh sierra]

the self.can_run check only the bot._checks

which seems to be only the global check

which is strange, both part of the code does the same thing

only difference is the ctx which is one time a prefix and another time a slash

[sage tendon]

can you change the print to just be a letter instead of ctx.commands

maybe that is null somehow? idfk

[fresh sierra]

also tried

just the can_run return always true

but trigger the decorator only for the slash command

[sage tendon]

no clue then

i hate bridge

[fresh sierra]

gonna try to debug that bug it seems to be from the context part

[fresh sierra]

this is true

so i will now have to check about the command invoke itself

[errant trout]

commands have their own can_run function

[fresh sierra]

yep im checking that rn

but i dont really understand why the prefix are not submitted to that decorator while the slash is

ctx.command class seems to be
<class 'discord.ext.commands.core.Group'>

so update now i check the different self.checks for each command, the one for the slash command find the
[<function Lumabot.has_perm.<locals>.wrapper at 0x7f37d120ac00>]
while the prefix return []

i fixed it

[sage tendon]

how?

[fresh sierra]

prefix command checks was a copy of the group

while it should have been a copy AND the checks

[vague plaza]

whats discord bots upload limit?

[sage tendon]

file size?

[vague plaza]

yeah like video

they keep changing it for users but idk for bots

[sage tendon]

25MB

[vague plaza]

thanks, and maximum text character limit?

[sage tendon]

4000

[little cobalt]

Is it still 25MB?

[sage tendon]

Discord says so

25MiB, actually

[hexed herald]

it's 10 MB as you can read in this article https://support.discord.com/hc/en-us/articles/115000435108-What-are-Nitro-Nitro-Basic

at least that's my lastest information

[sage tendon]

No.

https://discord.com/developers/docs/reference#uploading-files

Official Discord Documentation.

[lofty parcel]

Thats for users, not bots

[hexed herald]

huh, that's wierd that it's different for normal users and bot's but okay, learned something new

[sage tendon]

Bots have some nitro features to a degree

like animated avatars

and uhh

banners

[lofty parcel]

Or external emojis

[sage tendon]

stickers too right?

[fresh sierra]

80% of cpu usage for a 2k guild bot with 4 shards

should i begin to look about it or it seems normal to u

[sage tendon]

80% is usually the threshold to check what you can do

aka bigger VPS usually

[fresh sierra]

its on pterodactyl so like its for one of my bot only

[sage tendon]

thats great, but if you have 80% cpu usage you still need to get a bigger VPS sooner or later

[fresh sierra]

i have 4v core so im at like 150 in total out of 400

[sage tendon]

then why do you say 80% lol

[fresh sierra]

80% for one of my bot

so like does i have an issue in my code or its jsut normal for a bot with 2k

[sage tendon]

with no info about the CPU it could be anything

[fresh sierra]

i do have a lot of even/loop

[sage tendon]

its probably just the amount of events at 2k guilds

[fresh sierra]

my bot test has less than 5% so yeah might be just number of event

[sage tendon]

limit your intents if possible

[fresh sierra]

yeah cant since i use them, but i will thing about the on_presencd

implement more cache i think

does on_connect is supposed to be trigger for each shard connect ?

same for on_ready

[torn barn]

.rtfm on_connect

[sly karma]

7 results

discord.on_connect
discord.Bot.on_connect
discord.AutoShardedBot.on_connect
discord.ext.bridge.Bot.on_connect
discord.ext.commands.Bot.on_connect
discord.ext.commands.AutoShardedBot.on_connect
discord.ext.bridge.AutoShardedBot.on_connect

[torn barn]

There is on_connect and on_shard_connect

So for shards it dispatches the second one

And after connecting every shard it dispatches on_connect fron what I read

[fresh sierra]

yep it seems like it

        self.dispatch("connect")
        self.dispatch("shard_connect", data["__shard_id__"])

[torn barn]

Then thats it

[errant trout]

2000 for bots

(but they get 6000 through embeds so eh)

[sage tendon]

bruh right, forgor

dumb limit

[shadow sigil]

question, what the heck does this error mean

Ignoring exception in on_application_command_error
Traceback (most recent call last):
  File "...\.venv\Lib\site-packages\discord\client.py", line 409, in _run_event
    await coro(*args, **kwargs)
  File "...\bot.py", line 182, in on_application_command_error
    await ctx.respond(response)
  File "...\.venv\Lib\site-packages\discord\interactions.py", line 616, in respond
    return await self.response.send_message(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "...\.venv\Lib\site-packages\discord\interactions.py", line 957, in send_message
    await self._locked_response(
  File "...\.venv\Lib\site-packages\discord\interactions.py", line 1288, in _locked_response
    await coro
  File "...\.venv\Lib\site-packages\discord\webhook\async_.py", line 222, in request
    raise NotFound(response, data)
discord.errors.NotFound: 404 Not Found (error code: 10062): Unknown interaction

log file just contains this

the bot just gave a The application did not respond error on discord

but running the command a second time worked fine

[errant trout]

all your responses took too long

initial response took too long, so when it tried to respond discord gave you an error; but your error handler also tries to use the same interaction token to respond, so you errored again

then your logs also show that your autocomplete is taking too long

[shadow sigil]

...

i'm guessing it's probably due to it being idle for an extended amount of time

because basically the first thing the command does is await ctx.defer()

[maiden bloom]

Is there a code example for setting up entitlement checks and such?

[fresh sierra]

anyone experiencing issue with mention command ?

    return f"</{self.qualified_name}:{self.qualified_id}>"
                                self.parent.qualified_id
AttributeError: 'NoneType' object has no attribute 'qualified_id'

[clear lark]

is it possible to setup in app purchases on my bot, SKUs rather

[sage tendon]

sure

[sage tendon]

only exists for slash commands

[fresh sierra]

im using it for a slash command

[sage tendon]

100% sure?

[fresh sierra]

seems like something the bot.walk_application_command return []

even if the command work

[frail basin]

Am I doing something wrong here?
This is what it prints:

i=0 role.name='@everyone' role.position=0
i=1 role.name='test' role.position=1
i=2 role.name='new role' role.position=1
i=3 role.name='Akos test' role.position=2
i=4 role.name='12345' role.position=3
i=5 role.name='test2' role.position=4
i=6 role.name='----- Client Roles -----' role.position=5
i=7 role.name='Network Support' role.position=6
i=8 role.name='Technical Support' role.position=7
i=9 role.name='Staff' role.position=8
i=10 role.name='Management' role.position=9
i=11 role.name='.' role.position=10
i=12 role.name='LDG Dev' role.position=11

How can 2 roles have the same position?

[frail basin]

Oh, thanks

[echo wraith]

Why is it so tho ?

[sage tendon]

no idea

[frail basin]

Is this endpoint implemented? https://discord.com/developers/docs/resources/guild#modify-guild-channel-positions

I only found it's role counterpart in the docs.

[shell radish]

yes https://docs.pycord.dev/en/master/api/models.html#discord.TextChannel.move

[frail basin]

Isn't that only for moving a single channel?

I am basically looking for the channel counterpart of Guild.edit_role_positions

[sage tendon]

Don't think that's in pycord
and i'm wondering why the endpoint exists at all, like, with roles it makes sense, e.g. in the UI you can move them around fine and then click save
But channels, you usually only move one-by-one
and typically you dont move enough channels at once to justify a bulk operation

[frail basin]

Seems like the function exists for it actually, after looking into how TextChannel.move is implemented

[sage tendon]

yea but in a private attribute so not really meant to be used
odd

oh wait nvm

ah idk lol

[frail basin]

Well I am def not sorting 200+ channels by hand, alphabetically lol

[sage tendon]

1. why do you have so many channels
2. just do it in a loop really

[frail basin]

Discord would nuke the IP if I looped over all the channels

[shell radish]

if your use case is really necessary, then you can send an http request through the library

[sage tendon]

i dont think channel move limits are low

its just channel renames that are mega limited

[frail basin]

I am forced to use channel.edit, since in some cases I also need to move the channel to a different category

[frail basin]

Would it just be bot.http.bulk_channel_update?

[sage tendon]

no clue if discord specifically applies the rate limit to renames, or all edit operations

[frail basin]

all edit operations

even if it doesnt actually do anything

[sage tendon]

damn

[shell radish]

looks right

[frail basin]

This would ratelimit the bot after the 3rd - 4th cycle, even though nothing actually happens.

[sage tendon]

would this be worth properly putting into the library?

[sage tendon]

but if you used move()?

[shell radish]

you can always open a feature request

[sage tendon]

yea i'm just wondering if like, its worth it / has potential or if its a waste of time

i never really do feature requests ever for anything

[frail basin]

Seems like move() hits the same endpoint as edit()

[sage tendon]

but what about this?

[fresh sierra]

when you create a role you can put a position

and maybe it was a bug and they where lazy to correct it

[frail basin]

actually, its a different endpoint but same ratelimit bucket

[frail basin]

Well thats just the underlying function of move(), so its the same in terms of ratelimits

However I'll only need to send a single request this way.

[echo wraith]

That wouldn't even surprise me

[fresh sierra]

i dont see any other option, in what way having 50 roles with the position = 1 is a good thing

[ivory wind]

hello, i have an extension that's made up of a directory. i am able to load it with recursive = True but dont see a way to do that with re/unload

[fresh sierra]

for extension in self.bot.extensions:
if extension.startwiths ... continue else unload/reload

[meager token]

why is this happening

it works for everyone except me

[fresh sierra]

read the message

[meager token]

its not missing any permissions though

and theres no error

[fresh sierra]

you are i think

[meager token]

wat??

[fresh sierra]

you are missing permission to run that command

i think thats what missing perm mean here

[meager token]

it only checks if I have the role though and i have the role??

and thats only when it runs, its not running because of "missing permissions"

[fresh sierra]

did you check the default permission from discord ?

[meager token]

that exists?

how do I control that?

[fresh sierra]

/ try to reload ur discord

[meager token]

Ive allready done that

[fresh sierra]

yes but normally you shouldnt be able to see the command in that case but we never know

in the discord server setting

[meager token]

I have permission in discord

[fresh sierra]

you can choose about the permission requiered

[meager token]

wait so the channel has slash commands disabled??

why cant it say that

[fresh sierra]

maybe

[meager token]

bruh

[fresh sierra]

more fun for us to guess

[meager token]

wow it wont even let me /tableflip in taht server

and it doesnt fix when I reload discord