Basic Pycord Help

[errant trout]

you would have to tag everyone for every thread lol

idk if the everyone tag even works for threads, but direct mentions and role mentions (under 500 members) do

[pulsar ferry]

yeah the way i was thinking of setting it up was to tell people if they want notifications to just click this

so that way we could get around having to use roles for mentioning

but it doesnt seem to alrert the same way, like it doesnt have the red badge

[errant trout]

oh i didn't even know that existed

but yeah it only sends a notification without mentioning the user or making them join the thread

[pulsar ferry]

which is kinda what i want

because otherwise then their left hand sidebar would just get flooded

[echo wraith]

@tribal hornet

.tag slashnoshow

[sly karma]

Application Commands Not Showing Up?

• Refresh Discord by restarting or pressing Ctrl+R (or Command ⌘ + R)
• Uninstall libraries that conflict with the discord namespace (e.g. discord.py).
• Invite your bot with the application.commands scope.
• Load cogs before bot.run() (e.g. not in on_ready).
• Do not override on_connect.
• Update to the newest version of py-cord (see ?tag install).
• Turn off User Settings > Accessibility > Chat Input > Use legacy chat input.
• Share your code and errors.

[tribal hornet]

Ohhh okay, thanks

That's my full embed code pretty much

[echo wraith]

you can use ``` for multiline codeblocks

You can define your embed outside of the command btw

since it dosen't change between comman inokations

[tribal hornet]

Idek what that means 🤣
Had this bot made for me, but now trying to add a single simple new /slash command that just gives the member updates on what's to come

@echo wraith Here u go 👆

[echo wraith]

thy

[tribal hornet]

And this is the thing I'm trying to get added pretty much

[echo wraith]

this code is not pycord code

[tribal hornet]

What then?

[echo wraith]

@bot.command(name="menu", description="Open the methods menu")
async def menu(interaction: discord.Interaction):

[echo wraith]

idk

[echo wraith]

this is not pycord.

[tribal hornet]

I did pip show py-cord, says It's pycode

[echo wraith]

Also the code is messed up, it uses blocking code

I hope if you paid you didn't pay much for that...

[tribal hornet]

Paid $35

Is that bad?

[echo wraith]

yeah

[tribal hornet]

damn

[echo wraith]

you got kind of scammed there

[tribal hornet]

ah bruh

At least the bot works the way I wanted it to

[echo wraith]

yeah true

[tribal hornet]

So since the bot code is so messed up, I won't be able to add this new code / feature to it?

[echo wraith]

I'll do some cleanup for you.

[tribal hornet]

How's it coming along @echo wraith

[echo wraith]

The more I look at the code the more messet up it looks

It should probably in theory, but I can't test it out and understand why it wouldn't work if the code is all over the place.

[tribal hornet]

Damn okay🙏
I think the issue I'm having is the /slash command not showing after i do CTRL+C and then doing Python main.py

[echo wraith]

https://paste.pythondiscord.com/GGSA

try with this. You will need to

pip install python-dotenv aiofile

in your terminal first
and create a file called .env in the root, containing th eofllowing

TOKEN=<your bot token here>

[tribal hornet]

testing 🙏

[echo wraith]

works on my side btw

pip install typing_extensions too

what is your python version

[tribal hornet]

./

2.6.1 pycord

[echo wraith]

not pycord. python

python --version

[tribal hornet]

Python 3.11.9

[echo wraith]

yeah then you need to install typing_extensions

[tribal hornet]

Done 🙏

Do I do python main.py

or python .env?

[echo wraith]

python main.py 😆

[tribal hornet]

Some error

[echo wraith]

means your bot token is invalid

[tribal hornet]

Just double checked, it isn't

[echo wraith]

it must be

can you recreate one ?

[tribal hornet]

Yeah, let me do that

[echo wraith]

also make sure

.tag install

[sly karma]

1. Uninstall discord.py or any other forks of discord.py you might have with the namespace discord.
python -m pip uninstall discord.py discord -y

2a. Install py-cord
python -m pip install py-cord

2b. Update py-cord
python pip install -U py-cord

Installing other builds:
Note: You need to have git installed. Use ?tag git to find out how to install git.

Updating the module to master branch (unstable):
pip install -U git+https://github.com/Pycord-Development/pycord

[echo wraith]

just to be sure

[tribal hornet]

So first;

python -m pip uninstall discord.py discord -y

Then;
python -m pip install py-cord

And lastly;
python pip install -U py-cord

[echo wraith]

not the last one

just 1 and 2a

[tribal hornet]

Did that, I reset my token as well

And still shows same error

[echo wraith]

update the token in the .env file

and save that file too

you don't need the < >

also, please delete that picture

[tribal hornet]

It's online

[echo wraith]

neat

make sure to restart discord

[tribal hornet]

/menu is showing and works, but not /upcoming

[echo wraith]

for the commands to show

[tribal hornet]

Ohh wait

Yeah now it works

Thank you so so much @echo wraith

Really appriciate it

Oh wait

some of the /menu doesn't work

[echo wraith]

what specifically ?

on line 48, remove the await

that's mb

wait no

wait 1 sec

[tribal hornet]

🙏 alr alr

[echo wraith]

keep it as is, just change json.load to json.loads with an s

keep the await and all

[tribal hornet]

Works perfect now man, thank you so much 🙏

[echo wraith]

happy to help !

[pulsar ferry]

Is it possible to have multiple default emojis for a thread, or disable them all together and have the bot add 2 every time a thread is created?

[echo wraith]

the second option is possible afaik

[pulsar ferry]

Yeah it seems like on the forum view it still only shows one at a time ☹️

[echo wraith]

yes

[pulsar ferry]

trying to play with this... whats the proper way to react on a thread?

    thread = await channel.create_thread(name=thread_name, embed=embed, applied_tags=tags, auto_archive_duration=auto_archive_duration)
    # Add Verify and Invalid reactions to the thread
    await thread.add_reaction("✅")

[echo wraith]

I think you need to react to the original message

[pulsar ferry]

does the default emoji on the thread automatically sync to the first message?

ah nvm got it thanks! - looks to be so

    thread = await channel.create_thread(name=thread_name, embed=embed, applied_tags=tags, auto_archive_duration=auto_archive_duration)
    message = await thread.fetch_message(thread.id)  # Fetch the first message of the thread

    # Add Verify and Invalid reactions to the thread
    await message.add_reaction("✅")

[pulsar ferry]

playing with this some more and seeing some pretty weird behavior, when i do this

    message = await thread.fetch_message(thread.id)  # Fetch the first message of the thread
    await message.add_reaction("❌")
    await message.add_reaction("✅")

On the discord desktop app it shows the last reaction (the green check) as the default, but on the mobile app it shows the first X as the default?

[edgy nest]

you might have a default emoji set, and the client is handling it wrong

[torn barn]

Also, instead of making an api call you can use get_partial_message

.rtfm Thread.get_partial

[sly karma]

1 results

discord.Thread.get_partial_message

[maiden bloom]

When I try to install py-cord 2.6.1, I get this error:

building 'aiohttp._websocket.mask' extension
  error: Microsoft Visual C++ 14.0 or greater is required.

I went and downloaded what I thought was needed from the MS build tools and restarted, but it's still showing this error. Is there something else I need to install?
Using Python 3.12.8
SOLVED: I installed the ARM version of Python and after switching to the regular Windows x64 release, the py-cord install worked fine and I didn't need anything additional from Visual Studio.

[wise harness]

Hi, this piece of code splits options into multiple select menus to adhere to the limit of 25 options per menu:

def create_select_menus(self, options):
                MAX_OPTIONS = 25
                option_chunks = [options[i:i + MAX_OPTIONS]
                                 for i in range(0, len(options), MAX_OPTIONS)]
                print(len(option_chunks))
                for i in option_chunks:
                    print(len(i))
                select_menus = []

                for idx, chunk in enumerate(option_chunks):
                    select_menus.append(discord.ui.Select(
                        placeholder=f"Menu {idx + 1}",
                        min_values=1,
                        max_values=1,
                        options=[discord.SelectOption(
                            label=option, value=option) for option in chunk]
                    ))

                return select_menus

From the output, there are a total of 5 select menus (which is the limit) for this interaction and total of 124 options:

5
25
25
25
25
24

yet i'm receiving DiscordServerError: 500 Internal Server Error (error code: 0): 500: Internal Server Error..?

[sage tendon]

typically if you get 500 its not your fault

[wise harness]

well, i did try this exact same command which had 50 or so options and that one doesn't throw this error, so i assume it had to do something with running too close to the limit or something

[sage tendon]

oh, i think you can only have 5 UI rows on one message, and you have 6

weird that that gives a 500 tho

[wise harness]

No, the first print which shows 5 is the number of select menus

these are the number of options per select menu:

25
25
25
25
24

[sage tendon]

ah

yea then no clue

[wise harness]

i did some further testing by truncating the list of options, this works:

25
25
25
25
19

but this doesn't:

25
25
25
25
20

This is very likely some bug or it's a discord limit that i may not know of?

[sage tendon]

if anything its a discord bug

[wise harness]

is paginator.update() not too kind with ephemerality?

paginator = Paginator(
                pages=mypage, timeout=898, author_check=False)

            class MyModal(discord.ui.Modal):
                def __init__(self, *args, **kwargs) -> None:
                    super().__init__(*args, **kwargs)
                    self.add_item(discord.ui.InputText(
                        label="Enter Page number:"))

                async def callback(self, interaction: discord.Interaction):
                    page_num = int(self.children[0].value)
                    try:
                        await paginator.goto_page(page_num-1)
                    except IndexError:
                        await interaction.response.send_message("Invalid page number.", ephemeral=True)

            class MyView(discord.ui.View):
                @discord.ui.button(label="Switch page", row=1, style=discord.ButtonStyle.primary)
                async def button_callback(self, button, interaction):
                    await interaction.response.send_modal(MyModal(title="Travel to your desired page."))
            await paginator.update(custom_view=MyView())
            try:
                await paginator.respond(ctx.interaction, ephemeral=get_ephemeral_setting(ctx))

            except discord.Forbidden:
                return
            except discord.HTTPException:
                return

got this thing which throws AttributeError: 'NoneType' object has no attribute 'edit'

[sage tendon]

yea ephemeral messages are a bit weird so there's a change that function doesnt account for that

[echo wraith]

I have a pr open but it is not for this issue afaik. I will open another, can you just send me a short minimal reproductible example ?

[wise harness]

The code I appended is pretty much that. Basically there's the default paginator instance and i am appending a button (for modal) in the second row using the paginator.update() method, and goto_page() to travel to the page number.

Also, something related:
I tried enclosing paginator.update() in a try-except block, but i just do nothing in the except block. This allows me to get the combination that I want even in ephemeral state (in the screenshot below), however when goto_page() is invoked in the callback, this error is thrown: discord.errors.NotFound: 404 Not Found (error code: 10008): Unknown Message

[mossy star]

is there a reason an InputText returns different values for self.children[0].value and self.children[0].to_component_dict()['value']

putting in an self.children[0].value = self.children[0].value seems to fix it but it's a little weird

[sage tendon]

like what two values do you get

[mossy star]

self.children[0].value returns the value after the modal is submitted but self.children[0].to_component_dict()['value'] returns the value it was initialized with

and if value wasn't set during creation, it doesn't exist in the dict

[sage tendon]

to_component_dict isnt even documented

unless its a built in i dont know

[lapis dock]

to_component_dict is used internally for sending the data to discord iirc

[mossy star]

it was just a command the IDE popped up with that seemed to fit what i wanted

i just assumed it was from a parent class

[lapis dock]

yeah, It probably should be private. It is working as intended, and it is not intended to be used externally from the lib

[mossy star]

it's literally one of the first things pycharm brings up lmao

but got it, not intended to be used

but back to the other part of it, why does self.children[0].value = self.children[0].value cause anything to happen at all?

[wise harness]

just an update if anyone is interested, passing interaction parameter to both update() and goto_page() and enclosing them in try-except blocks provides a workaround fix for ephemerals

[mossy star]

adding this in causes the value to be used as the new default value when the modal is reused. but the value is reset to the original otherwise

[lapis dock]

I have no idea why this would change anything.

It could be a weird quirk with how the setters work on value

[sage tendon]

because pycord uses properties so a function is called

yea

[lapis dock]

Yeah, I looked closer .value is never supposed to be assigned externally either

[mossy star]

so i suppose the intended method would be to delete and replace the InputText instead?

or i guess at that point there isn't much reason to reuse the modal

i just thought it was a handy feature that setting value to itself would keep its values between uses

when you click the little [source] button next to InputText in the docs, it does seem to have a setter. and to_component_dict() is there too

but yeah i can see where the weirdness comes from now that i've read through it all

but imo value shouldn't be being reused for two different things (_input_value and _underlying.value) depending on the context. it's a bit confusing

[errant trout]

it's probably a minor oversight, if you think the behaviour should be different then submit a PR.

[halcyon totem]

hi, i know nothing about coding and stuff, i need help on downloading pycord

[sage tendon]

pip install py-cord

and if you're brand new to programming I highly recommend doing some very basic stuff in python first and not to start with py-cord as your first thing ever

@halcyon totem

[glad garnet]

yeah, you should create your own programming language first to get familiar with the tech before using python /s

[ebon swift]

would someone be willing to help me with a small issue, I shortened my bot to the straight basics, it has a cog with a \ping command, I enter a discord_token_for_test_server the ping is available and works but when I just switch discord_token_for_real_server ping command isn't listed and doesn't work, does anyone have any idea why this would occur ? (if you think I should create a new space on the forum let me know)

[errant trout]

did you reload discord

[ebon swift]

I did not

let me try that

even after ctrl+r it doesn't appear..

[errant trout]

can you show the command code

[ebon swift]

import discord
from discord.ext import commands
from discord.ext.commands import Context
from loguru import logger

class Fun(commands.Cog):
    def __init__(self, bot):
        self.bot = bot

    @commands.slash_command(name="ping", description="Check if the bot is alive.")
    async def ping(self, ctx):
        await ctx.respond(f"Pong! thanks for checking on me {ctx.author.mention} !")


def setup(bot):
    bot.add_cog(Fun(bot))

[errant trout]

can you show your main file

it's irrelevant to their issue

[ebon swift]

as for the import question I shortened the bot here and removed a ton of stuff so this is still here cause it as used before but not rn

[errant trout]

hmmm

are you sure you invited the bot? lol

[ebon swift]

stange isn't it

it is connected yes

should I re invite it ?

maybe ?

[errant trout]

does it show up in guild settings > integrations

see if it has any commands listed there

[ebon swift]

hooo

it does shows up

but one part is missing I think

[errant trout]

reinvite with the applications.commands scope

[errant trout]

though bot is meant to include that by default

[ebon swift]

see lara has slash command stuff

[errant trout]

reinvite it

[ebon swift]

ok

sorry if I sound dum which scope should I take appart from bot, I always forgot ?

is only bot enough ?

[sage tendon]

bot and application.commands

[errant trout]

it's included by default, but try add applications.commands

[ebon swift]

thank you

this didn't worked but that's definitely the issue

cause I have the slash command icon on my test server

maybe I'll jsut fully recreate it on dev portal

[errant trout]

eh... did you ever set guild_ids on any commands?

[ebon swift]

I didn't

[errant trout]

try changing the command name and restart the bot again

[sage tendon]

and restart discord if you havent yet

[ebon swift]

I did

I'll try to change command name

changing the command name indeed worked

could it be a conflict issue with an already existing bot ?

[sage tendon]

no

[ebon swift]

that what I thought, still the behavior is kinda strange

I'll dig into it and see if I can see where this comes from

[errant trout]

discord generally refers to commands by their name, so changing the name forces them to recognise it as a new command

[sage tendon]

discord's entire backend is strange

[ebon swift]

ok well thank you I'll keep that trick in mind if that ever happens again

[errant trout]

(though this is also somewhat the fault of our syncing method)

[ebon swift]

when I get my bot up and running as I want it I think I'll dig into pycord core

see how it's made

I'm sure I'll learn a lot of stuff

[sage tendon]

I'd say most of it is kind of just replicating Discord stuff like, well, guilds, channels, etc

i mean thats kind of the point of the library lol

[ebon swift]

yes I was more talking about the way to communicate with the discord api and such

[sage tendon]

http requests :]

[errant trout]

eh... maybe at the request level, but "replicating the library" is pretty broad

[ebon swift]

You know what I mean

[errant trout]

looking into any library to see how it works is worthwhile

[ebon swift]

thank u

[sage tendon]

yea

[ebon swift]

just need to solve some docker issues now , thanks again for the help see ya !

[little cobalt]

which permission should I use for bridge commands?

[sage tendon]

wdym?

[little cobalt]

discord.default_permissions or the basic one

[sage tendon]

ah, no idea

[red mist]

I mean iirc you can use that

Default perms

As it's discord.permissions object no?

and that object just gives out bools

Sooooo theoretically if you know what the bools values are gonna be you can even input them manually \o/

[little cobalt]

no

its bridge.has_permissions

[red mist]

doesn't that just ... Wait hold on a min.

[little cobalt]

.

;3

[red mist]

Bro-

[little cobalt]

xd

[wise knoll]

I have a list of users that the bot is intended to send DMs to and the basic function layout is like this

async def functionName(ctx):
    ctx.defer()
    for u in usersList:
        x = await get_or_fetch_user(u)
        x.send("hello")
        ctx.followup.send(f"sent message to {x.name}")
        asyncio.sleep(4)
    ctx.followup.send("sent messages to everyone")

My idea was if I wait 4 seconds between fetches + sends, I'd be able to make 30 requests to the discord api per minute to avoid being rate limited. The problem I'm facing now is that the bot stops are 45 DMs but there's no error messages in the console. How do I go about bettering this? Thank you for your time!

[lofty parcel]

You will just get your bot flagged for mass dming

Don't do that

[wise knoll]

Even with the users consent? They use a /join and that's how I get a list.
I'm not spamming them with ads or anything malicious. It's a secret santa bot so dms are somewhat necessary for secrecy reasons

[fresh sierra]

If you dm more than 50 people in less than an hour you will more likely be flagged I think

[sage tendon]

mee6 easily DMs 500 a minute

[keen relic]

Traceback (most recent call last):
  File "/home/container/.local/lib/python3.12/site-packages/discord/ui/modal.py", line 341, in dispatch
    await value.callback(interaction)
  File "/home/container/cogs/ticket.py", line 182, in callback
    await cog.create_ticket(interaction, self.name.value, self.problem.value)
  File "/home/container/cogs/ticket.py", line 293, in create_ticket
    await interaction.followup.send("You already have an open ticket!", ephemeral=True)
  File "/home/container/.local/lib/python3.12/site-packages/discord/webhook/async_.py", line 1794, in send
    data = await adapter.execute_webhook(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/container/.local/lib/python3.12/site-packages/discord/webhook/async_.py", line 222, in request
    raise NotFound(response, data)
discord.errors.NotFound: 404 Not Found (error code: 10015): Unknown Webhook```

[fresh sierra]

It’s a verified bot

[sage tendon]

just sending an error here with zero explanation isnt how you get help

[errant trout]

pls include the relevant code, but at a glance use interaction.respond instead of followup; it handles send_message/followup for you

[sage tendon]

Does https://docs.pycord.dev/en/stable/api/abcs.html#discord.abc.Messageable.typing not automatically end with ctx.respond?

This is my code, but it keeps typing for a few seconds or so after the respond

[lofty parcel]

Why are you using typing in a slash command

Defer already shows a thinking state

[sage tendon]

yea im also a bit unsure how to handle this rn because my command is kinda weird

lol

probably a dumb thought anyway
so ig typing only goes away with a .send?

What I really wanna do isn't really possible.. I kind of have just an LLM chatbot in my bot
currently i do it with a slash command and the result looks like this

which is kind of annoying to decouple in the code, the user-part and the bot-response part

[lofty parcel]

It should technically stop typing after the with block ends

[sage tendon]

i thought of doing it with pings, but idk, i kinda like doing it with a slash command

[sage tendon]

well, it stays until ~10s total have passed i think

which i thought it only does with trigger_typing

either way im really unsure whether a slash command or an @bot hello would be best
tho my users are used to the slash command now

[keen relic]

soory

[errant trout]

in my testing typing state does go away after respond

(strictly speaking discord handles that)

[sage tendon]

weird

[errant trout]

how long on average does your typing block take

[sage tendon]

2-3 seconds at most

[errant trout]

hmmm this is kinda silly but you could attempt to manually cancel the task py async with typing() as t: ... await t.task.cancel()but this shouldn't be necessary because your code takes under 5s and the respond should have it naturally end anyway

you could change it to trigger_typing and not bother with the typing block; that's only necessary if the code takes more than 10s

[sly ember]

can somebody send me the docs for peristent views? trynna make a button work after bot restart

[rugged lodge]

Here's the persistent example.

[sly ember]

thanks!

[cosmic swift]

can somebody tell me how i can set optional options for commands cause i couldnt find anything about it

[shell radish]

required = False

[cosmic swift]

thanks 🙏

[gray flame]

Hey everyone. I'm a lil dumb. Basically I'm trying to make a command that creates a new post in a forum channel. So, the command is executed in a forum post, which gets that post's parent (the forum), and creates a thread. I know usually it needs a starting message. I tried using the ...create_thread(name="test", message="test")but its not valid, neither is type. Pls help.

async def test(ctx):

    print(ctx.channel.id)
    main_channel = ctx.channel.parent
    print (main_channel.id)

    await main_channel.create_thread(name="test", auto_archive_duration=60)```

[deft kestrel]

I made a bot to give a role to online members and sync on message in a channel. It worked for my test server but didn't work for main server which is bigger. There is no perms or intents issue.

This is the code.

Does the server size have a impact on this ?

The target role in main server is a moderator role which has only 17 members

[sage tendon]

"didn't work" means what?
did you make sure to change the role IDs when changing servers first of all? hardcoded IDs are a nightmare

[sage tendon]

what error do you get

[deft kestrel]

Means it does record the message event but no role changes. Yeah I changed all server and role ids correctly

I also put a print message for on message to check if it gets triggered

[sage tendon]

does your bot have all the permissions

and is its role above the to be assigned role

[deft kestrel]

It's admin and above all roles

In both servers

Also intents is set to all and all intents are enabled

I believe this is related to server size

The main server has 215k+ members and test server had just 7 (my accounts and some bots)

[sage tendon]

do you get any error or anything at all?

[deft kestrel]

Nope

[sage tendon]

and no roles are ever assigned or removed

[deft kestrel]

Nope. All logs are clean

[sage tendon]

are you triple sure you got all the IDs right

[deft kestrel]

100%

I changed all one by one

All 4

Even added a print function to print that it actually recd the message event and it prints that it did

[sage tendon]

and do the prints inside the ifs print too?

[deft kestrel]

The for member in target role ?

I will make it print all members in mod role with their status

[sage tendon]

no, you have prints for "assigned role to name" and "removed role from name"
do those work

[deft kestrel]

Yes those work

Those were printed for test server

It correctly assigned and removed everytime

[sage tendon]

can you print your assigned_role

[deft kestrel]

Okay. I will add that and try

[sage tendon]

but you are testing it with other people and not yourself (if you are the owner) right?

[deft kestrel]

It's for moderator role. So it would assign online moderator role to PPL in moderator role who are online in main server

That's what I want to achieve

[sage tendon]

yea i know, i just wanna make sure you test it with a member that isnt you if you own the server

bots cant do a lot of stuff on server owners

[deft kestrel]

I am not owner in any. Although bot has second highest in main but it's above both roles and even actual online mods were not assigned

Also no error in console confirms no perms issue

[sage tendon]

unless you have a global error handler that eats the errors silently somehow lol

[deft kestrel]

I'll get more logging

Like log everyone in mod role and print the status. To see what's actually the problem

[gray flame]

Attribute errors or 404s

[sage tendon]

show the error

code looks fine as long as it's executed in a forum thread (or well, it works in a normal channel too I think)

[deft kestrel]

oh wow. I changed nothing. went home and just ran it again. it works now 😂

[sage tendon]

lol

[gray flame]

  File "C:\Users\Usuario\AppData\Local\Programs\Python\Python310\lib\site-packages\discord\commands\core.py", line 138, in wrapped
    ret = await coro(arg)
  File "C:\Users\Usuario\AppData\Local\Programs\Python\Python310\lib\site-packages\discord\commands\core.py", line 1082, in _invoke
    await self.callback(ctx, **kwargs)
  File "c:\Users\Usuario\Desktop\Bot Dev\vote2.py", line 16, in test
    await main_channel.create_thread(name="test", auto_archive_duration=60)
  File "C:\Users\Usuario\AppData\Local\Programs\Python\Python310\lib\site-packages\discord\channel.py", line 1319, in create_thread
    data = await state.http.start_forum_thread(
  File "C:\Users\Usuario\AppData\Local\Programs\Python\Python310\lib\site-packages\discord\http.py", line 374, in request    
    raise HTTPException(response, data)
discord.errors.HTTPException: 400 Bad Request (error code: 50035): Invalid Form Body
In message: This field is required

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "C:\Users\Usuario\AppData\Local\Programs\Python\Python310\lib\site-packages\discord\bot.py", line 1137, in invoke_application_command
    await ctx.command.invoke(ctx)
  File "C:\Users\Usuario\AppData\Local\Programs\Python\Python310\lib\site-packages\discord\commands\core.py", line 435, in invoke
    await injected(ctx)
  File "C:\Users\Usuario\AppData\Local\Programs\Python\Python310\lib\site-packages\discord\commands\core.py", line 146, in wrapped
    raise ApplicationCommandInvokeError(exc) from exc
discord.errors.ApplicationCommandInvokeError: Application Command raised an exception: HTTPException: 400 Bad Request (error code: 50035): Invalid Form Body
In message: This field is required```

[lapis dock]

You need to provide some sort of content, embed, attachments or etc
https://docs.pycord.dev/en/stable/api/models.html#discord.ForumChannel.create_thread

[proven valley]

has anyone used testcord with pycord?

[echo wraith]

It would most likely not work

[proven valley]

unfortunate

[sage tendon]

imo unit testing is just too much for a discord bot

unless its like dyno

[proven valley]

yeah but I would like to have it to make sure things are going into the data base correctly and I don't make an error when I change something, retesting everything by hand is a pain in the butt.

[sage tendon]

then verify the data in your db connector

[meager mantle]

Kind, does anyone know of any known issues related to the fact that discord.ui.Select().add_option(). It just stops the thread execution and the bot just doesn't respond to the command.

Now it is just in async def ApplicationCommand, maybe this is the reason, because in another command it works calmly from class foo(discord.ui.View).

[meager mantle]

Naturally this is not straight discord.ui.Select().add_option().

mselect = discord.ui.Select()
mselect.add_option(label=str(some<100),description=str(some<100), value=str(some<100))

[meager mantle]

It turns out add_option() doesn't do argument conversion or does it incorrectly.

I lied a bit above by writing value=str(some<100) because the int with role id was passed to value, which caused a hang. Actually adding str() fixed the problem.

[sage tendon]

because snowflake ID ints are too long for the type of integer discord uses

thats why they have to be passed as strings

@meager mantle

[lethal loom]

Hi, How can i translate commands?

[sage tendon]

like what

localisation?

[lethal loom]

when i do /rank in a english server and /rang in a german server

[sage tendon]

its never server dependent, it depends on user locale
https://docs.pycord.dev/en/stable/api/application_commands.html#discord.SlashCommand look at the _localizations here

same for command options

[lethal loom]

ah nice thx

[echo wraith]

If I want for some reason to make requests to the api in a raw manner, as long as I do it trough httpclient they are rate limited correctly ?

[silk spire]

Through pycord's HTTPClient.request method? Yes

[steady crater]

anybody have experience using environment variable to enable/disable commands? something like this:

# .env
ENABLED_COMMANDS=slash_command_1,slash_command_2

# config.py
import os
from typing import List

def get_enabled_commands() -> List[str]:
    commands_str = os.getenv('ENABLED_COMMANDS', '')
    return [cmd for cmd in commands_str.split(',')]

# bot.py
from config import get_enabled_commands

def setup(bot):
    enabled_commands = get_enabled_commands()
    bot.add_cog(AdminCog(bot, enabled_commands))

# main.py
from dotenv import load_dotenv

load_dotenv()

[sage tendon]

if cmd.strip() what

and no clue if thats easily doable with slash commands

[steady crater]

yeah not sure whats going on there lol it would just be [cmd for cmd in commands_str.split(',')]

[sage tendon]

also you dont need to import List from typing since like, 3.10 or something

[steady crater]

i didn't write this code, 4o did

i'm just asking about being able to use environment variables to toggle enabled commands

[sage tendon]

i mean sure you can

[steady crater]

are there code examples?

[sage tendon]

no, this isnt really something common to do lol

[steady crater]

is there a better way to easily enable/disable commands in production without rebuilding the image?

[sage tendon]

wdym rebuild the image

you'll most likely need this and this, and you probably need to sync commands after doing that as well

[steady crater]

docker image

i have it hosted in AWS ECS

[sage tendon]

the easier way is probably just to raise an error whenever a disabled command is called (within each command) then respond to that appropriately in your error handler
but with those functions the commands will be completely removed from view

[steady crater]

yeah that would be easier but i'd rather have the command not show up at all if possible

[lethal loom]

HI;

    async def update_rate(self, ctx: discord.ApplicationContext):
        xp_rate = await level_db.fetch_rate(ctx.guild.id)
        
        timestamp = ctx.message.created_at
        print(timestamp)```

Ignoring exception in on_application_command_error
Traceback (most recent call last):
File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\commands\core.py", line 138, in wrapped
ret = await coro(arg)
^^^^^^^^^^^^^^^
File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\commands\core.py", line 1078, in _invoke
await self.callback(self.cog, ctx, **kwargs)
File "g:\Development\Python\Xenority\extensions\level\main.py", line 455, in update_rate
timestamp = ctx.message.created_at
^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'created_at'

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\client.py", line 412, in _run_event
await coro(*args, **kwargs)
File "g:\Development\Python\Xenority\extensions\error_handler\main.py", line 121, in on_application_command_error
raise error
File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\bot.py", line 1137, in invoke_application_command
await ctx.command.invoke(ctx)
File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\commands\core.py", line 435, in invoke
await injected(ctx)
File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\commands\core.py", line 138, in wrapped
ret = await coro(arg)
^^^^^^^^^^^^^^^
File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\commands\core.py", line 1486, in _invoke
await command.invoke(ctx)
File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\commands\core.py", line 435, in invoke
await injected(ctx)
File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\commands\core.py", line 146, in wrapped
raise ApplicationCommandInvokeError(exc) from exc
discord.errors.ApplicationCommandInvokeError: Application Command raised an exception: AttributeError: 'NoneType' object has no attribute 'created_at'

[edgy nest]

applocation command has no associated message

[sage tendon]

it does, but only in some cases

[edgy nest]

you can find when an interaction was sent with discord.Object(ctx.interaction.id).created_at

https://docs.pycord.dev/en/stable/api/data_classes.html#discord.Object

[edgy nest]

message is only set on component interactions

[sage tendon]

what about a defer?

[edgy nest]

that is not a component interaction

[sage tendon]

yea but isnt the defer counted as message?

[little cobalt]

does it?

[sage tendon]

im 99% sure its included in channel.history (i know cuz of a command i have)

[edgy nest]

that has nothing to do with the context's message attr

ApplicationContext.message shortcuts Interaction.message which uses this value

[sage tendon]

hm ok weird

[lethal loom]

Why its not ehp?

    # Leaderboard Befehl mit Paginator
    @level_cmd.command(name="leaderboard", description="Zeige die Rangliste an.")
    async def leaderboard(self, ctx):
        await ctx.defer()
        guild_id = ctx.guild.id
        all_levels = await level_db.fetch_all_levels(guild_id)
        all_levels = [level for level in all_levels]
        
        if len(all_levels) == 0:
            await ctx.respond("Es gibt noch keine Benutzer in der Datenbank.", ephemeral=True)
            return

        # Sortiere nach Level und XP
        all_levels.sort(key=lambda x: (x["level"], x["xp"]), reverse=True)

        # Bereite Seiten vor
        pages = []
        for i in range(0, len(all_levels), 10):  # 10 Benutzer pro Seite
            
            description = "### Rangliste der Top-Benutzer \n\n"
            for j, level in enumerate(all_levels[i:i + 10], start=i + 1):
                user = ctx.guild.get_member(level["user_id"])
                if user:
                    ##description += f"`{j}.` {user.mention} Level **{level['level']}** XP **{level['xp']}**\n"
                    description += f"{j}. {user.mention} - Level: {level['level']} | XP: {level['xp']}\n"
            
            embed = await self.create_embed(
                ctx,
                title="XP Rate",
                description=description
            )   
            pages.append(Page(embeds=[embed]))

        # Paginator starten
        paginator = Paginator(pages=pages)
        await paginator.respond(ctx.interaction, ephemeral=True)```

[sage tendon]

because you deferred without ephemeral=True

[lethal loom]

await paginator.respond(ctx.interaction, ephemeral=True)

[lofty parcel]

Your defer isn't ephemeral

The response won't be ephemeral if your defer wasn't

[lethal loom]

ohhh

[rugged lodge]

Here's the persistent example.

[lapis dock]

@hardy abyss you need to use persistent views. See the example above

[lethal loom]

select = Select(placeholder="test", options=role_options)

is it possible to say i can select all values in teh dropdown

[sage tendon]

https://docs.pycord.dev/en/stable/api/ui_kit.html#discord.ui.Select

please read the docs

[runic sparrow]

hi

i have a question

how make this select menu?

i got it

[maiden bloom]

When a guild isn't chunked and I try to create a role in it, the attempt fails with a forbidden error even though the bot has the required permissions. Is there a way to create the role even if the guild isn't chunked?

[sage tendon]

you shouldn't get forbidden as long as you have proper permissions

[maiden bloom]

The code is pretty much just what's below, but users sometimes have to wait an hour or so for the guild to be chunked before the role is created. They send screenshots that the bot has the manage roles permission enabled and after not changing any permissions it works after waiting for the guild to be chunked.

try:
    await guild.create_role()
except discord.errors.Forbidden:
    # Send the user a message it didn't work.

[sage tendon]

and guild there is ctx.guild?

[maiden bloom]

Yes

[sage tendon]

what if you fetch the guild?

[maiden bloom]

Yeah good point, I just thought of that lol. I'll have to try it and see if it helps

[lethal loom]

https://hastebin.com/share/edafuyujoj.python

Traceback (most recent call last):
  File "C:\Users\jurek\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\site-packages\discord\ui\view.py", line 426, in _scheduled_task
    await item.callback(interaction)
  File "g:\Development\Python\Xenority\extensions\level\main.py", line 650, in add_channel_select
    embed = await Levelsys.LevelSetupEmbed(Levelsys, interaction)
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "g:\Development\Python\Xenority\extensions\level\main.py", line 607, in LevelSetupEmbed
    embed = await self.create_embed(
                  ^^^^^^^^^^^^^^^^^^
TypeError: Levelsys.create_embed() missing 1 required positional argument: 'ctx'

can where help me im confuse how can i fix this

[lapis dock]

You did not include the create embed function. My guess is you forgot to include self as the first parameter of that function

[stoic hazel]

is on_ready() but in a cog a thing? or would on_ready() just work in a cog

i have a self.something variable i'm setting in __init__ of the cog but it's not being filled properly because it runs too early (i think)

[errant trout]

can you show the code

__init__ is run the moment you load_extension/load the cog; assuming you load cogs before the bot starts, this is always before on_ready

[stoic hazel]

tbh i just moved the line from the command code to __init__ and tacked on a self. so it wouldn't repeat the line across all commands
idk if this is the right way to do this but it worked™ before and doesn't now

[errant trout]

so this'll result in logchannel being None

[stoic hazel]

yeah that's exactly what i'm getting

[errant trout]

because get_channel doesn't work before the bot's started

[stoic hazel]

(if it isn't the right way please enlighten me)

[errant trout]

if you want, you can have an individual on_ready listener in your cog using the @commands.Cog.listener() decorator

(but you should make sure to define in both init and on_ready since reloading the cog won't run on_ready again)

though, what i personally like to do is set @property for objects i frequently access across the cog ```py
@property
def guild(self):
return self.bot.get_guild(SOME_ID)

[stoic hazel]

(reloading the cog?)

[errant trout]

do you ever reload_extension

[stoic hazel]

no

what's that for

[errant trout]

it reloads a cog and the code for the listeners and commands inside it

so you don't have to restart for purely code changes, though sometimes it's finnicky with slash commands and if you want to update command metadata like names/options you still need to restart

[sage tendon]

that's just accessed with self.guild?

if so what's the point of making a property instead of just self.get_guild()

[glad garnet]

probably just looks better .guild.name rather than .get_guild(SOME_ID).name

[errant trout]

just for convenience across commands

guild was probably a poor example because most of the time you're using ctx.guild anyway

but if there's a common cached resource i'm accessing in a cog then setting properties is nice

[sage tendon]

yea but I'm wondering what the proper benefit is because get_thing() and thing() isn't really that much different
or what do you use it for

[errant trout]

there isn't, i just prefer self.thing vs self.bot.get_thing

90% of coding decisions is just preference

[sage tendon]

properties are 100% builtin right?

[errant trout]

yeah

[echo wraith]

Yes

[sage tendon]

Is this good for automatically creating commans or is there a better approach?

i tried just defining it with the decorator inside of the create_social_command function but it didn't do the trick

[sage tendon]

and im struggling to pass the cog correctly, no matter how i try it, ctx.command.cog is None

setting command.cog = self causes some weird error, and setting command._cog directly causes an error upon command invocation

overall i just cant find the intended way to do this apart from decorators which i (seemingly) cant use lolw

Okay i managed this now.. is this an IDE error? type(command_function) is SlashCommand, so its correct

ok apparently setting the cog can be achieved like this

[sage tendon]

this is inside of a function

[little cobalt]

I kinda do the same with name and description adding, I dont have to add it to every command anymore

[sage tendon]

also with decorators in a function?

[little cobalt]

I use bot._pending_application_commands for it

[sage tendon]

what

this is my entire code (for one part of the commands)

[fresh sierra]

what is ur issue ?

does everything work ?

[sage tendon]

yea, but im wondering if this is the way to do it

because it seems a bit weird

[fresh sierra]

create custom slash commands is the weird thing for me

[errant trout]

your method is correct

[fresh sierra]

like it doesnt make a lot of sens since it will be block pretty easily if its a public bot and if its not create prefix seems more logical since you will not have any limite and user will not have to restart their discord

but else if it works it works

[sage tendon]

The only reason I'm doing this because before, i had about 30 commands that were all the exact same save for a few strings

[fresh sierra]

o i see so its not to create it after the bot is running

mb then

[sage tendon]

im not creating them during runtime

[fresh sierra]

why not making a command like /action with an autocomplete after ?

[sage tendon]

yea i thought about that too but that would've required me to have the same mapping of a bunch of strings i already have like this, so i thought might as well do it the fancy way and keep the commands seperate

[fresh sierra]

ok i see

any idea how should i format my tree ?

because rn i feel like its messy

[errant trout]

eh, seems fine? unless you want to subdivide into even more folders

[sage tendon]

i'd say merge some of them

seems like a huge number

[fresh sierra]

each folder inside the slash has like

some of my issue are for exemple i have a folder name event where i put some config but not all of them are event

like there is some cog to check if there is some blocking code, some other to just wrap all the istener to check if there is an error and get it etc

[sour hedge]

yes

I forgot to respond earlier

but yes, that's how you should do it

[sage tendon]

why do you have a single folder for every command

that's not really how cogs are intended lol

[fresh sierra]

its a folder for each group of command

and i have a file for each command because i find it more clear

[sage tendon]

yea, IMO you divided everything much, much too far

at the very least you can put all subcommands into a single cog

also makes them easier to manage and you dont need multicog

[fresh sierra]

nah that will be way too much messy consiering the length of the command

[sage tendon]

unless its >300 lines its fine imo

[fresh sierra]

i have a custom thing that make everything os i dont really have to do anything with multicog or bridge

[fresh sierra]

it is more

also some of my other issue is my yaml file that is 10k line...

[sage tendon]

wtf

for what

[fresh sierra]

10k for french, then 10k line for english etc

this is like all the answer of the bot

7k sorry

like this is hell to maintain

maybe i can do for each folder of subcommand something like that :
one folder for all translation for this group
one folder for all view
one file for each utils functions
and other like normal thing

but even this seems big asf

[little cobalt]

what?

[fresh sierra]

7k line of text per lang

my bot has a lot of command

[echo wraith]

Localization

[frail basin]

If an ephemeral view has it's timeout set to None, await view.wait() is called, and the user presses the Dismiss message button, what happens to the function that called .wait()? Will it hang forever?

Example code:

@discord.slash_command(name="test")
   async def test(self, ctx: CustomApplicationContext):
       view= TrueFalseSelectView("Yes", "No")
       await ctx.respond("Test", view=view, ephemeral= True)
       await view.wait()
       print(view.final)

class TrueFalseSelectView(discord.ui.View):
    def __init__(self, true_button_label: str, false_button_label: str):
        super().__init__(timeout=None)        
        self.final: bool | None = None

        false_button = discord.ui.Button(style=discord.ButtonStyle.red, label=false_button_label)
        false_button.callback = self.false_callback
        self.add_item(false_button)
        
        true_button = discord.ui.Button(style=discord.ButtonStyle.green, label=true_button_label)
        true_button.callback = self.true_callback
        self.add_item(true_button)

    async def false_callback(self, interaction: discord.Interaction):
        await interaction.response.defer(ephemeral=True)
        await interaction.delete_original_response()
        self.final = False
        self.stop()

    async def true_callback(self, interaction: discord.Interaction):
        await interaction.response.defer(ephemeral=True)
        await interaction.delete_original_response()
        self.final = True
        self.stop()

Based on my testing, it never returns, however I'd like some additional confirmation on this.

And if this is indeed the case, is there some sort of workaround to check whether the message has been Dismissed?

[little cobalt]

for what do you need .wait()?

[frail basin]

It would be a universal "Confirm" / "Cancel" view for warnings.

Take the following as an example use case

[sage tendon]

btw you can use ctx.respond as a shortcut

[lapis dock]

You can check if the message still exists. By getting the interactions original response. Ideally you would have a timeout instead though. Ephemeral messages are supposed to be temporary so you should not treat them as persistent. I don't think you can check if the message exists after 15 minutes as this is when the interaction token expires (which is used to get the message)

[frail basin]

Timeout it is then. Purely out of curiousity, would my code in its current state leak little bits of memory with each Dismissed view?

[lapis dock]

Not 100% sure but I think so. Unless the message gets marked as unavailable some other way later on

[frail basin]

The intellisense does not like it for some reason, and I can never spell ephemeral

[sage tendon]

how lol

[frail basin]

¯_(ツ)_/¯

[sage tendon]

man vsc really is incapable of even the most basic IDE task

[lapis dock]

Same on the spelling part XD

[sage tendon]

same

ephemereal usually

[lapis dock]

Ephermial I add an I for some reason.

[frail basin]

I guess it does not like that its a property?

[sage tendon]

not even sure why it is

like its a function why is it a property lol

oh nvm makes sense

[fresh sierra]

is there an error for a subcommand that doesnt exist being invoke ?

like command not found but subcommand not found

[pulsar ferry]

I am getting this when issuing a command but then it seems like the command still runs its just taking a bit? is there a default time pycord uses when issuing a command that can be modified?

[little cobalt]

do you use ctx.send or ctx.respond?

[pulsar ferry]

yes the command does work in some scenarios, but when there is more processing in backend that has to happen it seems to timeout

[little cobalt]

?

[pulsar ferry]

ah sorry thought you were asking if using either. yes using ctx.respond @little cobalt

[lapis dock]

Discord gives you 3 seconds from the moment the user runs the command until they expect to recive the response.
For longer running commands you can use ctx.defer(). This tells discord that you need extra time and then allows you to have 15 minutes to respond. You should defer at or near the top of your command callback as the defer needs to be received by discord in 3 seconds as well.

[lapis dock]

does not look like it

[pulsar ferry]

Exactly what i needed, thank you... follow up though as to the question now in trying to establish why the command is taking so long

The command I have looks into previous threads to see if a specific link was already included in the embeds posted in each thread. I only want it to look at threads that have not been archived, and the default archive time I have the threads is 24 hours. IT seems like even with this code though it may be searching through every thread, including the archived ones... is there something im missing here?

async def find_existing_thread(channel, link):
    for thread in sorted(channel.threads, key=lambda t: t.id, reverse=True):  # Sort by ID descending
        if not thread.archived:
            first_message = await thread.fetch_message(thread.id)
            embed = first_message.embeds[0] if first_message.embeds else None
            if embed and any(f.name.lower() == "claim" and f.value.lower() == f"[click here]({link})".lower() for f in embed.fields):
                logger.info("Found existing thread with matching claim link.")
                return thread
    return None

[sage tendon]

you should use https://docs.pycord.dev/en/stable/api/models.html#discord.Thread.starting_message and not make an API request for every thread to get its message

[pulsar ferry]

Does this work if the first message is only an embed? following code prints None

    for thread in sorted(channel.threads, key=lambda t: t.id, reverse=True):  # Sort by ID descending
        if not thread.archived:
            first_message = thread.starting_message
            print(first_message)

[sage tendon]

well if the message isnt in cache you need to fetch it

[pulsar ferry]

arent i then hitting the api again though?

[sage tendon]

yea but you should default to using cache

[pulsar ferry]

so something like this?

first_message = await get_or_fetch(bot, "message", thread.id)

[sage tendon]

so just do first_message = thread.starting_message or thread.fetch_...

[pulsar ferry]

Neither of above worked, this seemed to do it though

first_message = thread.starting_message or await thread.fetch_message(thread.id)

That being said though, still taking very long to run and i think its still going through the archived messages

[sage tendon]

oh yea you need to await

[sage tendon]

i mean then print the thread names and check lol

[lapis dock]

https://docs.pycord.dev/en/stable/api/models.html#discord.Guild.active_threads
This gets a list of all active threads in a guild, not specific to a channel. I dont think discord provides a channel specific list

[sage tendon]

i mean their code should work no?

[fresh sierra]

yeah thats what i thought, thanks tho

[pulsar ferry]

yeah i was just doing this as i typed it. either its automatically not fetching the archived, or nothing is actually getting archived

because this didnt print anything...

        elif thread.archived:
            print(thread.id)

[lapis dock]

channel.threads gives you all the threads in the bots cache

oop wrong reply

[sage tendon]

print the thread names and check if it prints any unarchived ones

[pulsar ferry]

Tried that above with the elif statement

Nothing printed

Is there a way in the discord UI to determine if a thread is archived? …perhaps nothing is actually getting auto archived?

Because I’m still seeing everything even that’s older than like 2 days and I have it set to 24 hrs

[lapis dock]

In forum channels: It shows under the "Older Posts" header

[sage tendon]

no, print all thread names you get in your original code please

[lapis dock]

Also I dont know how reliable the auto archive is.

[sage tendon]

maybe it doesnt even apply to forum posts

[lapis dock]

It does, we have it here

[pulsar ferry]

yeah i dont have that. - should "close post" be already selected on archived posts?

ah shit - baby just woke up gotta run, will be back in a bit to keep tinkering thanks so far

[lapis dock]

👍 Yeah, "archived" is the API name "closed" is the public name

If it is archived it will say "open post"

[pulsar ferry]

Yeah so not getting archived then…guess the auto archive isn’t working :/. alright at least I know the issue now will try and figure out why that’s not working later

[lapis dock]

actually you might be right. I dont think anything more recent than a month is archived here and the default is 3 days

[sage tendon]

threads and posts should have been 2 seperate classes

not sure if discord themselves differentiate it

[lapis dock]

Discord treats them the same.

[sage tendon]

archived for forum posts might mean forcibly closed

[lapis dock]

Posts like this one were probably not forcibly closed but are still archived
https://canary.discord.com/channels/881207955029110855/1306742100439859281

[sage tendon]

it is closed though

[lapis dock]

It is closed but I dont think anyone closed it I think it was automatic

Unless someone closed threads not using the /close command

[sage tendon]

i do that rarely

[shell radish]

oh i remember manually closing that rhread because the bot was down

[lapis dock]

... well that was a big coincident.
Do you do that a lot? There are a bunch of other threads that follow the same pattern

[shell radish]

not recently

[edgy nest]

the older posts section is all archived posts

archive is set to 3 days

[lapis dock]

yeah, but why are posts that are more than 3 days old not auto archived. I just closed a bunch but they were inactive for 20+ days

[edgy nest]

im not sure why some threads are still open

discord stroke ig

[lapis dock]

It happens a lot

[sage tendon]

remember

its a feature added past discord's prime

so it HAS to have some bugs in its core functionality

its the law

[lapis dock]

[pulsar ferry]

Hey checking back in here - seems like this is an actual issue with discord itself not py-cord?

Could it have anything to do with the “auto_archive_threshold” arg?

and if not, does the api call to manually archive the thread still work? Maybe we just need a background job every hour to check total age of the thread and archive accordingly

[lapis dock]

Yeah, it seems that there is something strange with how discord closes threads. A manual API call to archive would work.

[pulsar ferry]

Alright I’ll try writing something up when back at pc later tn

Perhaps it’s something that should be built into pycord though? Since all the args for setting thresholds prob aren’t working for everyone using them

[echo wraith]

Is there a possibility to have an attachment as a parameter to an ext.command just like slash or should I iterate over ctx.message.attachments ?

[fresh sierra]

you can just do like so :
async def upload(self, ctx: LumabotContext, file: discord.Attachment):

[echo wraith]

that didn't work :/ I think I might have messed up smth

oh well

in the end I am just iterating over the attachments so I can have my own logic

[fresh sierra]

strange

maybe it does not work with more than 1 arg

[sage tendon]

1. why prefix

[echo wraith]

Idk. But users want it so I don't really care, I give both options with, most of the time, more options for slash, and they do whatever they want

[pulsar ferry]

Alright not entirely sure what i did, was just playing with archiving manually and it seems like ive now entirely hidden a few channels 😅

I can still get to them by doing like #rules and clicking it but its hidden from the sidebar... how do i get it back?

[sage tendon]

did you enable onboarding

or just disable "show all channels"

[pulsar ferry]

i do have onboarding enabled but where is that show all channels setting?

because now its not even showing in the 'browse all channels'

[lapis dock]

Did you somehow enable server guide?

[pulsar ferry]

ah okay that did it - but how do i force those channels to always show like how you guys have the #rules in info

my rules channel has this in the settings...

and theres no 'dont hide' option

[sage tendon]

enable them as default in onboarding

[sage tendon]

thats for threads

[lapis dock]

Is your rules channel a forum channel? nvm

[pulsar ferry]

they are

[pulsar ferry]

i dont think so? lol

[sage tendon]

yea you enabled it as rules channels in server guide

it wont show as channel then

[pulsar ferry]

the same issue is happening for FAQ though

[sage tendon]

also part of server guide

disable server guide

[lapis dock]

(or remove those channels)

[sage tendon]

not sure you can remove the rules

[pulsar ferry]

ahhh you guys have server guide and SEPERATE channel for rules

thats really confusing... discord has like 3 seperate places for 'rules' and none of them are linked

[lapis dock]

we did here

[sage tendon]

really just 2

nope

you need to make a speerate channel

[pulsar ferry]

onboarding, rules channel, and then a seperate server guide rules

[sage tendon]

server guide uses the normal rules channel

[pulsar ferry]

its not possible to make the server guide not hide the channels?

[sage tendon]

no

unless you duplicate them

well thats only true for the rules channel

i think

[pulsar ferry]

ah alright think i got it set up how i want... how do you remove the events and other crap up there?

[sage tendon]

cant

and thats not really pycord related now lol

[pulsar ferry]

yep fair - i saw it wasnt in your guys thing at the top so figured id ask

1 last pycord question... is this the most efficient way to get a channel by its name?

async def find_channel_by_name(channel_name: str):
    guild = await get_or_fetch(bot, "guild", GUILD_ID)
    # Look for a text channel with the provided name
    channel = discord.utils.get(guild.channels, name=channel_name)
    return bot.get_channel(channel.id) if channel else None

[sage tendon]

the first question is why do you need to do that

[pulsar ferry]

trying to go through all my code now and optimize where i dont need to be fetching every time

its a helper function.. i have multiple commands that can be invoked that then should write to specific channels... i could hardcode the channel IDs i guess but wanted it to be somewhat dynamic in case i change channels later

[sage tendon]

only mods see it if no event is active

im pretty sure you can just remove the bot.get_channel line since discord.utils.get should already return the channel object

[pulsar ferry]

ah yep thanks!

[sage tendon]

same for the if channel else None

[pulsar ferry]

async def find_channel_by_name(channel_name: str):
    guild = await get_or_fetch(bot, "guild", GUILD_ID)
    # Look for a text channel with the provided name
    return discord.utils.get(guild.channels, name=channel_name)

works thanks!

[sage tendon]

but keep in mind getting channels by name is really not ideal

[pulsar ferry]

Yeah i know - this is a custom bot just for my 1 server though so cutting a few corners

[errant trout]

just note that fetch here is pointless because a guild from fetch_guild doesn't include channels

(a small bot like yours will always have the guild cached, so you rarely need to fetch anything really)

[pulsar ferry]

Well wouldn’t it always just use get first then if cached with this function?

[errant trout]

right, which is why im saying there's not much reason to use get_or_fetch

[pulsar ferry]

Yah I guess my style is more “have and not need” than “need and not have”

There’s no harm in having as it’ll use cache and if it doesn’t need to fetch

[hardy abyss]

How to make slash_command only visible in a certain guild

[errant trout]

(my point being that if it ever fetches, your system breaks)

[errant trout]

add a list of guild_ids to the command decorator

[round heart]

That's a code change. Say you have a bot that is in multiple guilds (I don't, but let's pretend); is there a way to selectively enable/disable a command for a particular guild?

[errant trout]

no

[round heart]

Is that a Discord limitation, or a pycord one?

[errant trout]

well for starters, discord has no concept of "disabling" a command

it either exists or it doesn't

when we set guild_ids, it creates a new individual command on every guild through their individual endpoint

[round heart]

So when you refresh commands with Discord, you have to send them all? There's no way to say 'this specific command is no longer available'?

[errant trout]

there's a few different sync methods

pycord by default does a bulk sync and overwrite

the important part is if for some reason a command is "deleted" or not included in an overwrite sync, you'll also lose any user customizations such as permissions

which obviously isn't ideal for a public bot

[round heart]

I'm not saying it's not complex, but is it theoretically possible - given available methods from Discord - to manage available commands per server on-the-fly? Whether via a bulk sync or with some of the other type of sync you hinted at

[errant trout]

oh it absolutely is

just a pain lol

ideally pycord v3 will also make it easier, the current sync system sucks for anyone trying to even remotely do their own thing

[round heart]

It seems like such a giant oversight to not allow permissions to be added with command registration. It's not like they don't have Role settings on every other model.

[errant trout]

well it wasssss

but when permissions v2 came, it was moved to user control

thus if you want to programatically set permissions you need oauth instead (which is why large scale dashboard bots can do it)

[round heart]

mm, that makes sense (but is overly complex for smaller bots, especially bespoke ones.)

I can just imagine there's been a non-zero number of times privileged commands were accessed by commonfolk because the integrator forgot, or didn't get to it in time

[errant trout]

at least we do have default perms which is a decent enough compromise, but yeah at the time it was a pretty big deal

[maiden bloom]

Are there any downsides to using py-cord[speed]?

[little cobalt]

for what is speed even?

[maiden bloom]

I found it on the pypi page

[echo wraith]

I think it's uvloop, or what is it ?

[edgy nest]

like 3 extra dependencies

that are intended to make the library run marginally faster

[maiden bloom]

Sounds worth it to me

[pulsar ferry]

Sorry, why would it break?

[errant trout]

Because fetch_guild doesn't return channels

That was my original point

[pulsar ferry]

Ah

Okay maybe I need to update that one then

[drowsy grove]

I made my own help command and added it to my bot’s help_command setting. However, it isn’t responding with it. Does it take a while to load the command like how slash commands do?

[sage tendon]

did you restart discord and the bot

also i think help_command is kind of a remnant of prefix times, not sure how well it works with slash

[drowsy grove]

I don’t have it set up as a slash, I was just wondering if it had the same load bearing as slash does when it comes to being a global command. I restarted the bot, but not discord. Everything else didn’t need a discord restart so tbh I didn’t think about that.

I’ve tried even getting just the Minimal Help Command to appear, but it just won’t register the command

[lapis dock]

Prefix commands are not registered to discord so you only need to restart your bot. If it is not working immediately after restarting your bot you have something wrong in your code.

It is recommended to use slash commands at this point in time though.

Ah, i see you talked about it in #general

[drowsy grove]

Yeah, I'm just going to bail on prefix commands. It's good to know now that they're useless. It's getting easier as I translate my commands. Finished file 1/7 so far, and was able to cut out a lot of commands since a lot of it was 'help adjacent'. Makes it so much easier since discord will handle that.

[sage tendon]

everything is easier with slash commands

absolutely everything

[drowsy grove]

Yeah, I'm very quickly learning that lol

Each feature is reducing in code by a large margin

So much error handling is automatically taken care of if I do typing

[sage tendon]

you don't even have to do typing lolw

but you should

[drowsy grove]

I'm def going to do typing for my own sanity

[pulsar ferry]

hey sorry what is the proper way you are saying i should query the guild then?

[errant trout]

just bot.get_guild because it's always cached in your scenario

[stray cape]

Hey guys I made a bot that uses views (with the bot ui kit), but it seems like that if I run my project on my local computer, it runs on time, it doesn't have any delay or anything. But when I run it on my vps, it has slightly delay with timeout, it timeout's a minute later what might be the cause?

[lapis dock]

How are you setting your timeout?
I sounds like you mean that the timeout is actually 1 min 15 seconds on your VPS or similar?

[sage tendon]

VPS system time might be wrong would be my first guess

[stray cape]

This is how I create my views and set my timeout:

class AnnounceView(discord.ui.View):
  def __init__(self, someone: discord.Member, *args, **kwargs):
      super().__init__(timeout=600, *args, **kwargs)
      self.someone = someone

view = AnnounceView(...)

its like a minute delay

[sage tendon]

check if the time of your vps is correct

[stray cape]

its more a shared hosting but does this have to do something with the timezone difference?

[sage tendon]

doesnt matter if its shared or not, check the time of the server lol

and no, timezones are 30m minimum

[stray cape]

what specifically do you want me to check about the time of my vps?

[sage tendon]

whether its correct or off by a minute lol

[stray cape]

1 hour delay thats it

luxkatana@nest:~/cautious-computing-machine$ date
Mon Dec 30 10:06:24 PM UTC 2024

[sage tendon]

so the minute is correct? hm

[stray cape]

the timeout is dynamic by the way, but it has some pattern

[lapis dock]

That is strange. If the VPS time was off it should not effect something that happens X seconds in the future as that is a duration not a specific time.

[sage tendon]

it might have taken the time from which the message is sent as a timestamp, thats how i imagined it

[lapis dock]

The timeout resets every time you interact with the view. Are you aware of this?

[stray cape]

hm?

I sent the view already, at that point the timeout time should go off, right?

This is what I mean (the time currently is 23:17 or 11:17 PM for you dumb americans)

This is how the timeout is decided:

Current time + 10 minutes
current time seconds will be zero so that'll timeout when it is exactly after 10 minutes

[stray cape]

So you mean that, if someone interacts with any of the buttons or selects, then the timeout will reset?

[lapis dock]

Yes the timeout is how long the view needs to be inactive before it does its timout things

[stray cape]

Oh?

I want to make this behaviour that it should last for n amount of seconds how can I do that then?

[echo wraith]

With a task

[sage tendon]

run view.stop() after that amount of time

and before that, view.disable_all_items() i think

so the buttons become inactive

[stray cape]

so I should be cnstantly checking the time using time.time()?

manually

[sage tendon]

no lol

hm actually not quite sure how id do it tho rn

probably running a function in a new asyncio runner that then waits 10m exactly and then disables all items and stops the views

[stray cape]

but the time that it should disable is dynamic

[sage tendon]

tasks can have dynamic timings

[echo wraith]

Yeah

Wait I sent an example some time ago

[stray cape]

@tasks.loop

[echo wraith]

No not that

An asyncio task

#1132206148309749830 message

[sage tendon]

oh i also thought you were talking about pycord tasks lmao

so yea we mean the same thing

[stray cape]

cant I just wait using asyncio.sleep for the amount of time directly? And disable the items of the views after the sleep?

[echo wraith]

Yeah but that would block your code execution

[stray cape]

oh that it'll block access to another commands?

[lofty parcel]

It doesn't block the thread

[stray cape]

thats impossible right because its asynchronous.. right?

if you use time.sleep then it will block?

[lofty parcel]

asyncio.sleep doesn't block the bot

[lofty parcel]

Yeah

[stray cape]

so it shouldnt be a problem to use asyncio.sleep ig

[sage tendon]

It does

[lapis dock]

On my stuff like that I just check the time when someone interacts and if it is passed the end time I say "sorry to late" then disable/stop the view. This works well for longer timeouts

[sage tendon]

because it still pauses code execution on that branch

so, unless im wrong, the view will never respond to events or your command code wont resume (assuming something happens in the command used to send that view)

[stray cape]

after 10 minutes, it'll do another stuff (creating channels, handling other views and stuff)

[sage tendon]

might be wrong tho

[stray cape]

my project only consists of just one task not any other command :p

i should be trying stuff out ig, but thanks guys!

i read my role wrong thought it said transistor

[lofty parcel]

I think it doesn't block the view either

I remember using it somehwrre

[errant trout]

asyncio.sleep will only block the code after it, nothing else

(that's the entire point)

[sage tendon]

yea but i worried about the init never finishing and thus that branch of code never returning the command until after those 10m but since there's apparently no command is fine

[toxic echo]

quick question, what feature is this bot using to generate profiles programmatically?

[torn barn]

Webhooks

[pulsar ferry]

Hey guys qq - I am using autocomplete for an arg on one of my commands, but the command is still able to be executed with a value taht is not in the list... is there a way to force the user to only be able to execute by using a value from the autocomplete?

@bot.slash_command(guild_ids=[GUILD_ID], description="Share a promo the community!")
async def promo(
    ctx,
    name: discord.Option(str, "specify the name", autocomplete=discord.utils.basic_autocomplete(name_autocomplete)),
):

[sage tendon]

use options

[pulsar ferry]

I cant because there are more than 25 or whatever the limit was

[lofty parcel]

Then you will need to check against the autocomplete options and return an error message

Or use multiple select menus

[pulsar ferry]

yep okay thats fine i can write a manual check just didnt know if there was a quick arg that could be set

[drowsy grove]

New day, new question. I have a global error handler to handle (obviously) errors. I also have command specific errors. My global handler is interfering and dealing with errors that only my command specific error handlers should touch. Is there a way to prevent this?

[sage tendon]

are you actually setting handlers or do you just do the handling right in the command code

if you set handlers, you can add this to the top of your global handler and it should work

if ctx.command.has_error_handler():
    return

[pulsar ferry]

Hey guys another quick one - is it possible to respond as ephemeral when returning the function?

I am trying to do something like this but it seems like the bot response is still showing public to everyone

    msg = f"[{category.capitalize()} Posted]({thread.jump_url})"
    logger.info(msg)
    return await ctx.respond(msg, ephemeral=True) 

If not, is there a way to hide the args that were passed to the command when clicking on used <command> ?

[sage tendon]

you need to defer ephemerally if you wanna respond ephemerally

[pulsar ferry]

oh so every function essentially then that has a response should have the first line be the ctx.defer?

[sage tendon]

no

but you are deferring there, no?

[pulsar ferry]

ohh wait you mean pass the ephemeral arg as True to the defer?

[sage tendon]

yes

[pulsar ferry]

got it thanks!

[drowsy grove]

I’m setting error handlers

[sage tendon]

.

[drowsy grove]

Thanks!

I had something similar that worked with prefixes, it was a hasattr(command, ‘error’)

It broke once I swapped to slash commands

[pulsar ferry]

is my server subscription tier bugging out, or does discord upcharge the fee on ios devices?!

[echo wraith]

They do

I'm pretty sure

[sage tendon]

yep

ios purchases are more expensive

[pulsar ferry]

wow okay thought i was going crazy with messing up a tier

[pulsar ferry]

Anyone know if this actually works either or do I need to write a custom greet message in pycord? Have people signing up and no message :/

I wonder if its because there is a week free trial, so maybe after the trial ends it procs?

[sage tendon]

yea i guess so

since it does say "purchases"

[pulsar ferry]

but they do put in the credit card and everything when they click 'subscribe'

[sage tendon]

not charged tho

[pulsar ferry]

hmm yah i guess

have an actual py-cord question now... are there any examples of how to determine when someone joins, what invite they used and then subsequently who created that invite (i.e. who invited them?) ..Thinking of doing a referral type system

[sage tendon]

only indirectly

you need to actively track the number of uses of each invite and recheck after a join

but joining is just on_member_join

[pulsar ferry]

Yeah so i worked up something like this with that thinking, was hoping there was a better way though

@bot.event
async def on_member_join(member):
    guild = member.guild
    # Fetch the current invites from the server
    invites = await guild.invites()

    # Find the invite with increased uses
    for invite in invites:
        if invite.uses > 0:
            inviter = invite.inviter
            if inviter:
                print(f"{member.name} joined using invite by {inviter.name}.")
                # Store inviter-invitee relationship
                inviter_tracker[member.id] = inviter.id
            break

[sage tendon]

yea thats how you need to do

[pulsar ferry]

but this would obv need some tweaking to check against a cache

[stray cape]

hey guys I'd like to know when I could use persistent views

and when not

[sage tendon]

when you need them past a bot restart

[drowsy grove]

Ok I’m losing my mind here. How do I get the SKU id for a premium shop item??? I want to verify that a user has the correct SKU, but have no idea how to do so because discord does NOT explain how to do so in their documentation. I’ve been staring at it for literal hours.

Am I just missing something??

[fresh sierra]

Isn’t it entitlement.id ?

And user_id for the other v

[lapis dock]

I think all purchases on the mobile apps are more expensive. Tell your users to buy on PC or discord web if they only have a phone. I know there is a fees breakdown in one of discord minimization writeups

[fresh sierra]

This is not gonna work

You need to compare the old invite count to the new one, here it will just always show the same person who has an invite greater than 0

[drowsy grove]

That’s what I thought, but I don’t know where to get that from. Is it from the ApplicationContext being passed through? Via obtaining the bot/client from the context for the bot’s SKU and obtaining the user for the user’s SKU? In that case, how do I know which SKU belongs to the subscription I am trying to reference? From what I can tell there are no names within an entitlement object, just id’s. I haven’t looked at a SKU object, so maybe it’s there, but I find it weird that I can’t find the SKU id before having to grab it via the application command. That would mean I’d have to call the method to read all entitlements available and then pick and choose until the right entitlement is selected

[rugged lodge]

Searched in pycord

discord.Guild.entitlements

[edgy nest]

If you open the edit shop menu the sku ID is the last ID in the url

you can also use Client.fetch_skus

[drowsy grove]

This is the exact answer I was looking for. Thanks

[drowsy grove]

Does this grab the entitlements that are in the guild? Or the one that the invoking user has within the guild?

[edgy nest]

that would grab entitlements attached to guild subscriptions

[drowsy grove]

Ah ok. I think I have mine set up to user subscriptions instead. Would I instead just grab the user from context and do user.entitlements, giving the same result?

[edgy nest]

yes

[drowsy grove]

Thanks 🫡

[edgy nest]

both of these use Client.entitlements, they just specify the guild/user parameters respectively

[drowsy grove]

Cool. So if I did Client.entitlements it would pull all entitlements, regardless of guild/user?

[edgy nest]

yes

[drowsy grove]

Perfect

Thanks a lot!

[edgy nest]

additionally if you're using interactions, Interaction.entitlements should return everything that applies to that interaction

(both user & guild)

[drowsy grove]

I am indeed using interactions. I have application commands returning ApplicationContext, which give me access to the user and the interaction within it

So either way I can grab the entitlements

I just had no clue where to grab the SKU I need for comparison

[edgy nest]

Inter.entitlements is preferred because it's not an extra API call :3

[drowsy grove]

Even ChatGPT told me it was displayed on the discord applications page 😭

[drowsy grove]

I’ll keep that in mind then

[edgy nest]

yes but actually no

[drowsy grove]

Yeah. It displays it, just not next to the SKU (which is what it said lmao)

[edgy nest]

it does in analytics but you can't copy it :p

[drowsy grove]

Weird

[edgy nest]

i guess they expected people to just fetch it over the api

[drowsy grove]

Is there a way in discord to pin that response? I won’t be able to get around to the fix until day after tomorrow

[drowsy grove]

Yeah I think so

I just don’t want to open another API connection

I’m lazy

So lazy that I already started making my own database for entitlements instead of fetching from the API

[edgy nest]

yeah

you can cache them in memory, you just have to fetch them on startup

then listen to events

[drowsy grove]

I have event listeners up for entitlements, just didn’t wire them to the DB yet. Going to do this solution instead

[pulsar ferry]

Yeah I’m going to try and write something today - was hoping someone else in the community already had an example to work off of

[fresh sierra]

Check the invite create to save all of the invite with there use

Then when a member join check which invite is greater than the one you save

And then you will now which one invite the member

[ionic wasp]

whats the interval at which a bots latency refreshes?

as per bot.latency

[sage tendon]

logically it would be every API call or even every heartbeat ig

[stray cape]

hey guys got a discord.File object how can I get the link of it to use it for example as an image for an embed or as thumbmail?

[sage tendon]

you need to upload it as attachment to the message then make the image link a certain thingy

wait, there's a tag for it

[sly karma]

f = discord.File("some_file_path", filename="image.png")
e = discord.Embed()
e.set_image(url="attachment://image.png")
await messagable.send(file=f, embed=e)```

[sage tendon]

yes that

[stray cape]

oh okay

thanks!

do I have to close the file tho?

[sage tendon]

show your code, depends on what you do

[stray cape]

Currently I am taking a file from a slash command

[sage tendon]

yea show your code

[stray cape]

async def apply_as_helper(ctx: discord.Interaction, image: discord.File) -> None:
    if SEARCHING_FOR_HELPERS is False:
        await ctx.response.send_message("Helper applications are closed.", ephemeral=True)
    else:
        STAFF_CHAT = await ctx.guild.fetch_channel(1323056105835991050)
        embed = discord.Embed(title=f"{ctx.user.name} is applying for helper", description=f"{ctx.user.mention} is applying for helper")
        await ctx.response.send_message("Cool, thanks for applying as a helper, we're gonna make our decision later!", ephemeral=True)

[sage tendon]

ctx is not discord.Interaction

and I'm pretty sure files uploaded to slash commands are discord.Attachment

and you can use ctx.respond() as a shortcut
and :3 you shouldn't fetch the channel but rely on cache (get_channel)

[stray cape]

ctx is discord.ApplicationContext?

[sage tendon]

yes

[stray cape]

is there a small chance that the channel is not cached?

and that it returns None?

[sage tendon]

and no if you get an attachment from a slash command like that you don't have to close anything

[sage tendon]

not if your bot is only in one or a few servers
typically only becomes a problem with big bots

[stray cape]

ohh alright

[sage tendon]

also idk what ctx.user exactly points to but usually you use ctx.author

sorry for being annoying I love to nitpick

[edgy nest]

ctx.author is a shortcut for . user

[sage tendon]

ok

[stray cape]

is there a way to remove webhook messages of a channel?

[sage tendon]

delete them like normal messages

[pulsar ferry]

For anyone else who may be interested... this seems to be working

server_invites = []

@bot.event
async def on_ready():
    # Cache Server invites
    global server_invites
    guild = bot.get_guild(GUILD_ID)
    server_invites = await guild.invites()

@bot.event
async def on_member_join(member):
    global server_invites
    new_invites = await member.guild.invites()
    for old, new in zip(server_invites, new_invites):
        if old.code == new.code and old.uses < new.uses:
            update_referral(member.id, new.inviter.id) # Custom function to do something with the data
            break
    server_invites = new_invites  # Update cached invites

[sage tendon]

why do you declare the global twice lol

or rather why make it global at all

oh bc its not in a cog

[fresh sierra]

You lack one thing, if a invite was not saved it will not work

So you will have to update the invite also during invite create

[pulsar ferry]

yeah i actually just just realized that after posting... trying something like this now

@bot.event
async def on_member_join(member):
    global server_invites
    new_invites = await member.guild.invites()
    old_invites = {invite.code: invite.uses for invite in server_invites}
    for invite in new_invites:
        if invite.code in old_invites and invite.uses > old_invites[invite.code]:
            update_referral(member.id, invite.inviter.id)
            break
    server_invites = new_invites

and i think the assumption i still need to add is that IF there is a new code, that is the one to use... not sure how to account for if more than 1 code is created though before someone new joins

[fresh sierra]

Just also use the invite create event to update the event

[pulsar ferry]

oh wait great idea... just use the event create to update the server_invites

[fresh sierra]

Basically what you can do is
invite_create and invite_delete => update ur list of all invite.
member_join : check which invite is greater and update the list of invite

And consider saving the invite inside ur on_ready too

[pulsar ferry]

something like this?

async def refresh_invite_cache():
    global server_invites
    guild = bot.get_guild(GUILD_ID)
    server_invites = await guild.invites()

@bot.event
async def on_ready():
    logger.info(f"Caching server invites...")
    await refresh_invite_cache()

@bot.event
async def on_invite_create(invite: discord.Invite):
    await refresh_invite_cache()

@bot.event
async def on_invite_delete(invite: discord.Invite):
    await refresh_invite_cache()

@bot.event
async def on_member_join(member):
    global server_invites
    new_invites = await member.guild.invites()
    old_invites = {invite.code: invite.uses for invite in server_invites}
    for invite in new_invites:
        if invite.code in old_invites and invite.uses > old_invites[invite.code]:
            update_referral(member.id, invite.inviter.id)
            break
    server_invites = new_invites

[edgy nest]

invite create & delete you can just manually update the cache instead of refetching all of them

since only one invite changed

[pulsar ferry]

but i wouldnt get the uses if i manually added the code...would I? ah wait nvm i could add the object

but feels safer to just pull a refresh?

[edgy nest]

safer technically but unnecessary

[echo wraith]

This is very unreliable and prone to issues

[sage tendon]

f

[pulsar ferry]

latest but if you have more recommendations please share! #1132206148309749830 message

[echo wraith]

If you want to do it that way you'll need to add a bunch of error handling including for cases when for some reason your db is not in sync with the guild invite information and a database to work between restarts.

[fresh sierra]

There is no db use

[echo wraith]

For the guild invites...

[fresh sierra]

And he used the on_ready to sync again between restart

For a single guild bot this seems to be good

[echo wraith]

If it's a single guild then yeah ig...

[fresh sierra]

Because I don’t really see any scenario where it will not works

Or it will not work for 1 member if discord has an issue

[echo wraith]

Still needs error handling imo, e.g. server is a bit unresponsive and multiple users join at similar times

Race condition basically

[fresh sierra]

Yeah, for a single bot it should be good but ofc it can be better

[echo wraith]

The race condition issue doesn't depend on the number of guilds. But yeah I get the fact that it's for a single guild. Although imo that tracking method is still hacky and duct taped

[echo wraith]

I would make it so that server_invites = new_invites is in a finally block or else it could link a user to the wrong invite if an issue happens for another user at first

At least

Then have some kind of lock on server_invites and run the appropriate logic with that lock

[pulsar ferry]

Sorry stepped away for a bit thanks for the tips here - yes single server use only but lock is a good idea for race condition even if it’s an edge case

[ionic wasp]

can i asign custom values to slash commands directly from
@discord.slash_command(name="", description="")

so that these custom values show up in bot.application_commands

[echo wraith]

What do you mean by custom values ?

[ionic wasp]

@discord.slash_command(name="", description="", custom_value="banana")

So then in bot.application_commands i could do
command.custom_value and it would return banana

[sage tendon]

for what purpose

[echo wraith]

uh

[ionic wasp]

a commands webpage, id like to mark owner only ones etc

[sage tendon]

I mean you should be able to do command_function.attr = " "

right under the definition of the function

[ionic wasp]

i dont think that works

[echo wraith]

that works

But I'm unsure why you would want to do that

[sage tendon]

to say it doesn't work without even trying is wild

[echo wraith]

@ionic wasp Do you mind sharing a bit more about your exact use case so we can find maybe a better implementation ?

[ionic wasp]

^

[sage tendon]

if you address the functions like this it'll be SlashCommand objects btw so just make sure you're not overwriting a preexisting name

[ionic wasp]

thanks

[serene mauve]

how to get user's bio (about me)?

[sage tendon]

not possible

[echo wraith]

You would probably be better of subclassing SlashCommand in that case to add your own attributes. Something like

class MySlashCommand(discord.SlashCommand):
    def __init__(self, *args, **kwargs):
        self.owner_only = kwargs.pop("owner_only", False)
        super().__init__(*args, **kwargs)

def slash_command(**kwargs):
    return discord.application_command(cls=MySlashCommand, **kwargs)


@slash_command(yada yada)
async def cmd(ctx:...):
    ...

[ionic wasp]

👍 cheers

[pulsar ferry]

Thanks for the tip, how do you think this looks?

server_invites = []
server_invites_lock = asyncio.Lock()

async def refresh_invite_cache():
    global server_invites
    async with server_invites_lock:
        guild = bot.get_guild(GUILD_ID)
        invites = await guild.invites()
        server_invites = invites

@bot.event
async def on_invite_create(invite: discord.Invite):
    await refresh_invite_cache()

@bot.event
async def on_invite_delete(invite: discord.Invite):
    await refresh_invite_cache()

@bot.event
async def on_member_join(member):
    global server_invites
    async with server_invites_lock:
        new_invites = await member.guild.invites()
        try:
            old_invites = {invite.code: invite.uses for invite in server_invites}
            for invite in new_invites:
                if invite.code in old_invites and invite.uses > old_invites[invite.code]:
                    update_referral(member.id, invite.inviter.id)
                    break
        finally:
            server_invites = new_invites

[echo wraith]

This looks good ! If you wanted, you could have all that in a Cog so that you could do self.server_invites instead of using global but that's more a preference than anything.

[stray cape]

hey guys what's the default timeout of a View

[pulsar ferry]

I was thinking of doing cogs but it seemed to add a lot of other challenges as i have a few other things happening in this file as well that would cause some circular import references. As it stands the whole file right now is only like 500 lines of code so was trying to just keep it simple

[sage tendon]

5 or 15m I believe I can never remember

[sage tendon]

cogs make everything easier

the only real challenge with cogs is spreading a single command group across multiple cogs

everything else can easily be put either in helper files or synced via the bot object

[pulsar ferry]

Well so for example the issue i was having is that in my main.py right now, i also have the bot spinning up a fastapi webserver. so in a theoretical webserver.py where i define the fast api app and all its endpoints, i couldnt figure out how to call commands that require the bot because in the main.py where i have the bot, it also needs to import the app so it can spin up the server

@tasks.loop(count=1)
async def run_fastapi():
    logger.info("Starting FastAPI server")
    config = uvicorn.Config(app, host="0.0.0.0", port=8080)
    server = uvicorn.Server(config)
    await server.serve()

so in main.py i need to import app (defined in webserver.py) but in webserver.py i would need to import bot (defined in main.py)

so i just kept it all in main.py 😅

[stray cape]

oh

[sage tendon]

you do not ever import bot from your main file

ever

[pulsar ferry]

how do i run it then?

if __name__ == "__main__":
    bot.run(SWEEPY_TOKEN)

[sage tendon]

check the cog guide please

[pulsar ferry]

hmm okay i did give it a try but maybe i misunderstood will try reading again

I was on this page and thought i understood how to essentially define all the cogs as modules, then add them in main.py with something like bot.add_cog(Greetings(bot)) .... but i though you still would have to actually start it after adding all the cogs call bot.run() It doesnt say on this page how to start the bot a different way unless its documented somewhere else?

https://docs.pycord.dev/en/master/ext/commands/cogs.html

[sage tendon]

check the guide

there's a guide specifically for cogs

[pulsar ferry]

oh is this not the right page?

[sage tendon]

that's the docs

[echo wraith]

you do bot.add_cog(...), then bot.run(...)

That should be it

[pulsar ferry]

where can i find the guide?