development

[craggy pine]

Derp, yeah you're not wrong. After work brain is mush right now 😓

[sharp geyser]

@wheat mesa sir, you've used opengl right?

[humble gyro]

omg the goat @craggy pine

[crude totem]

Hello, as the new owner of top.gg I was wondering if you could get someone to update the python sdk that interact with the top.gg api? A user added malware to the library weeks ago and we was told someone would be made aware of it and fix it but this isn’t the case

[humble gyro]

can you show me where?

[crude totem]

Let me find it

@humble gyro

I was also informed someone on pycord did a deep dive and turns out it is indeed malware

[spark flint]

yep malware downloader

fake downloader

how were they able to commit?

[crude totem]

I don’t think no one ever looked into it from top.gg

Still a very good question

[spark flint]

oh this person has perms for another repo

ok

cc @earnest phoenix

since its your github account

want to explain the malware

[earnest phoenix]

what’s up

[earnest phoenix]

what malware?

[spark flint]

your github account commited malware to the top.gg python lib

[earnest phoenix]

huh

hold on

i haven’t even contributed to that repository

[spark flint]

verified commits from your github

maybe compromised?

[earnest phoenix]

i’m only contributing on the c++ one

[spark flint]

https://github.com/top-gg/python-sdk/commits?author=editor-syntax

[earnest phoenix]

bro what

[humble gyro]

if you have push rights (i dont yet 😭) could you revert it?

[spark flint]

see https://medium.com/@demonia/lets-dig-deep-into-pypihosted-malware-part-1-94ada4737442 for more info, explains how it makes a bunch of fake commits and hides it in one

i think your github account was compromised then

[earnest phoenix]

i’m so confused but how am i able to revert it back when i didn’t even do that

i think yes

[humble gyro]

if its your github acc then you are able to

[real rose]

[earnest phoenix]

can u tell me how can i revert it back (lol i havent use git in awhile)

[humble gyro]

git revert (hash)

[real rose]

i reported internally when it came up guess it wasnt sorted

[crude totem]

Indeed wasn’t. Sadly. I’m sure more users are affected by this now

[spark flint]

buffer i think you ran an info stealing malware

[earnest phoenix]

apologies, i think my github account was indeed compromised. Can someone remove my access to every repository

[humble gyro]

i can't yet lol

yea looks like all the commits are garbled nonsense

[earnest phoenix]

holy

[spark flint]

thats part of the malware to hide the commit

it just adds into random places

[humble gyro]

yea

zzz

[real rose]

there is a lovely concise list of people with access to those public repos

[earnest phoenix]

is there a chance that the possible malware is still on my computer rn

[spark flint]

lmfao its hosted on galaxygate

[humble gyro]

atleast no new version was added

[spark flint]

likely

info stealers scrape your creds

problem is, since its from code and not like a file you've downloaded idk what to suggest 😭

[humble gyro]

enjoy rotating every single password you own

[earnest phoenix]

[spark flint]

lemme get someone who knows more to help

[earnest phoenix]

thank you

im opening my pc rn

[spark flint]

oh someone beat me to it

thats a different IP to what i can see tho

[humble gyro]

ig we can just ff to the latest non-buffer commit then

good thing there's no release since

[spark flint]

i dont think it released publicly on pypi

[humble gyro]

pretty sure it is

[spark flint]

so the affected user count will be absolutely minimal

[humble gyro]

we have a pip acc

[spark flint]

oh

ueah but this was only commited to github

andn ot published

[humble gyro]

yea

unless ppl use the repo directly

[spark flint]

the malicious package is down anyways

[crude totem]

So the issue is

[spark flint]

so the malware won't execute

[humble gyro]

great

[crude totem]

I think pip comes with discord.py

So anyone using any other library had to download the one from GitHub

[spark flint]

yeah

[earnest phoenix]

how can i revert the multiple commits

[spark flint]

that was another commit someone wanted to make

[earnest phoenix]

or something

[spark flint]

im happy its not affected any other libs

[humble gyro]

force push to the commit latest non-you commit

[crude totem]

Discord.py requirement should be removed from pip version.

And just let the user install what library they want to use. Cause it does clash with forks of discord.py that uses the import discord part I guess

[humble gyro]

ig open an issue

it's community maintained

[crude totem]

It is on the GitHub version just needs pushed to pip

[humble gyro]

ah

[crude totem]

After the malware is removed

[humble gyro]

yea i can prolly release a new ver whenever i have access

until then no bueno

[crude totem]

And whoever have access to the pip

[lyric mountain]

if the lib converts topgg info into dpy objects, then this isn't possible

since there has to be a minimum AND maximum versions

since dpy changes some things ever so often that break compat

[earnest phoenix]

the requirements.txt was already fixed

[lyric mountain]

also buffer, enable 2fa on ur acc

[earnest phoenix]

will do

[lyric mountain]

enable everywhere possible tbh

[earnest phoenix]

its already enabled via login but let me do it again

pc is cleaned

but they need to remove my access on the repository

weird only the topgg python sdk was attacked

[lyric mountain]

probably the weakest link

this is some 4d chess move, breach into maintainer's account so you can harvest bot tokens that use dpy + topgg

[real rose]

probably because the domain is a lookalike pypi installer

and they dont have a npm module for it or smth

eventho those do exist be warned !!

[wheat mesa]

Yeah what’s up

[earnest phoenix]

weird

i ran a safety check commandline

and it outputted this

https://data.safetycli.com/v/62044/97c/

not sure if this is helpful enough lol

but yeah i think its all resorted

[earnest phoenix]

i see

[earnest phoenix]

oh damn

ill be more alert when installing something now

[lyric mountain]

I kinda now understand why it's so godamn bureaucratic to post packages to maven central

even if someone breaches my machine, and get my key, they still need the pass to said key

it's like 3fa

[lyric mountain]

how did u even get breached?

discord url?

[earnest phoenix]

ion even know

im still finding it out

[modern sable]

the last commit that removed the packages was signed with the same gpg key as the malware commits btw

[lyric mountain]

gotta change it yeah, in case the invader got ur key

[earnest phoenix]

ill revoke and rotate my gpg key

[eternal osprey]

is this TCP protocol with fast retransmits correct?

i don't fully get it yet. Like, suppose we fast retransmit a failed package after 3 duplicate acknowledgements, so we resend that data segment, the server receives it, but what do we do in the client side after receiving acknowledgement 1001?

Like do we just stop as that was all data, or do we need to resend all data from 1001 -> 2500?

[sharp geyser]

Who exactly is in charge of the sdks?

I’d thought at least the cms had control over it

[sharp geyser]

Was just wondering what did you use to learn it

[lyric mountain]

"Pain, suffering and a hint of insanity" - Waffle

[wheat mesa]

learnopengl.com

[wheat mesa]

And a bit of this ^

I’d highly recommend making sure you understand memory and how to manage it before doing opengl though

Because opengl is another layer of memory management, you have to manage RAM but also manage VRAM

[dusky idol]

Life update, it doesn't any you for any legal documents for getting 2nd bot verified

[deft wolf]

Because they already have them, why would they require them a second time

[tulip ledge]

I'm trying to count the amount of comparisons of merge sort and plotting on a graph. This is my merge sort implementation

fn merge_sort(arr: &mut Vec<i32>, comparisons: i32) -> i32 {
  let length = arr.len();
  if length <= 1 {
    return comparisons;
  }
  let mid = length / 2;
  let mut left: Vec<i32> = Vec::new();
  let mut right: Vec<i32> = Vec::new();
  for i in 0..mid {
    left.push(arr[i]);
  }
  for i in mid..length {
    right.push(arr[i]);
  }
  let mut comparisons = merge_sort(&mut left, comparisons);
  comparisons = merge_sort(&mut right, comparisons);
  merge(arr, &left, &right, &mut comparisons)
}

fn merge(arr: &mut Vec<i32>, left: &Vec<i32>, right: &Vec<i32>, comparisons: &mut i32) -> i32 {
  let mut i = 0;
  let mut j = 0;
  let mut k = 0;
  let left_length = left.len();
  let right_length = right.len();
  while i < left_length && j < right_length {
    *comparisons += 1;
    if left[i] < right[j] {
      arr[k] = left[i];
      i += 1;
    } else {
      arr[k] = right[j];
      j += 1;
    }
    k += 1;
  }
  while i < left_length {
    arr[k] = left[i];
    i += 1;
    k += 1;
  }
  while j < right_length {
    arr[k] = right[j];
    j += 1;
    k += 1;
  }
  *comparisons
}

But when plotting it I get a linear graph, so there's something wrong with how I'm calculating the comparisons

[solemn latch]

Does anyone know of a bitvise(ssh/sftp client which supports a dark mode.

I'm tired of being blinded by bitvise

[sharp geyser]

WinSCP is a good sftp client

I use it

You have 3 options with it iirc

[solemn latch]

Thanks ❤️

[sharp geyser]

System, Dark and Light

[solemn latch]

I'll download it the second I get a chance.

[sharp geyser]

np

Only annoying thing about it is the constant updates

but

do what you have to do to remain secure

[eternal osprey]

i basically manipulated an image from my host pc and add it as a MessageAttachment to the embed setImage field by using attachment://file.png.
After a few months it basically removed this file from the setImage, displaying a greyish image.
But other people seem to upload images and it permanently stays. Is this because a normal image does not get deleted while an embed image does? You can for example see the ephemeral-attachment instead of normal attachment.

[spark flint]

hm

it should auto refresh

[lyric mountain]

embed images dont get auto-deleted

ephemeral embed images however...maybe

since the whole point about ephemerals is that they only exist for a while

[ebon inlet]

Hey, I want to make a dashboard for my bot. Is it smart to program an API where the bot and the dashboard can access it? So that I have a large Redis cache and everything then goes together to the database at certain intervals.

[lyric mountain]

it's not "smart", it's the only way you'd be able to make a dashboard lul

as for redis, you dont really need it unless you expect to have a VERY large flow of data

also depends on the size of said data

[ebon inlet]

I would like to add a Redis cache so that I am prepared for the future.

Atm it is not alot of data

[lyric mountain]

adding a cache layer without the need for one can be detrimental to the performance

also you need to handle syncronization between cache and database properly as to not lose nor show stale data

databases are well within confortable zone at 10k reqs/s

you'll inevitably have to rewrite the bot along its growth, dont be too preemptive

[ebon inlet]

Yes thats true.

One side question: How can I get the „Bot Developer“ role?

[lyric mountain]

have your bot aproved on top.gg

[ebon inlet]

Yes

[lyric mountain]

wasn't a question lul

[sharp geyser]

huh

[lyric mountain]

if you have a bot there, then ask some mod for it

since luca might (will) break and not give the role automatically

[ebon inlet]

Sorry I am f…ing dumb😂

[sharp geyser]

talk to @harsh nova (if he remembers to do it)

[eternal osprey]

how can i remove it?

[lyric mountain]

remove what?

[harsh nova]

I am here

[eternal osprey]

well, how can i turn it into a normal attachment

[lyric mountain]

did u put the image in an ephemeral embed?

[eternal osprey]

no

[lyric mountain]

that's weird then

oh wait

is that a thread?

might want to check whether images sent in threads are temporary

[sharp geyser]

do you remember your own name

[harsh nova]

duh it's jh

[eternal osprey]

ahh it might

no nvm

it'sjust a regular channel

i will post the code here soon

maybe something goes wrong in there

[sharp geyser]

no

its dinner

😋

🍽️

[harsh aspen]

hello, how can I code a discord bot

im using js

[eternal osprey]

i just realized that prisma is actually dating back all the way to fucking 1990

[real rose]

prisma is amazing

[eternal osprey]

i've never used it but read a dutch article dating back from 1990 where they noted that it was ran for database queries on a shared nothing dfs with 100 nodes.

[wheat mesa]

look up discord js guide

That’ll give you a “good” start (I don’t agree with everything they say to do)

[eternal osprey]

though each node only had 16 mb of memory

[harsh aspen]

ok

thx wfaffle

[radiant kraken]

prisma is older than JavaScript, Python, and Java

[surreal sage]

nobody:
my phone's termius session that i forgot to close 3 days ago: ☹️

[pale vessel]

"its termius time"

[lament rock]

I love the part of the ssh where he said "its termius time" and started termius-ing everywhere

[pale vessel]

can u send some to me?

[sterile lantern]

lol so random

[wheat mesa]

Dropshippers when they find out they provide nothing to humanity, actually raise prices, and promote child labor: 😱

[sharp geyser]

Mmm my favorite

[solemn latch]

@drifting chasm no ads please

[sharp geyser]

@solemn latch uwu

[solemn latch]

OwO

[sharp geyser]

clipped turned into an emoji

[muted pollen]

[sharp geyser]

what got a problem?

[muted pollen]

bored

[slender wagon]

quick question, i am working with a vue project and by default all the console logs go in the browser's console. Is there a way to move them to the machine's console or editor's console?

[lyric mountain]

only if vue has a "Run" option instead of "Build"

else you're packing everything into production, there's no vue at that point, only html + css + js

https://vuejs.org/guide/quick-start

it does, npm run dev

[slender wagon]

i see

[slender wagon]

so i have been testing on development for awhile but for some reason i cannot get it to log on the machine's console

unlike nextjs

[lyric mountain]

you probably cant then

[slender wagon]

😭

[lyric mountain]

that said, you could try webstorm

jetbrains usually go out of their way to add qol features

[slender wagon]

Are you aware of a way to organize the logs in the console at least?

[slender wagon]

i am using webstorm

[lyric mountain]

use other levels aside from log

[slender wagon]

like can i use a custom logger?

[lyric mountain]

console.error, console.trace, console.debug, etc

[slender wagon]

with colors and allat

[lyric mountain]

doubt it

[slender wagon]

on my strapi backend i used winston

[lyric mountain]

you could, however, output to a file

[slender wagon]

no not an option

I'll just see if i can organize the console logs

maybe create a middleware for th logs

and handling logging properly?

[lyric mountain]

did u try running through "debug" instead of "run"?

also, are u running from ide buttons or terminal?

[slender wagon]

i configured th ide buttons my self

all of them run dev

[lyric mountain]

webstorm doesn't have it by default?

[slender wagon]

i think it does

but i changed them

look how beautiful the backend ones are, i am proud of these

[lyric mountain]

reason?

[slender wagon]

to run them on dev + run them both at the same time with a shortcut

they use npm by default but i have added yarn in the configs. so it does run on yarn as develop

[lyric mountain]

wait did u create a js project or vue project?

[slender wagon]

this is vue

and the backend is strapi

[lyric mountain]

I meant when creating the project in webstorm

[slender wagon]

I am not the one who created it, but i am pretty sure it was created with the create vue app thingy

is that what you want to know?

or am i getting it wrong

[lyric mountain]

I mean this

webstorm add facets to the project when you create through there, perhaps your project isn't being detected as a vue project but as a regular js one

[slender wagon]

could be, but still not everyone runs the project on webstorm

so i should not base it on the editor

yk what i mean

i think i am better off just formating the browser logs and making them clean

since as you said in production it comes as just html js and css

unlike the backend

[lyric mountain]

for now u can just use the different levels of logging js has

console.log is just the generic INFO level

[slender wagon]

yes i know that

but i'll also need some sort of formater

[lyric mountain]

https://www.jetbrains.com/help/webstorm/debugging-javascript-in-chrome.html try this btw

[slender wagon]

so for example when there is an error thrown, i'll get rid of the non important info and only keep some of it

will do

thanks

now it's time for a little break

[lyric mountain]

@wheat mesa can't say how much I hate people who do this

like, ffs don't deploy alpha/beta versions as releases

[wheat mesa]

Real

[charred nest]

ok i need help, because i have been hititng my head against the wall abt this for days. my interactions keep doubling as per my console logs. for background info:

• there is only one instance being run
• i use an event handler -> separate files for events. other events do not double
• the interactionCreate event is only listened to once
• i defer reply for one command, but every command is affected
• this applies to every interaction, including buttons (where i noticed this issue)
  interactionCreate.js:

const pool = require('../bot')
const config = require('../config.json')

module.exports = {
    name: Events.InteractionCreate,
    async execute(interaction) {
        console.log(`interaction type: ${interaction.type}`);
        
        if (interaction.isChatInputCommand()) {
            await handleChatCommand(interaction);
        } else if (interaction.isButton()) {
            await handleButtonInteraction(interaction);
        } else if (interaction.isAutocomplete()) {
            await handleAutocomplete(interaction);
        }
    }
};
async function handleChatCommand(interaction) {
    const command = interaction.client.commands.get(interaction.commandName);
    if (!command) {
        console.error(`No command matching ${interaction.commandName} was found.`);
        return;
    }
    await command.execute(interaction, pool);
    console.log(`${interaction.user.username} used a command | ${interaction.commandName}`);
}
async function handleButtonInteraction(interaction) {
    if (interaction.customId === 'nextPage') {
        //handle buttons. too long and not related to issue
    }
}
async function handleAutocomplete(interaction) {
    const command = interaction.client.commands.get(interaction.commandName);
    await command.autocomplete(interaction).catch(err => {
        console.log(err);
    });

    if (!interaction.deferred && !interaction.replied && interaction.isChatInputCommand()) {
        interaction.reply({
            content: `uh oh, an error occurred.`,
            ephemeral: true
        }).catch(console.error);
    }
} //added to try to mitigate issue. did not make a difference

[charred nest]

jk i fixed it

something in my main.js was duplicating shards

do i know what? no

will i ever know? no

[solemn latch]

I was going to blame

        if (interaction.isChatInputCommand()) {
            await handleChatCommand(interaction);
        } else if (interaction.isButton()) {
            await handleButtonInteraction(interaction);
        } else if (interaction.isAutocomplete()) {
            await handleAutocomplete(interaction);
        }

for not being a switch case ;p

[charred nest]

nerp just something fucking up the shards LOL

[wheat mesa]

I was going to say that you can't convert this to a switch case before remembering that those are just basically instanceof checks 😭

[solemn latch]

Discordjs is so abstracted it makes better code harder here.

Even if the performance benefit here is basically 0

I'd do the same thing, if statements

[charred nest]

tbh i moved them to functions bc i could not keep scrolling up and down checking those if/else

coz im like Did i put a bracket somewhere wrong???

[frank saffron]

Hey, so I was told to head here from the lovely people over on the SRA discord (SomeRandomAPI).

I am developing a bot, which in short will send an "OfTheDay" each day at 10pm. You can select 5 options from a list of selections, for example: "Dog Of The Day", "Question Of The Day" etc etc.

This can all be customised via a /setup command
(See photo below)

So far, I only have three commands:
/setup, /userinfo, /ping.
Whilst these commands do what is required of the bot, I want some more stuff to add and I just cannot think of anything.
Does anyone have any sort of ideas or challenges for me that I could add to the bot?

|| On a side note, this bot is made for a specific purpose.. I don't want it to be one of those really common multi-purpose moderation bots. ||

[wheat mesa]

better than my handling of just slash commands ```ts
if (payload.isCommand()) {
if (payload.isChatInputCommand()) {
const slashCommand = client.commandManager.getSlashCommand(payload);
let shouldRun = true;
try {
if (payload.inGuild()) {
shouldRun =
(await slashCommand.userPermissionCheck(client, payload)) &&
(await slashCommand.botPermissionCheck(client, payload));
}

                shouldRun = shouldRun && (await slashCommand.onBeforeRun(client, payload));

                if (shouldRun) {
                    await slashCommand.run(client, payload);
                } else {
                    await slashCommand.onCancelRun(client, payload);
                }
            } catch (e) {
                if (e instanceof CommandRunError) {
                    await slashCommand.onRunError(e, payload);
                } else {
                    await slashCommand.onUnexpectedError(e, payload);
                }
            }
        }
    }

back when I actually did discord dev

[charred nest]

[solemn latch]

oh man 👀

[wheat mesa]

though I was quite happy with this result

it led to very nicely structured commands

Like so ```ts
export class TestSubCommand extends BaseCommandOption {
constructor() {
super({
name: 'subcommand',
description: 'desc',
permissionsClient: [],
permissionsUser: [],
});
}

async run(client: PaperClient, interaction: ChatInputCommandInteraction) {
    await editOrReply(interaction, 'yay');
}

}

heavily inspired by detritus

[charred nest]

i'll stick with my silly code thanks LOOOL

[solemn latch]

Customizable time of day would be an interesting thing to work on.

[wheat mesa]

had permission checking built in and optionally overridable functions for extra pre-run checks

[charred nest]

ppl will ask me for dev advice and all i can say is.... augh.... im a silly dev pls dont rely on me

[solemn latch]

But all the async calls 👀

[wheat mesa]

What's wrong with that :c

[frank saffron]

I will give it a go, this would actually help out with my testing. Usually I would have to set the time for the cronjob, restart the bot, then wait for it to come online and post the thing to see if it worked. Really delays me lol

Oh, wait. Is it even possible with a cronjob?

[solemn latch]

You would likely want to move away from cron there.

The way i would handle it would be a 10 second interval, checking the database for upcoming requests, then setting a timer to handle each one as it comes up(or just handling them in bulk, they might be off by 10 seconds, but not a big deal)

[wheat mesa]

timer 😰

[solemn latch]

It should be using callbacks, or avoid async code.

async is great, but for base structures imo its better to not.

Puts every async call at the end of the event loop

[wheat mesa]

eh I'm too lazy to care about the ins and outs of the event loop in terms of a discord bot

I don't usually use js/ts much anymore

[solemn latch]

I think understanding the event loop has made me over concerned with it tbh.

[wheat mesa]

Plus if I was interested in high performance code, I'd be using a different language

[solemn latch]

In a bot these days I'd make everything callback based inside the command handler

yeah fair ^-^

[wheat mesa]

Just made it because it was ease of use

[mild swallow]

[fringe nova]

ello
im updating my server count every 30 minutes ish, i used the token from the topgg site itself
but nothings happening 💀

[sharp geyser]

why is intellisense in vscode broken for rust 😭

[wheat mesa]

Are you using rust-analyzer

[sharp geyser]

yes

[wheat mesa]

sometimes it spazzes out

restart vsc

[sharp geyser]

yea no its spazzing out a lot

I'm constantly having to refresh

[neon leaf]

Why is computer craft so complex

Took me like 4 hours to make a working wireless network with access points and signal strength detection

[humble gyro]

hiya

finally got access to the github, removed the commits on python sdk

anyone using those specific commits should get an error and re-install to the latest commit hash

[radiant kraken]

any plans on fixing top-gg/docs?

[humble gyro]

ye

first fixing the site tho

[radiant kraken]

aight

[humble gyro]

😭

top.gg servers are so old that if i want to update nodejs i need to manually build and install glibc

[acoustic bough]

wtf lmfao

what version does it run rn?

[humble gyro]

u dont wanna know

[acoustic bough]

12-16?

idk I thought it would run in docker containers lol

[humble gyro]

they will soon

[acoustic bough]

lol nice

me like updating a all deps every few weeks

[spark flint]

lol

[sharp geyser]

no no we do actually

you said this in the dev channel

big mistake

[true star]

Not enough caffeine in the universe to help me fix these messes‽

[deft wolf]

Huh?

[rugged dawn]

smh

wall spammer

[harsh aspen]

hi

how are

[deft wolf]

Weird question for development chat

[sharp geyser]

Hi how are you NyNu

[harsh aspen]

sorry lost my internet connection while typing this

so, how can I make a modal pop up with html & css
so if I clicked on a button it will pop up

[wheat mesa]

You’ll need js for that as well, granted it’s a small amount that you can just inline it

[harsh aspen]

thank you

but gpt helped me better

[sharp geyser]

💀

the AI + Copy and Paste era is going to kill the tech industry

[harsh aspen]

"enhance the tech industry"

[wheat mesa]

If it works for you then who cares but AI is definitely not going to enhance the tech industry (at least for programming quality) for a very long time

[harsh aspen]

idc

if it works then Im happy

[wheat mesa]

Battleless vibes

[sharp geyser]

are we certain its not?

[wheat mesa]

Wouldn’t be 100% surprised but this account is 3 years old

[sharp geyser]

trying to use diesel_cli on mac is annoying

[limpid rain]

Hi

[lament rock]

to be fair, I did ask chatgpt to help me with my unity game a few times, but its knowledge is quite outdated and it produced quite a few errors, but got me on the right track

[queen needle]

I usually use it for small mundane tasks, like replacing a variable or some small math I can't think of right away

[sharp geyser]

Replacing a variable?

Your IDE can do that for you

💀

[queen needle]

I realized after I sent it I didn't know what exactly I meant to say but it wasn't that

I swear I know about find and replace

[sharp geyser]

[past field]

if anyone has time today, can i get a little help with utilizing sqlite lol

[quartz kindle]

what is your issue?

[harsh aspen]

dontasktoask

.com

[deft wolf]

Funny

[lyric mountain]

this is an excellent example as to why you shouldn't ask to ask

2 hours already and neither the question nor the solution are known

[pale vessel]

it's a bad habit of theirs

[grim aspen]

good example for not specifying the issue

[eternal osprey]

does anyone understand the difference between them?

[lyric mountain]

I dont get it, what ur supposed to do?

implement findElement?

[eternal osprey]

no thats ez, i am trying to understand this theory
but i don't get it at all

like wat'ss the difference betweeen the 2 characteristic creations

[lyric mountain]

without context it's pretty hard to understand what those powerpoint slides are saying

[past field]

I want to connect my bot to a database and hear that sqlite is best for beginners, just not sure where to start

[rose warren]

When I moved all my projects from a VPS to a dedi my RAM usage went from 25GB on the VPS to 15GB on the dedi. Like wtf? I didn't change any code.

[deft wolf]

Maybe better RAM?

[rose warren]

yeah but 10GB?!?!

could it really be that much of an improvement?

[solemn latch]

It stabilized at 10gb? 👀

[rose warren]

was my vpe really that bad?

[solemn latch]

or at 15

[rose warren]

yeah! my bot has been up for a 3 days now and it's stabilised at 15GB total usage for the whole server (my mod bot, my main bot, my db, my microservices...). It used to all use 25 to 28GB out of the 32GB it had on my vps and now it's all using 15GB total

[solemn latch]

👀 thats so weird.

[rose warren]

this was before

[solemn latch]

Do VPS's have some overhead that happens within the VPS itself?

[rose warren]

this is on the dedi

[solemn latch]

This feels like a question for dylan ;p

[rose warren]

i mean, you're sharing the hardware with others, so i guess there's some inefficiencies there?

[pale vessel]

did u compare the processes too?

[rose warren]

but 10GB of efficiencies seems huge

[rose warren]

no i don't have a record of the old ones

but what's insane is i went to town with cache allowances on the dedi since i have the space. I doubled the cache my db is allowed to store and i doubled the max node memory or whatever it is (this is not my field)

and yet it's still using 10GB less

i don't understand

[lyric mountain]

the first move is to understand what is SQL and how to use it

sqlite is easy for beginners because it has very low complexity regarding features, and is really fast

there are a lot of tutorials and courses around regarding how to use sqlite on every language

[rose warren]

Also if you wanna know how fast a dedi is vs a vps @solemn latch

A bump in 2 versions of postgres + the new dedi hardware

[quartz kindle]

if you're using js, you start with installing an sqlite library from npm

the best one around is better-sqlite3

the library will take care of managing sqlite itself, you just give it the sql commands

now, SQL is a separate language as a whole, different from JS. its the language that defines what sqlite should be doing

so its a good idea to learn a little about the SQL language itself

basically what you will be doing in JS, is loading better-sqlite3, then with the library, you send commands to sqlite, for example like this:
database.prepare("SQL command here").run()

with the SQL language, you can create and run all sorts of commands, like create a database, write data, read data, etc...

the good thing about SQL is that its an independent language, meaning it works the same way in JS, in python, in C# and in any other programming language

so its always a good thing to learn

[solemn latch]

I swear we need to make a website called timanswers

Just copy paste the questions and answers

[pale vessel]

timoverflow.com

[quartz kindle]

lmao

[lyric mountain]

timotei.com

except it's already a site

[quartz kindle]

lmao

[solemn latch]

When will I get a section of this dedigrandpa?

^-^

[quartz kindle]

dedi literally means grandpa in slovenian

[sharp geyser]

maybe the cpu cores?

on a vps they are shared and not dedicated to you, so it might be a cause for such high ram usage as the ram has to pickup slack I assume anyway

[rose warren]

TimGPT

[rose warren]

Yeah I know there'd be some overhead but like 10GB worth?

Seems a little crazy

[sharp geyser]

indeed

what provider?

[rose warren]

Pulseheberg

[sharp geyser]

never heard of em

[rose warren]

They're a small French company

Honestly I've been super impressed with their services

[quartz kindle]

did you check if its not the swap configuration?

[rose warren]

But I've outgrown them

[quartz kindle]

maybe the dedi is swapping more

[rose warren]

There's nothing in swap cache

and i did have swap configurated on the vps

[quartz kindle]

the os image is the same?

[woeful zenith]

So I used this API

from discord.ext import tasks

import topgg

# This example uses tasks provided by discord.ext to create a task that posts guild count to Top.gg every 30 minutes.

dbl_token = "Top.gg token"  # set this to your bot's Top.gg token
bot.topggpy = topgg.DBLClient(bot, dbl_token)


@tasks.loop(minutes=30)
async def update_stats():
    """This function runs every 30 minutes to automatically update your server count."""
    try:
        await bot.topggpy.post_guild_count()
        print(f"Posted server count ({bot.topggpy.guild_count})")
    except Exception as e:
        print(f"Failed to post server count\n{e.__class__.__name__}: {e}")


update_stats.start()

It said on my end that it posted the server count, I go to the top.gg page and it hasnt updated?

[solemn latch]

Whats the ID of the bot?

[woeful zenith]

1177973930225569802

I've even tried this snippet of code

import topgg

# This example uses topggpy's autopost feature to post guild count to Top.gg every 30 minutes
# as well as the shard count if applicable.

dbl_token = "Top.gg token"  # set this to your bot's Top.gg token
bot.topggpy = topgg.DBLClient(bot, dbl_token, autopost=True, post_shard_count=True)


@bot.event
async def on_autopost_success():
    print(
        f"Posted server count ({bot.topggpy.guild_count}), shard count ({bot.shard_count})"
    )

[woeful zenith]

And a direct request to their API

import aiohttp
import asyncio
async def updateStats():
    url = f"https://top.gg/api/bots/1177973930225569802/stats"
    headers = {
        'Authorization': ""
    }
    payload = {
        'server_count': 49956
    }
    async with aiohttp.ClientSession() as session:
        async with session.post(url, data=payload, headers=headers) as resp:
            if resp.status != 200:
                print(f'Failed to update server count: {resp.status}')
            else:
                print('Server count updated successfully.')
            
asyncio.run(updateStats())

None worked

[woeful zenith]

yeah

[solemn latch]

is 49k not right?

[woeful zenith]

it is right

but i need it to update automatically

[lyric mountain]

press ctrl + f5

on the page

[woeful zenith]

right now we need to go into edit, press save

[solemn latch]

Thats likely just site cache

It will update over time

[woeful zenith]

didnt do anything

[woeful zenith]

how long approx does it take? because we tried waiting ebfore and we waited weeks and nothing happened

[deft wolf]

I'm totally lost in this situation guys

[lyric mountain]

if ctrl f5 didnt work, then it's likely yet-another-topgg-cache-issue

probably the expiration aint working

[solemn latch]

So, in your logs every 30 minutes for weeks it logged

"Posted server count ({bot.topggpy.guild_count})"

in your console?

👀

[woeful zenith]

Ye

It does that

But dosent update on the site

[solemn latch]

have you edited in the past hour?

[woeful zenith]

Yes

I just edited right now

[woeful zenith]

Direct API call

I used that

It didnt update

[solemn latch]

It wont update right away.

For now, don't edit and leave it auto posting every 30 minutes.

In 3-4 hours check https://dblstatistics.com/bot/1177973930225569802.

dblstatistics checks the backend, and should tell us if the backend is updating or not.

[woeful zenith]

Okay

ill leave it auto updating

[solemn latch]

If you use the direct api code youll have to run it every 30 minutes with a timer, or just manually run it.

[woeful zenith]

If I leave it like this, its good?

dbl_token = ""
    bot.topggpy = topgg.DBLClient(bot, dbl_token, autopost=True, post_shard_count=True)

[woeful zenith]

Or do I go back to this

[solemn latch]

You need the

@tasks.loop(minutes=30)
async def update_stats():
    """This function runs every 30 minutes to automatically update your server count."""
    try:
        await bot.topggpy.post_guild_count()

code too iirc

I dont know python, its over my head on how it works.

[woeful zenith]

Alright

[rose warren]

I think it's a newer version

[quartz kindle]

really weird then

the only things i can think of that would explain a 10gb difference is either ram compression, swapping or the workload not being the same

[eternal osprey]

 let embed = EmbedBuilder.from(interaction.message.embeds[0]);
          embed.setDescription(
            `> Profit report posted by ${interaction.member}.`
          );
          embed.setTitle("🚀 | Profit Submission");

          let k = await outputChannel.send({
            embeds: [embed],
          });```
why is my discord bot sending ephemeral embed images such that they eventually dissappear after a few months?

 const attachment = new AttachmentBuilder(await canvas.encode("png"), {
      name: "report.png",
    });
    const actionRow = new ActionRowBuilder().addComponents(
      new ButtonBuilder()
        .setLabel("Accept")
        .setCustomId("complete")
        .setStyle(ButtonStyle.Success)
        .setEmoji("✅"),
      new ButtonBuilder()
        .setLabel("Decline")
        .setCustomId("discard")
        .setStyle(ButtonStyle.Danger)
        .setEmoji("❌")
    );
    const botInfoEmbed = new EmbedBuilder()
      .setColor("DarkButNotBlack")
      .setFooter({ text: `🚀 Report Bot | Profit System` })
      .setTitle("🚀 | Are you sure?")
      .setDescription(
        "# Image Metadata" +
          "\n\n> ``Name:`` " +
          `${name}` +
          "\n> ``Quantity:`` " +
          `${quantity}` +
          "\n> ``profit:`` " +
          `$${profit}\n\nPlease click on one of the buttons below to send this image to the ${outputChannel} channel or to discard it.`
      )
      .setImage("attachment://report.png")

      .setTimestamp();

    let z = await interaction.reply({
      ephemeral: true,
      fetchReply: true,
      embeds: [botInfoEmbed],
      components: [actionRow],
      files: [attachment],
    });```

[sterile lantern]

months or minutes?

if youre asking why its sending ephemeral messages, it's because youre sending it as an ephemeral message

[acoustic bough]

discord only stores files for ephemeral replies as long as the reply exists

(a bit longer than that but no guarantees)

[eternal osprey]

i got it working but what's this type of bullshit?

embeds[0].image.url returns me this a link, that once you click on it downloads a random ass file

[lament rock]

A:\Windows\Documents\GitHub\Amanda\packages\runtime-worker>yarn start
A:\Windows\Documents\GitHub\Amanda\node_modules\backtracker\dist\index.js:32
    try { throw new Error(); } catch (e) { stack = e ? e.stack : ""; }
                                                         ^

[getStack (A:\Windows\Documents\GitHub\Amanda\node_modules\backtracker\dist\index.js:32:58),getPrefix (A:\Windows\Documents\GitHub\Amanda\packages\logger\dist\index.js:1:227),post (A:\Windows\Documents\GitHub\Amanda\packages\logger\dist\index.js:1:609),getStack (A:\Windows\Documents\GitHub\Amanda\node_modules\backtracker\dist\index.js:38:13),getPrefix (A:\Windows\Documents\GitHub\Amanda\packages\logger\dist\index.js:1:227),post (A:\Windows\Documents\GitHub\Amanda\packages\logger\dist\index.js:1:609),getStack (A:\Windows\Documents\GitHub\Amanda\node_modules\backtracker\dist\index.js:38:13),getPrefix (A:\Windows\Documents\GitHub\Amanda\packages\logger\dist\index.js:1:227),post (A:\Windows\Documents\GitHub\Amanda\packages\logger\dist\index.js:1:609),getStack (A:\Windows\Documents\GitHub\Amanda\node_modules\backtracker\dist\index.js:38:13)]

Node.js v20.8.0

The fuck is this supposed to mean lmao???

even just new Error().stack does that so dont get me for syntax

https://github.com/PapiOphidian/BackTracker/blob/master/src/index.ts source code

Error is from line 55 (there are some changes I made while trying to rapid iterate fix, but nothing made any effect on the error)

oddly, works in the repl

Oh wait. It was console.logging which I overwrite so circular reference. I'm dumb

[tired jay]

hullo... i am trying to post the server installs for my bot using the following code:

def update_installs(install_count: int):
    response = requests.post(
        url=f"{BASE_API}/bots/{BOT_ID}/stats/",
        headers=_headers(),
        data=json.dumps(
            {
                'server_count': install_count,
            },
        ),
        
    )
    response.raise_for_status()
    return response.json()

def _headers():
    return {
        'Authorization': BOT_TOKEN,
    }

the POST call doesn't return an error, but also doesn't seem to have updated my servers count:

>>> from services import top_dot_gg
>>> response = top_dot_gg.update_installs(50)
>>> response
{'shards': []}
>>> top_dot_gg.get_installs()
{'shards': []}

(consistent with https://dblstatistics.com/bot/1186494834446454784)

could anyone point me towards what i'm doing wrong?

[solemn latch]

You're missing the content-type header

application/json

[tired jay]

ty... i just added that and tried again with the same results:

def _headers():
    if not settings.TOP_GG_BOT_TOKEN:
        raise Exception("missing TOP_GG_BOT_TOKEN")
    return {
        'Authorization': settings.TOP_GG_BOT_TOKEN,
        "Content-Type": "application/json",
    }

>>> from services import top_dot_gg
>>> response = top_dot_gg.update_installs(50)
>>> response
{'shards': []}
>>> top_dot_gg.get_installs()
{'shards': []}

[tired jay]

also your cat is indeed cute.

[solemn latch]

its been awhile since I've posted using the raw api, you might need to provide all three values.

shardId and shardCount, and serverCount.

Its been awhile, so I could be wrong

Nope, I was wrong

[tired jay]

yah... same results with:

def update_installs(install_count: int):
    response = requests.post(
        url=f"{BASE_API}/bots/{BOT_ID}/stats/",
        headers=_headers(),
        data=json.dumps(
            {
                'server_count': install_count,
                'shard_count': 1,
                'shard_id': 0,
                'shards': [install_count],
            },
        ),
        
    )
    response.raise_for_status()
    return response.json()

[solemn latch]

I'll play around with the API and see if I can figure out what we're missing ^-^

[tired jay]

tyvm!

[solemn latch]

hmm

[solemn latch]

Theres no way its the trailing slash is it?

url=f"{BASE_API}/bots/{BOT_ID}/stats/",
to
url=f"{BASE_API}/bots/{BOT_ID}/stats",

[tired jay]

huh. i think that was it. ty!

Python 3.11.7 (main, Jan 17 2024, 06:37:55) [GCC 12.2.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> from services import top_dot_gg
>>> top_dot_gg.get_installs()
{'shards': []}
>>> top_dot_gg.update_installs(50)
{}
>>> top_dot_gg.get_installs()
{'server_count': 50, 'shards': None, 'shard_count': None}
>>> 

[solemn latch]

goofy api

[sharp geyser]

hey woo

[sharp geyser]

hows ya

[solemn latch]

hungry

[sharp geyser]

same lol

I want to do some open source projects

Any ideas?

[solemn latch]

Hows your nextjs skills?

[sharp geyser]

anything frontend

is terrible

[solemn latch]

Well, this is in between ^-^

lol

[sharp geyser]

[solemn latch]

I need a lib that takes

https://nextjs.org/docs/app/building-your-application/optimizing/fonts#google-fonts

And makes it so I can dynamically load any font in here.

Thanks

Just make it open source, ezpz

kek

[sharp geyser]

you got 5$?

[solemn latch]

$3

I'm bored and making a description builder.

https://i.woo.pics/de12a2dca8.mp4

And I want to let users use whatever font they want

But it sucks 😦

[sharp geyser]

uh

I mean

you can just import everything from that lib

and just search it like an object

I think

so import Fonts from 'next/font/google' then Fonts["Inter"]({})

[solemn latch]

If I restrict them to variable fonts only that might work

Better than nothing

[sharp geyser]

import * as Fonts from 'next/font/google' if it doesn't have a default

I don't see making a lib for something like this being super helpful tbh

no offense

[solemn latch]

Ah, I remember why this wont work.

[sharp geyser]

I might work on my i18n project again

[solemn latch]

Fonts need to be defined as a const, its a restriction on the next/font/google thing

[sharp geyser]

Better yet, I will make an entire discord js framework

[solemn latch]

ask tim for help ^-^

[sharp geyser]

oh god

tim would demolish me and essentially rewrite the entire project 😭

I just want something to do on the side so I don't get bored of just one thing

[quartz kindle]

wat

[pale vessel]

do something outside of discord

[quartz kindle]

nein

[green kestrel]

touched grass as instructed.
instructions unclear, got muddy hands.
C-.

[earnest phoenix]

[fringe nova]

ello
my server count does not get added on the top.gg website ;-;
this is the code i use, i based it on what the api said:

    const body = [
      { server_count: client.guilds.cache.size },
    ];
    await fetch(`https://top.gg/api/bots/939565109787902052/stats`, {
      body,
      headers: { Authorization: "token removed due to security reasons" },
      method: 'POST',
    }).catch((err) => console.error(`#${cache.index + 1} ${this.name}: ${err.stack}`));

im pretty sure its not an issue with your api, just im probably doing something wrong ;-;

[acoustic bough]

I dont think the body should be an array

also, you have to stringify the body before sending it

[fringe nova]

oh

oh okay my bad i was reading docs wrongly

if its arrayed it acts like shards

and my bot is not sharded

thank you :D

[acoustic bough]

yw

[fringe nova]

once i post it, would i need to refresh data? or would it load automaticly

[solemn latch]

It updates over time on its own, but you can encourage it to happen faster by editing and saving.

[rancid parcel]

What Is The Best Hosting?

[solemn latch]

best in terms of what?
Just raw performance ignoring price?

[sharp geyser]

Has anyone here integrated stripe into their platform? If so what did you use? I was reading the documentation on it and they quite frankly have a lot of different options. Payment links, Checkout, Connect, etc. I know I will likely need to use a combination of Checkout and Connect if I wanted to make my own payment flow with them, meaning I would be making my own customer accounts when it comes to it.

[rose warren]

I use it through Chargebee. Super simple to set up and Chargebee manages all my subs for free.

[sharp geyser]

oh?

[rose warren]

You can even add other payment processors like PayPal, Amazon Pay etc. It's an excellent product.

[sharp geyser]

hm

[rose warren]

All I deal with is webhooks sent to my bot from Chargebee to update it on the status of people's subs.

Everything else is handled by Chargebee

[wheat mesa]

The real struggle is keeping track of taxes on such platforms 😉

[rose warren]

Stripe have a taxes add-on that calculates it for you

[wheat mesa]

Oh good

[rose warren]

Chargebee is free up to like €100k in lifetime revenue or something. But honestly I'm nowhere near that but I'd happily pay for it already.

[sharp geyser]

I mean, stripe I can use, I just don't know what flow to go with.

My usage will be fulfing orders and shipping to the customer.

[rose warren]

Oh wait it's actually $250k they increased their free tier

[rose warren]

Oh not subs?

[sharp geyser]

no

[rose warren]

Chargebee doesn't make sense for you then

[sharp geyser]

yea which is why I was about to say chargebee won't work

I was just looking at their website

[rose warren]

I haven't worked with the Stripe API so I'm not much help

[sharp geyser]

all good

the stripe discord kinda blows ngl

See idk

I think I understand what I need to do but please if anyone's ever used stripe's api reach out to me 😭

[spark flint]

Stripe API is amazing

I use it for several projexts

[sharp geyser]

please

teach me the way

😭

[spark flint]

I’m at work atm but will help after

[sharp geyser]

ok

[spark flint]

What I do is check if a user has a stripe ID linked to them in the DB

[sharp geyser]

I am free for the next 3-4 days

so

[spark flint]

if they don’t, generate one

Then generate a checkout session with their stripe ID and assign an order ID in metadata

[sharp geyser]

Well here's the thing. I plan on making comissioner accounts with my platform, wouldn't I need to integrate stripe connect into my workflow as well?

[spark flint]

Then use stripe webhooks, and watch for incoming payment events matching that stripe ID and match the custom meta data

Ah true

[sharp geyser]

How would that play out with the whole payment stuff

[spark flint]

Pretty sure stripe checkout works with that

[sharp geyser]

[spark flint]

IIRC you pass a stripe connect ID too and it will link to their stuff

[sharp geyser]

I think I will first need to work on integrating stripe connect which is something I didn't plan on doing until later cause comissioner accounts wasn't something I was worried about yet

it seems like the most viable thing tho to get it out of the way

Then I can work on payment processing and figuring that out.

[sharp geyser]

oh cool

I am using rust for this, so the package I am using is kinda wack.

[spark flint]

Stripe makes everything easy asf when you figure it out

[sharp geyser]

It's docs page is a mess

[sharp geyser]

Honestly I wasn't too confused on how to use stripe, just what to use

[spark flint]

Fsirrs

Fairs

[sharp geyser]

My workflow is like this

Customer creates an account on my platform -> I think I use stripe here? -> They browse the products -> They comission a design to go on those products (or choose to just take the design as an image file or smth) -> Checkout -> Pay -> Order ships or is emailed to them if they choose the second option

Now the thing is

I have no fucking clue on how to define products. I know stripe has a products page, but that seems very limiting

[spark flint]

I wouldn’t use that

[sharp geyser]

I'd rather have a json file of the products we offer

and have defined values there

[spark flint]

It’s quite slow and buggy and some adblocksrd stop it

[sharp geyser]

Man making a service is a lot harder than I thought when it comes to planning this shit out

I might need to rework my auth flow to work with stripe connect

[sharp geyser]

I will bother you about this later if you don't mind, but to clarify, have you ever used stripe connect?

[wheat mesa]

json? :c

[sharp geyser]

I was saying i'd prefer that over the stripe products page

[frank saffron]

Hey. I know this may be simple js but whatever..
If I want to map every role that a user has (apart from @ everyone), would i do:

<member>.roles.cache.map(role => `<@&${role.id}>`).filter(role => role.id !== interaction.guild.id);

Or does the filter have to come first

[real rose]

filter first

because you want to filter out the everyone role

then map over the result

[frank saffron]

Okay cool, thank you!

[wheat mesa]

Yeah, filtering here would be filtering over a list of strings that doesn't have an id property

So in practice your filter predicate would always return true

[sharp geyser]

@spark flint

excuse me sir may I bug you?

[stiff dust]

@green kestrel well im here

[spark flint]

sure

[sharp geyser]

trying to figure out how to make a table relation rn.

Essentially there are two tables. students and campuses. A campus can have multiple students, but a student can only belong to one campus. How would this be modeled in sql?

raw sql table relations are new to me, I usually had an orm like typeorm handle this for me

I was thinking that the foreign key on the students table that references the campus e.g campuses_id had to be unique but if I do that, then that means that campus_id cannot appear anywhere in the table again

[spark flint]

hm

[sharp geyser]

wait

I just need to make the student id unique

ez

[green kestrel]

hmm a ping was deleted

[pale vessel]

wasn't it this

[rancid parcel]

a good and trusted hosting

[pale vessel]

hetzner?

[deft wolf]

Personally, at the moment I am using datalix (https://datalix.eu/) to host the bot and databases and contabo (https://contabo.com/en/) to host photos/videos because they had a promotion in which they did not charge the setup fee at all

Both work without major problems and contact with support is very pleasant and easy

[spark flint]

i use datalix and hetzner

and a few others

im liking https://inceptionhosting.com/vps.html

[charred nest]

i use digital ocean for both my vps and cloud db

[stiff dust]

I didnt delete it

[deft wolf]

[green kestrel]

yeah it was

[green kestrel]

so yeah do you have your bot source on github

i could take a quick look and stab a guess at what discord didnt like

no promises 🙂

[slender wagon]

question

is it ok if i store images and files uploaded by a user on the file directory on my vps

Not for mass usage only used by admins

i only store the directory on the postgres so i can retrieve it easily later

[charred nest]

storing would need to be stated in the privacy policy, usage of uploaded data im not sure. i also recommend having some sort of filter on user uploads

[slender wagon]

aka where the file is located

[deft wolf]

If it's included in your privacy policy, there shouldn't be a problem

[slender wagon]

oh that's not the issue. I am just wondering about my practices

Like what else could be there besides storing in the file system

[charred nest]

as long as its stated in the privacy policy, it should be fine imo

[slender wagon]

oop practices as in speed etc etc

[charred nest]

oh my bad HAHAHA

[slender wagon]

I am storing files uploaded by the owner of the site so a policy isn't important

it's a webite

[charred nest]

i don't think it would be slow ? not sure though since im not the most knwoledgable on that

[slender wagon]

oh that's fine, yeah it aint that slow. I'd have to wait for production though and i really need some strict rules as someone could fuck my entire vps with a few uploads

[charred nest]

i store my images/files on cloudflare and havent had any speed issues in retrieving tbh

[slender wagon]

you're going to wonder if the endpoints are secure

nope

[slender wagon]

never tried

we could be talking about 100mb + files

so

[charred nest]

i think for cf its $5 for 100k images stored

$5 a month*

[slender wagon]

what about files?

like size

[charred nest]

https://www.cloudflare.com/developer-platform/r2/

[slender wagon]

ty

[charred nest]

nppp

[acoustic bough]

I use r2 and love it

[slender wagon]

i don't get the 10gb/month

like i can store 10gb worth of data this month and another 10 next month lmfao?

or 10gb all time storage

[acoustic bough]

nah

10gb is always free

[slender wagon]

so i can have up to 10gb stored on cloudflare

[acoustic bough]

with aws you only have 10gb free for the first year

[acoustic bough]

yes

[slender wagon]

i see

[acoustic bough]

that’s what i’m doing

optimize all uploads to be 200kb

and just live off the free tier

[slender wagon]

Can't i got a bunch of large files

so i think i will continue with the vps storage

but i will def use it on some other project

[acoustic bough]

well I only store webp’s

100-200kb on average

[slender wagon]

i see

that's cool

[surreal sage]

@solemn latch consider contabo for vps'

[pale vessel]

galaxygate is better than contabo

contabo has those specs for cheap because they oversell

that tends to cause cpu steal

[slender wagon]

Galaxygate sucksass if you have an albanian ip

[solemn latch]

👀

[slender wagon]

Its weird it just timesout logs you out and makes you struggle

So i quit

[solemn latch]

Have you poked dragon about it?

[slender wagon]

No

This was a few months ago

But i might give it another shot as they are cheap

[charred nest]

dumbassery here

[solemn latch]

You could check for any shard that has the user cached, if not fetch the user.

[charred nest]

its client that's undefined tho no?

[solemn latch]

Client would be undefined in your sharding file right?

Since it has no bot client

[charred nest]

ah yeah, true

[solemn latch]

It's been a long time since I've thought about sharding I don't remember much about it tbh ^-^

[charred nest]

no worries ! any ideas on what would be best practice then to get the user? im trying to fetch the user that voted -> send a dm but obv only the currency update portion works

[solemn latch]

If that's all you really want to do, use client.users.send(userID, standard message stuff)

You'll want to eval that into your first shard

[charred nest]

ahhh gotcha !

[solemn latch]

I forget how, to eval to a specific shard, but you'll figure it out ^-^

[charred nest]

i haven't touched vote stuff in a hot min so im like

this helps a lot, thank u !!

[surreal sage]

nah

63% battery gone in an hour

discord doesn't optimize shit 💀

[wheat mesa]

the react native experience

[surreal sage]

being in a lot of servers

and all new messages get sent to you

doesn't matter if you haven't looked in that guild for a month

used like 230mb in a hour

celular

[harsh aspen]

how a 55min of discord takes 63% of the battery

that doesn't make sense

[surreal sage]

it's not a shitty battery fyi

like 1-2yo phone

2022? release

[pale vessel]

double the screen time, half the battery, and on an even less stable version of discord

[harsh aspen]

that does make sense

[pale vessel]

yeah i got a good phone

[harsh aspen]

what is it

I assume that you both have the same OS

[pale vessel]

it's the red star OS

my government made it mandatory to use it

[harsh aspen]

oh bruh

||they are tracking u, made it spoiler so they can't read it, Idk if I'm dumb like that||

[wheat mesa]

you've gotta be trolling right

[deft wolf]

My first guess would be Russia

[stiff dust]

its private repo but I can give u access to it

[sharp geyser]

[green kestrel]

which government

you're not in dprk unless your surname is Kim and you're part of the ruling elite

nobody else has internet enabled phones iirc that can get outside services

[pale vessel]

i'm an exception

[lament rock]

Flaze is next in line to be the world ruler

[pale vessel]

the world suler

[lament rock]

Lmao

[pale vessel]

[frosty gale]

😭

i swear they said react native would save battery better than their native app

but thats what happens when you insist on using javascript and react everywhere on platforms not designed for it

you have more translation layers than good code

actually not surprised they dont care bc they can get rid of their java native developers and only keep their web dev react team that can also develop the app

theyre saving money and were paying in terms of battery usage and performance

[sharp geyser]

@lyric mountain hey buddy are you busy?

I heard you know a lot about sql databases

[stiff dust]

@green kestrel

[twilit beacon]

Hi, I'm using the api: https://top.gg/api/bots/1082860291152420954
to get information from my BOT and I needed the "server_count" parameter, could anyone help me find out why it is not being returned?

[deft wolf]

Because the bot does not have this value

You never sent the number of servers through the top.gg api so this information does not exist on the top.gg side

[surreal sage]

https://tauri.app

native webview x rust

[surreal sage]

if only javascript could be compiled natively 💔

[wheat mesa]

That would defeat the purpose

[lament rock]

You could theoretically transpile it into assembly

Though would likely need to be TypeScript based

and you'd lose the benefits of being runtime interpreted

err I guess asm could be interpreted

[radiant kraken]

skill issue

[lyric mountain]

Not a lot no, but depends

[sharp geyser]

I just had some questions regarding my table setups and was wondering if you could give me some insights

[lyric mountain]

Sure

[past field]

lord

[nova zephyr]

how can i change the color of the main theme of my bot page?
whenever i try it only seems to change the bg color of the content in the long description

[solemn latch]

What css are you using?

[nova zephyr]

<style>
body {
background-color: black !important;
font-family: Manrope;
color: #f3f3f3;
}

    img {
        max-width: 100%;

    }

    blockquote {
        border-left: 4px solid #3498db;
        padding-left: 10px;
        margin-left: 0;
    }

    .image-container {
        display: flex;
        justify-content: space-between;
        margin-bottom: 20px;
    }
</style>

this to test

[solemn latch]

Looking at your bot page, you're including things that are unneeded.

The website itself already has html, head, body tags.

<!DOCTYPE html>
<html>
  <head>
  </head>
  <body>
  </body>
</html>

^ I think by including these things you're creating your own "scope" within the description, which means your css cant affect the right things outside of the description.

[nova zephyr]

yeah that worked thanks

[sharp geyser]

Does anywhere here know about docker and iptables? I was looking at setting up mailcow again and they have a new warning on their page

Is this important to do or is it fine to still go with ufw ?

[grizzled raven]

Is pairing ufw and docker known to cause common issues? Because I'm new to docker and I'm running into an issue that causes apk packages to not install at all unless I disable ufw

perhaps it's a dns issue or a skill issue but idk thought i'd ask

[sharp geyser]

Well

Apparently docker has issues with the ufw or firewalld from what i've been reading, but I don't have any clue on how iptables work and its a more intimate aspect to controlling your firewall where ufw abstracts you away from all that mess.

[harsh aspen]

I recommend using windows

[sharp geyser]

[harsh aspen]

[tulip ledge]

Anyone who might have a clue why this merge sort implementation is doing way more comparisons than expected?

fn merge_sort<T: Ord + Copy>(arr: &mut [T], aux: &mut Vec<T>, low: usize, high: usize, comparisons: &mut usize) -> usize {
  // Base case: if low is greater than or equal to high, the array is sorted
  if low < high {
    let mid = (low + high) / 2;
    // Recursively sort the left and right halves
    merge_sort(arr, aux, low, mid, comparisons);
    merge_sort(arr, aux, mid + 1, high, comparisons);
    merge(arr, aux, low, mid, high, comparisons);
  }
  *comparisons
}

fn merge<T: Ord + Copy>(arr: &mut [T], aux: &mut Vec<T>, low: usize, mid: usize, high: usize, comparisons: &mut usize)  {
  // Copy the array into the auxiliary array
  for i in low..=high {
    aux[i] = arr[i];
  }

  let mut i = low;
  let mut j = mid + 1;

  // Compare the elements of the two halves and merge them into the original array
  // The smaller element is always copied into the original array first
  for k in low..=high {
    if i > mid {
      arr[k] = aux[j];
      j += 1;
    } else if j > high {
      arr[k] = aux[i];
      i += 1;
    } else if aux[j] < aux[i] {
      arr[k] = aux[j];
      j += 1;
      *comparisons += 1;
    } else {
      arr[k] = aux[i];
      i += 1;
      *comparisons += 1;
    }
  }
}

It's supposed to do ~nlogn comparisons in worst case but measuring this I find that it does ~1.6nlogn comparisons

[wheat mesa]

O(1.6nlogn) is nlogn

Big O notation does not take into account coefficients

It cares about the way things grow

[harsh aspen]

Sorry I cant help

[spark flint]

...then don't reply?

[sharp geyser]

question for you

how would you handle authorization of endpoints?

[spark flint]

using middleware

[sharp geyser]

okay but I mean

what concept, like jwts, or some other kind of auth model

[spark flint]

jwt works

i usually just use uuid

[sharp geyser]

huh

uuid?

🤔

[spark flint]

https://uuid.rocks/plain example UUID

every language has a package to generate a UUID

[sharp geyser]

I know what a uuid is

[spark flint]

yes

good

[sharp geyser]

I am wondering what you mean when you say you use uuids

that doesn't quite track with me

[spark flint]

its just a random unique token

pass as an authorisation header

[sharp geyser]

okay

[spark flint]

match to a user

[sharp geyser]

That could work yes.

I orignally had that idea

but I saw no way of managing this with https://rocket.rs

[solemn latch]

Isnt that just jwt with extra steps?

[sharp geyser]

no

jwt has embeded data and is meant for short lived authorization

they can expire.

[solemn latch]

You want them to expire no?

[sharp geyser]

Nope

[solemn latch]

oh api

mb

[sharp geyser]

The only thing that should expire is the login session.

I think ima just make a token for each user when they signup for the service and use that to authenticate routes.

I just don't know how to do this with rocket

😭

I wish I could just implement an auth system like discord and just call it a day

[lament rock]

Yeah I'd just generate a token like Discord does

can embed the user ID in the token

[silk flame]

Can I run two node.js client instances? One for a slash command and one for a prefix?

[pale vessel]

ideally you should be combining them to one instance

[vale moss]

Preferably not. That's just asking for trouble. It's also is more work to run two files (you need two terminals instead of 1 and two commands)

[silk flame]

How would I combine them then?

[lyric mountain]

Just listen to both event types

[craggy pine]

Those being interactionCreate and messageCreate

[craggy pine]

https://scs.twilightgamez.net/eEFpM.png

[tulip ledge]

Yeah we use tilde notation

Then we do care about factor

[lament rock]

Need opinions on how to get stuff ordered in sql or if I'm already doing the best thing.
Currently, I have tracks in a playlist and the structure is a little something like: { playlist_id: number; video_id: VARCHAR(50); next: VARCHAR(50) | null; }
The next is obviously linking to a video ID that's next in the track list if any and to get all the tracks in order, I need to go backwards starting from where next is null and then searching the entries where the next field is the current video_id until there is none left. Cannot be recursive. I have to be able to reorder the tracks at any time and updating all entries after an index wouldn’t be a good idea if I went with another number field to indicate index. Any ideas on improvements? (The searching is done in code rather than on the sql server's side)

[sharp geyser]

Im not sure if I was misreading your question, or if I am not quite understanding what you are trying to achieve.

You said you need to figure out how to order stuff in your sql?

Well, the simplest solution is to follow spotify's example.

The way they have playlists is you add songs to them, and they immediately get pushed to the back. You can use a numerical index to track this as if its an array starting at 0 (or 1) if you prefer. Now the next would be an id to the song directly after it, if there is one. Simple enough to achieve this I trust you to be able to do this. You'd need a numerical index to be able to make sure that no matter what if you were to visually load this playlist, it'd remain in the same order. You can ofc allow them to reorder it if you want, which is when you figure out the logic yourself and set it's new "place" value

Again, not sure if I misunderstood what you are trying to achieve.

[glacial acorn]

Please make for me a profile with ROYAL GHOST 111

[solemn latch]

👀 what does this mean?

[deft wolf]

At least he wrote "please"

[charred nest]

any ideas as to why a message the bot sends wouldn't be found in the cache? im trying to edit my bot's starboard message when an additional reaction is added but it returns unknown message

• it is the correct msg id
• verified that the bot has the correct channel permissions
• the bot sends the msg thats trying to be edited
• it isnt restarted in between
• logging the message shows that it isnt in the cache ?? idk why

                const channelID = '964997630910427197';
                const channel = client.channels.cache.get(channelID);
                        client.starboard.set(message.id, {
                            originalMessage: message,
                            count: count,
                            author: message.author.username,
                            authorURL: message.author.avatarURL()
                        });

// .... other code here

try {
                    const fetchedMessage = await channel.messages.fetch(msgID);
                    console.log(`Fetched message: ${fetchedMessage.content}`);
                } catch (error) {
                    if (error.code === 10008) {
                        console.error(`Message with ID ${msgID} not found.`);
                    } else {
                        console.error(`Error fetching message with ID ${msgID}:`, error);
                    }
                }

sorry if the code is a bit all over the place, ive tried dozens of things atp

[pale vessel]

fetch also grabs it if it isn't in the cache, so are u sure u got the correct message ID and channel?

[charred nest]

yep, i already double checked

channel and msg id are correct and match to what's being logged, i double checked permissions and those are fine as well

[pale vessel]

what's msgID?

[charred nest]

oop, shit, sorry i thought i included it

                const msgID = starboardData.messageID

i checked what it logs and it does log a string

[pale vessel]

by right clicking the message on discord and copying the id too?

[charred nest]

yeah

[pale vessel]

[charred nest]

yeah LMFAO me too

i legit dont know wtf is happening

[pale vessel]

u have an error check there, but what's the actual error message if it's 10008?

is it really just "Message not found"?

[charred nest]

unknown message

yeah

[pale vessel]

ah

does the bot have read message history perm? that might give a different error though

[charred nest]

yeah it does

i thought i was going crazy so i asked the owner of the server to give me perms to double check the channel perms and theyre fine

i used a msg content cmd too and it works so i know its reading msgs there

[pale vessel]

a full code might help

[charred nest]

one sec

client.on(Events.MessageReactionAdd, async (reaction, user) => {
    console.log(`reaction added`)
    const reactionThreshold = 1;

    if (reaction.emoji.id == '1041127014066835590') {
        const message = reaction.message;
        const count = reaction.count;

        const channelID = '964997630910427197';
        const channel = client.channels.cache.get(channelID);

        let starboardData = client.starboard.get(message.id);

        if (count >= reactionThreshold) {
            if (starboardData) {
                console.log(starboardData.originalMessage.id)
                const footerText = `♡ ${starboardData.count}`;
                const embed = new EmbedBuilder()
                    .setColor('#f0deda')
                    .setAuthor({
                        name: `${starboardData.author}`,
                        iconURL: starboardData.authorURL,
                    })
                    .setDescription(`· ·  ![f2_hrtbow_DNS](https://cdn.discordapp.com/emojis/1146941520239394816.webp?size=128 "f2_hrtbow_DNS")   new heart  !!  + ⁺ .\n\nʚ・"${starboardData.originalMessage.content}"・ɞ\n\n![f2_hrtbeat_DNS](https://cdn.discordapp.com/emojis/1146941657095340052.webp?size=128 "f2_hrtbeat_DNS")   [jump to source](${starboardData.originalMessage.url})   ![f2_hrtbeat_DNS](https://cdn.discordapp.com/emojis/1146941657095340052.webp?size=128 "f2_hrtbeat_DNS")`)
                    .setTimestamp()
                    .setFooter({ text: footerText });

                const msgID = starboardData.messageID
                console.log(`Attempting to fetch message in channel ${channelID}`);

                console.log(`Channel permissions for bot: ${channel.permissionsFor(client.user)}`);
                console.log(`Is message in cache? ${channel.messages.cache.has(msgID)}`);

                await new Promise(resolve => setTimeout(resolve, 1000)); // even tried delaying to see if that would help give more time to get it cached

                console.log('Cached messages:', channel.messages.cache.map(msg => msg.id));
                try {
                    const fetchedMessage = await channel.messages.fetch(msgID);
                    console.log(`Fetched message: ${fetchedMessage.content}`);
                } catch (error) {
                    if (error.code === 10008) {
                        console.error(`Message with ID ${msgID} not found.`);
                    } else {
                        console.error(`Error fetching message with ID ${msgID}:`, error);
                    }
                }
            } else {
                console.log(message.id) //returns correct msg id
                client.starboard.set(message.id, {
                    originalMessage: message,
                    messageID: message.id,
                    count: count,
                    author: message.author.username,
                    authorURL: message.author.avatarURL()
                });

                const footerText = `♡ ${reaction.count}`;
                const embed = new EmbedBuilder()
                    .setColor('#f0deda')
                    .setAuthor({
                        name: `${reaction.message.author.username}`,
                        iconURL: reaction.message.author.avatarURL(),
                    })
                    .setDescription(`· ·  ![f2_hrtbow_DNS](https://cdn.discordapp.com/emojis/1146941520239394816.webp?size=128 "f2_hrtbow_DNS")   new heart  !!  + ⁺ .\n\nʚ・"${reaction.message.content}"・ɞ\n\n![f2_hrtbeat_DNS](https://cdn.discordapp.com/emojis/1146941657095340052.webp?size=128 "f2_hrtbeat_DNS")   [jump to source](${reaction.message.url})   ![f2_hrtbeat_DNS](https://cdn.discordapp.com/emojis/1146941657095340052.webp?size=128 "f2_hrtbeat_DNS")`)
                    .setTimestamp()
                    .setFooter({text: footerText});

                channel.send({embeds: [embed]})
            }
        }
    }
});```

[lyric mountain]

why don't you try fetching the message instead of using cache?

for messages it too unreliable to use cache

[pale vessel]

they're doing that

[lyric mountain]

where?

[lyric mountain]

ah

can you fetch messages globally? or must you do it on a channel?

[pale vessel]

must be in a channel

[charred nest]

im doing it ina channel for the starboard

[lyric mountain]

in java at least we do client.retrieveMessageById(), being able to do it on a channel too

the issue might not be on the message itself, but on the channel being wrong

log the fetched channel name to be sure

[charred nest]

logs the correct name
id: '964997630910427197',
name: '♤﹒﹒﹒﹒',
rawPosition: 24,
parentId: '964372181880553532',

[pale vessel]

why are u fetching the message though?

[sharp geyser]

its a starboard command or smth no?

you need the message you're going to starboard

[pale vessel]

you're also storing the user message not the starboard message

unless the starred message was sent inside the starboard channel it won't work

[sharp geyser]

wym by that

[pale vessel]

maybe u meant to send the message first and store the ID of this message instead

[sharp geyser]

they are storing/getting the message that is reacted to no?

[pale vessel]

it won't work if they want to edit the starboard message since the message ID is the reacted message's ID which wouldn't be from the bot

u can't edit someone else's message

here

messageID should be different, it should be the bot's sent message ID

[pale vessel]

from here ^

but right now it's the same message ID as originalMessage

[charred nest]

ah shit u right, i meant to do that