🧩-plugin-development

[green vessel]

:)

[gritty sparrow]

insane
thats the admin panel
nothing else

it has been recreated from our screenshots anyway

or at least the user page

[green vessel]

Send me it rq

The one currently is different,

it's the updated 2024 one.

They changed some stuff around.

[gritty sparrow]

its closed source i dont have access to it duh

[green vessel]

Well anyway, it's embeded into here:

"Discord"

or at least it's suspected

but its not true

[gritty sparrow]

dawg what???

[green vessel]

?

Most users think the admin panel is on that top button,

[gritty sparrow]

idk who is your so called "most users" but no its not

[green vessel]

Meaning most users who have no idea into the Discord Data Mining community

etc

[gritty sparrow]

okay, tell me where the admin panel is then

the website they're using

[green vessel]

Ask someone in the DDM

Only few amounts of people know about it,

[gritty sparrow]

@carmine spade

[gritty sparrow]

yeah and that would be me
moving on, give me the link

[green vessel]

You should know it :)

Wouldn't you?

[gritty sparrow]

yes, but do you know it?

just curious

[green vessel]

I do, but again. not willing to break h1 policies over a skid on Discord.com trying to compare code for code.

Which is more cringe.

Would of expected better from someone,

¯_(ツ)_/¯

[gritty sparrow]

buddy the admin panel is no secret

[green vessel]

It doesn't matter,

If it's reported. it's policies of H1.

Safe Harbor moment.

[gritty sparrow]

https://admin.discord.co

oh no!

i will get now banned from their h1

scary

[green vessel]

so real!

[gritty sparrow]

i sure hope no one reports it!!!

[green vessel]

It's legit cf's auth for 0 trust 💀

[gritty sparrow]

yeah no shit

[green vessel]

Meaning shit

[gritty sparrow]

it wouldn't be open

[green vessel]

let me get it

one sec

[gritty sparrow]

what will bro get 💯

[potent fox]

@gritty sparrow give me admin panel credidentals

I need for one important job

[gritty sparrow]

never

[potent fox]

I think we can make a agreement

[gritty sparrow]

nuh uh

[potent fox]

you are so bad

[gritty sparrow]

ikr

[green vessel]

crazy!

[dull magnet]

Locked out? Message IT in the #it-team channel.

[gritty sparrow]

what is the purpose of this screenshot

idg

[gritty sparrow]

i will

[dull magnet]

you both need to stfu tbh

[gritty sparrow]

:3

[dull magnet]

who cares bro

[gritty sparrow]

im just making fun of buddy

😔

[potent fox]

awful

[signal hemlock]

@green vessel did you seriously just send my screenshot of the admin panel

💀

[gritty sparrow]

💯

[green vessel]

I got IT to reset it 🥶

[signal hemlock]

bro cropped out application name

[dull magnet]

password is EpicDiscordoPassword24. don't tell anyone else

[gritty sparrow]

he actually didnt

[potent fox]

dont worry its secret of mine

[gritty sparrow]

its still up there

[signal hemlock]

explod

[gritty sparrow]

@dull magnet send 2fa code pls 🙏

[dull magnet]

273891

[potent fox]

husk is this 5 year old admin panel or smthn

[gritty sparrow]

thx worked

[gritty sparrow]

6 actually

the same one the osinted guy sent

[dull magnet]

don't let discord know i gave you my login

[gritty sparrow]

i wonder where he got that one from

[dull magnet]

they will fire me

[gritty sparrow]

dw my secret will be safe

erik wont know

🙏

[signal hemlock]

erik doesn’t snitch you can tell him

trust me bro

[gritty sparrow]

awww he left

:(

[signal hemlock]

:(

[dull magnet]

made him leave

bully

[gritty sparrow]

true...

[dull magnet]

no way bro forked every discord repo and starred his fork

https://github.com/stars/Osinted/lists/discord-stack

[gritty sparrow]

https://cdn.discordapp.com/attachments/1029315212521771020/1196090408720146562/image-146.png?ex=65b65cff&is=65a3e7ff&hm=403ed54541fc6fcb45c055c7e7a2eb01d5d4fda1bc297f5694c4dce182fed813&

he sent this

[glass oracle]

self appreciation 🙏

[gritty sparrow]

so from my knowledge of discord staff

[potent fox]

his only id is 203526789849415680

[gritty sparrow]

i tracked him down

[potent fox]

@chrome umbra

[gritty sparrow]

that discord staff is 1112974718518186075 - erik.burbulla

also i was preparing a ban gif for him

[dull magnet]

i skimmed the bio and thought wow 15 year old staff they hire young nowadays

[gritty sparrow]

real.

[potent fox]

I still dont get why people think discord admin panel is so hidden

[dull magnet]

why would he censor so poorly

[glass oracle]

me when post a screenshot of i <insert admin panel gif>

[potent fox]

like it of course exists

[dull magnet]

at least hide the bio as well

[gritty sparrow]

unfortunately the original was deleted

[gritty sparrow]

all i needed was the join date

got it from the start of the snowflake

[potent fox]

this is only for discord staff testing

[dull magnet]

[potent fox]

(only readme file)

[gritty sparrow]

yeah literally

[glass oracle]

i could make a new one

[dull magnet]

discord staff use that repo don't you know

[gritty sparrow]

chat is this real?

[glass oracle]

yeah they totally upload their DB migrations on a public repo

no shit sherlock

[gritty sparrow]

they would upload on a github gist

they love their gists

[potent fox]

what if this guy was actually discord employee

and he is gonna ban all of us now

[gritty sparrow]

then i would know

:3

[glass oracle]

bro would know

[austere mauve]

inside tetr.io's asar theres a readme thats supposed to be for their repo lol

[gritty sparrow]

and why would erik ask him if he has access to admin panel

duh

[austere mauve]

it had a snarky message at the bottom like "if you have a feeling you shouldn't be seeing this, go away"

[dull magnet]

adding that to all vencord repos

brb

[austere mauve]

true

[gritty sparrow]

he would get 50 rep for high/crit

• 7 for triage

he would have 57

unfortunate

[glass oracle]

me when an employee doesn't know all other 999 employess

[gritty sparrow]

he wouldnt be on the h1 list

[gritty sparrow]

tin flashbacks:

[glass oracle]

well it's 170 employees less now

[gritty sparrow]

its 830 now

[glass oracle]

bro's got count

[dull magnet]

831

[potent fox]

wth is reputation

[gritty sparrow]

h1 rep

[dull magnet]

china social credit system

[gritty sparrow]

yeah

[dull magnet]

remember tencent bought discord

[potent fox]

how do I incrfease

[dull magnet]

post china propagands

[potent fox]

https://tenor.com/view/social-credit-social-credit-score-credit-score-score-china-gif-23125701

[dull magnet]

every time you post this song you get 2 reputation

https://youtu.be/OjNpRbNdR7E

[gritty sparrow]

https://open.spotify.com/track/0tPCA7r3nk3T3kPQ6nAKSt

[dull magnet]

for every person that clicks the link you get an extra 0.5 reputation

[pliant pasture]

hehehe :3

[glass oracle]

find discord_web source maps, send in my dms, report on h1
ez reputation

[gritty sparrow]

yeah

i tried that one

[glass oracle]

kedas jumpscare

[austere mauve]

https://kagari.moe/outer_assets/tetrio/prvwarn.png

[glass oracle]

GET THE FUCK OUT

[austere mauve]

true!

[glass oracle]

gonna add that to all my repos

[dull magnet]

amazing alt text

[austere mauve]

tetr.io

tetr.io

[dull magnet]

https://ak.kawen.space/objects/ed41e0d3-bfb1-467a-a7d1-edab07fe1309

[proud parrot]

Katka :heart_bi: ([e-mail])

something poetic about this

Image

Click for image

Akkoma · Jan 14, 2024, 1:37 PM

[dull magnet]

they should make better rust

[glass oracle]

use even better toml

[austere mauve]

why did they do that

[gritty sparrow]

while at it you can find
discord_api
discord_ai
discord_developers
discord_android
discord_ios
discord_analytics
discord_devtools_extension

[austere mauve]

bettertoml is just taplo lsp

[dull magnet]

when Best Toml

[glass oracle]

better rust deprecated since 0.1.0 use even better rust instead

[potent fox]

wth is discord_ai

[glass oracle]

clyde

[gritty sparrow]

yuh

[glass oracle]

[potent fox]

isnt that only gpt

[green vessel]

@shrewd tundra

[potent fox]

venbot is dumb as hell

[gritty sparrow]

yeah venbot is actually hosted on discords servers

[austere mauve]

dont tell anyone its actually me

im the experiment

[gritty sparrow]

you're clyde

[potent fox]

okay

[pliant pasture]

Use the Even better TOML extension!

[glass oracle]

discord_analytics
discord_android
discord_api
discord_app
discord_design_system
discord_devtools_extension
discord_ios
discord_media_proxy
discord_protos
discord_static
discord_web
discord_marketing
discord_tools

[carmine spade]

discord_cuties

[gritty sparrow]

as in this?

https://cdn.discordapp.com/attachments/1142115777122676807/1150772358433484887/image.png?ex=65ae61c6&is=659becc6&hm=5ed9b9c180feac500f72e172abd44725082c2e012c26ab11f8ec0195e484b0ff&

[glass oracle]

someone should've leaked discord_api when being layed off

[austere mauve]

agony

[glass oracle]

yuh uh now show folder size

[gritty sparrow]

at least

3

[glass oracle]

you

[carmine spade]

no u

[glass oracle]

3tb

[dull magnet]

only noobs need source code

[gritty sparrow]

you know chat

[dull magnet]

true chads reverse engineer the minified code

[carmine spade]

real

[austere mauve]

discord_nsa

you will be banned

[gritty sparrow]

we should've asked brandon about admin panel guys

[austere mauve]

2ho the hell is tachiyommyo

[dull magnet]

jason to brandon the day he gets laid off: let's go brandon

[gritty sparrow]

manga reading app

(real)

[viral roost]

discord_pipebomb

[dull magnet]

idk why i always think of let's go brandon when someone mentions brandon

secret maga supporter

[glass oracle]

discord_watchlist

[viral roost]

the fuck is a lets go brandon

[glass oracle]

you will code in s-expressions

[gritty sparrow]

me includeed

[glass oracle]

(write eww widgets)

[dull magnet]

do you not know

google it

[glass oracle]

youre the only one in that directory

and your alts

[gritty sparrow]

{"type":3,"name":"ai-target","description":"Specifies which target/model to build and deploy for discord_ai"}

:3

[glass oracle]

oh right summaries too

[carmine spade]

am i included too? 🥺

[gritty sparrow]

all of hn is

[glass oracle]

leaving rn

[carmine spade]

true

[gritty sparrow]

(people that joined before nick incident)

[glass oracle]

husk

[carmine spade]

UNKNOWN_FLAG_23 - hn member

[gritty sparrow]

fuck that flag

[glass oracle]

we all have 9999 🙏

[carmine spade]

overflow

[glass oracle]

works on my machine

[carmine spade]

[viral roost]

mhm yeah i respect the tos

[potent fox]

you need minified code first tho

[glass oracle]

what theme is that

[viral roost]

i am totally not actively breaking it nop i am a good discord citizen

[dull magnet]

purple alien fart theme

[carmine spade]

we're all "breaking" it by using veecord

[glass oracle]

purple dolfies fart theme

thanks

[carmine spade]

wha

https://tenor.com/view/i-saw-what-you-deleted-cat-gif-25407007

[gritty sparrow]

sad shit

[carmine spade]

(apply HAS_UNREAD_URGENT_MESSAGES to yourself to make it more realistic 💯)

[gritty sparrow]

true

[sand portal]

[nocturne apex]

vencord shikiblocks not available ?

[fluid fjord]

my plugin is in the userplugins folder, I assume that should work too.
I ran pnpm build and it built the files in the dist folder.
can I make VencordInstaller.exe use these files instead of the main vencord ones?

[dull magnet]

pnpm inject

[fluid fjord]

yea that works for me but I didn't know how I can make my friend install it without having nodejs etc. installed
I see now that there's a VENCORD_DEV_INSTALL env var for the installer, I think that's what I needed. I'll try that thanks

[dawn echo]

is it just me or did discord discontinue the "streaming" activity type

"type must be one of [0,2,3]"

[frigid bay]

theres someone streaming right now. so

unless that works differently now

then idk

[dawn echo]

man what the hell

[fluid fjord]

okay that worked 👍
I made a zip file like this with the batch file being

set VENCORD_DEV_INSTALL=1
set "VENCORD_USER_DATA_DIR=%cd%"
start VencordInstaller.exe

[dawn echo]

the discord api makes me want to hurt people

[frigid bay]

do it

[dawn echo]

like discord is just telling me to fuck off lol

"type must be one of [0,2,3]" 🤓

[gritty sparrow]

@clear parcel you love

[dawn echo]

https://discord.com/developers/docs/topics/gateway-events#activity-object
.....
"name" is not allowed..

WHAT THE FUCK

FIELD TYPE DESCRIPTION
name string Activity's name

ITS RIGHT THERE

ITS REQUIRED

WDYM "NO"

the documentation is gaslighting me

[frigid bay]

theres unofficial docs for discord stuff

maybe those are better

https://discord.sex/

🛸

[dawn echo]

discord.sex

gahhhhh it copied the official documentation

[frigid bay]

its not 1:1

theres additional stuff

[dawn echo]

is my discord application sterilized or smth

[frigid bay]

maybe

[dawn echo]

why cant i use like 90% of the features

"id" is not allowed, "created_at" is not allowed, "name" is not allowed, "url" is not allowed

[viral roost]

nahhh who bought that

[frigid bay]

lol i just went on userdoccers and it redirected to that

btw i dont actually know how to help you so dont feel lost when i cant help

i dont do discord stuff its just that ive heard of unofficial docs and thought those might help

[dawn echo]

k

ill go cry in a corner or smth

my ass is NEVER developing discord stuff again

wait i can leech off of other peoples discord applications xdd

[gusty maple]

Me

[clear parcel]

where are you getting this

rpc?

because rpc never allowed streaming

[dawn echo]

a

alrighty

what allows streaming

[clear parcel]

what are you tryna do

[dawn echo]

the goal is to make a streaming thingy

basically rip off the vencord rpc thing

so i can change the status to streaming without actually streaming anything

[clear parcel]

as in a plugin?

[dawn echo]

yeah

[clear parcel]

why duplicate it lol

but either way set the activity within the client

[dawn echo]

k

[clear parcel]

don't try to use rpc

[dawn echo]

because i am a filthy betterdiscord user addicted to betterdiscord plugins that dont work on vencord

[dawn echo]

welp that explains the errors..

back to hunting for solutions!!!

[green vessel]

For some reason can't talk in #👾-core-development but is it possible to stop the client from compressing images upon upload?

[frigid bay]

it doesnt do that for me on pc

only changes metadata

i tried with a jpeg

[warm gulch]

(Flatpak Linux testing)

Hellou vencord users

I'm setting up Rich Presence (RPC) support to Flatpak Anki

If you want to test out see

https://github.com/flathub/net.ankiweb.Anki/pull/133#issuecomment-1891043930

[balmy sky]

Is there a way i can use the .split() method without removing the split character? if i split "one. Two" by . for example, it will return one and two but i want it to include the punctuation in the array element

[dull magnet]

i dont think pring this to a random flatpak is a good idea

users can do this globally

[warm gulch]

@dull magnet

Could we add Vesktop in Downloads

https://vencord.dev/download/

https://flathub.org/apps/dev.vencord.Vesktop

[dull magnet]

yes soon

[dull magnet]

holy shit @trail ginkgo you gave me a good scare

i was skimming vencord bundle.js

and i got a mini heart attack from this

thinking i messed up bundler somehow

[trail ginkgo]

LOL

how is that even there

ohh its the shiki example

[dull magnet]

its the shiki example

xd

[austere mauve]

the eslint disable header goes hard

[trail ginkgo]

LMAO

functional and helpful

[austere mauve]

big part of the example

[dawn echo]

tyyy :3

(that took WAY too fucking long)

[clear parcel]

np

[dull magnet]

@grim hare btw, plugins now can have their own native code

you can likely use electron frame created event to patch the frames src and it likely can automatically bypass csp

#1032200195582197831 message

[grim hare]

ooo

i'll take a look maybe tomorrow since I have a holiday

anywhere I could quickly reference to see a plugin with native code?

[dull magnet]

fixspotifyvolume

it does exactly what you need

patching iframe

it's incredibly simple you just create a native.ts file and that will be bundled with native code

any exported functions will automatically be turned into ipc methods

really need to work on docs

[grim hare]

got pulled away but

that sounds great

I'll definitely take a look tomorrow

[grim hare]

sad day

frame.url is readonly so I can't assign to it to change the frame URL

I can use JS to redirect (what I tried above) but that hits CSP still

unless I'm missing something pretty obvious I don't know that there's another way to patch the frame's src property

oh wait I have an idea

nvm, since vencord's csp patcher noops webrequest.onHeadersReceived I can't do what I was planning to

however

since plugins now have an easier way to hook into native I think building that csp api thing that has been thrown around in the past would be pretty easy

I might see if there's something I can do

[grim hare]

alright I got something that works but it does require some slight changes to src/main/index.ts

[dull magnet]

can you somehow cancel the frame creation and create a new one with the context from old one?

i honestly don't know much about the frame api

Also can you change webframe options? if yes you could disable csp enforcement for it

[vast karma]

I'm trying doing import traverse from "@babel/traverse"; in a node project, which according to docs should give a function, but instead it gives an object where .default is that function

What am I doing wrong?

[dull magnet]

nothing

[vast karma]

Are the docs wrong then?

[dull magnet]

the module is bundled wrong it seems

// le scuffed module
import * as babelTraverse from "@babel/traverse";
const traverse = babelTraverse.default;

[vast karma]

I love the js ecosystem

[dull magnet]

wait

nvm it would be .default.default

// le scuffed module
import babelTraverse from "@babel/traverse";
const traverse = babelTraverse.default;

funny

[grim hare]

I can't seem to find a way to cancel frame creation (preventDefault doesn't do anything on the event object and stopPropogation doesn't exist contraray to what the types say) - even if I could cancel creation there's not a way for me to create a new frame

admittedly the event kinda feels like a glorified way to tell if an iframe has been created on the page or not, it's incredibly limited aside from being able to execute some js or send a message to the frame

[dull magnet]

hmm

i think i know a way

hmmmm

you can disable csp for the frame via https://www.electronjs.org/docs/latest/api/web-contents#contentssetwindowopenhandlerhandler

but

you can't actually use that because you will overwrite discords handler

[grim hare]

that also doesn't apply here unfortunately

the heiarchy as I can tell is BrowserWindow -> webContents -> webFrameMain (what an iframe is)

I can't change the webSecurity property on an iframe directly, it's only on webContents

an iframe doesn't have its own webContents

[verbal blaze]

can i share programs i made in here?

[thorn drift]

i wanna see

[verbal blaze]

alright so i made a program that makes markiplier name soundalikes💀

[jagged briar]

As long as they aren’t viruses

[thorn drift]

lmao

most sane use of python i've seen this year

[verbal blaze]

idk how to make viruses idek how to make vencord themes😭

[verbal blaze]

fr

[thorn drift]

How do you guys feel about committing documentation changes directly to main without creating a PR and then merging it yourself? Is it bad or is it fine?

[verbal blaze]

if only i knew what that means😭

[thorn drift]

so basically updating the readme without asking nice

[verbal blaze]

ohhhh

[thorn drift]

the tradeoff is it being more of a pain to see who changed what and when vs having a cleaner pull request history.

I think its fine but I also thought a lot of other things were fine that aren't so that's why I'm asking chat.

example I realized was bad today: leaving a super long error dump log fully expanded when opening an issue instead of collapsing it

(I went back and fixed all of them)

[thorn drift]

i've decided I'm just going to keep doing it and then say sorry if someone gets angy

[rocky idol]

how do you get members in a server

[vast karma]

Kidnapping is efficient

[high thistle]

[thorn drift]

writing a function somewhere else that only ever gets used once to call it is taking clean code too far @frigid bay

[dull magnet]

not necessarily

splitting your code into separate functions is good

even if you only call the function once

its easier to read if you split every section of the "pipeline" into separate functions

[thorn drift]

so if the function is in a completely different file you can understand that faster than having it inline at the single location its needed?

[dull magnet]

completely different file is weird

it should be same file if its use is related

[frigid bay]

I never said different file, it's the same file

maybe I should've kept the messages my bad, I deleted cus I thought question was vague and think I decided I'll keep it for now at least

I have a function that inits everything but there's groups of things that are inited if that makes any sense

something like this for example (pseudocode)```cpp
// all the init functions return a status that can be success or failure
status init_x() {
if (init_xx() != success) return failure;
if (init_xy() != success) return failure;
if (init_xz() != success) return failure;
return success;
}

status init_y() {
// similar to init_x
}
status init_z() {
// similar to init_x
}

status init() {
if (init_x() != success) return failure;
if (init_y() != success) return failure;
if (init_z() != success) return failure;
return success;
}

then init_x inits things related to x, init_y inits things related to y, and so on

[dull magnet]

whats success defined as

[frigid bay]

i was wondering if i should have it like that, or if i should inline them

[frigid bay]

i actually just make it a bool and its true on success, i wrote it like that so its easier to read for u

ik its a bit ugly but

oh well

offtopic anyway

its dumb but shush

[dull magnet]

then why not

bool init() {
  return init_x() && init_y() && init_z();
}

[frigid bay]

ig something like that could work too yea, didnt think of that

[frigid bay]

although the init_x/y/z funcs are just more of this

idk if i like it but

oh well

[frigid bay]

i edited this a little

i was wondering if instead of this i should do this: (give me a minute to write it :p)

status init() {
    if (init_xx() != success) return failure;
    if (init_xy() != success) return failure;
    if (init_xz() != success) return failure;

    if (init_yx() != success) return failure;
    if (init_yy() != success) return failure;
    if (init_yz() != success) return failure;

    if (init_zx() != success) return failure;
    if (init_zy() != success) return failure;
    if (init_zz() != success) return failure;

    return success;
}

thjere would be more than just 3 of each

doesnt rly matter in the end but i cant decide which is nicer

actually the code could be different than just similar copies of if (init_stuff != success) return failure

and it would init/do different things too

thats why i was thinking of separating it

ill just keep the separate functions at least for now

[frigid bay]

either way ty for helping and also ty for suggesting this even tho its unrelated to what i originally asked, i remember wondering what the cleanest way to do something like this would be, maybe ill end up using this in the future if its applicable

[dull magnet]

you could make an array with ur init funcs

and loop over it

[pure temple]

horror

cleaner if you inline :)

[dull magnet]

or make a macro

#define INIT(NAME) if ((init_ ## NAME)() != success) return failure;

status init() {
  INIT(xx)
  INIT(xy)
  INIT(xz)

  return success;
}

maybe too cursed?

[pure temple]

i don't think you know what "clean code" means

you love :3

std::optional<CommandSpec> commander_parse(const std::string &in) {
    CommandSpec result;

    std::istringstream stream(in);

    std::string ids;
    if (!std::getline(stream, ids, ' '))
        return {};

    std::istringstream ids_stream(ids);
    std::string id;
    while (std::getline(ids_stream, id, '|'))
        result.ids.emplace_back(std::move(id));

    std::string word;
    while (std::getline(stream, word, ' ')) {
        std::string name;
        if (word.starts_with("--")) {
            name = std::move(word.substr(2));
            if (!std::getline(stream, word, ' '))
                return {};
        }

        FlagSpec flag{};
        if (word == "?")
            flag.type = FlagType::Void;
        else {
            if (word.ends_with('?')) {
                word.resize(word.length() - 1);
                flag.mode = FlagMode::Optional;
            } else if (word.ends_with("...")) {
                word.resize(word.length() - 3);
                flag.mode = FlagMode::Greedy;
            }

            if (!FLAG_TYPES.contains(word))
                return {};

            flag.type = FLAG_TYPES[word];
        }

        if (name.length() != 0)
            result.named_flags[name] = flag;
        else {
            if (result.unnamed_flag.has_value())
                return {};

            result.unnamed_flag = {flag};
        }
    }


    return {result};
}

[frigid bay]

this is kind of what I did at first but inside the init funcs

there's init funcs on top of init funcs lmao

idk its kinda shit now but its consistent

or maybe not shit idk

im not quite happy with it but its either that or no error handling (or exceptions.. which i wanna avoid)

[pure temple]

if you do this, what's wrong with you

[frigid bay]

the macros? nah

i did at first

now its.. template functions

[pure temple]

why would you segment your init function like that

[frigid bay]

theres a lot of different stuff i init inside it

i dont wanna inline it all cus it would be a bit of a mess

[pure temple]

aren't they operating on different data anyway so you'd need to use a different function :p

unless this is C(-style)

[frigid bay]

ill maybe show some more code in a moment

so its easier to understnad

but maybe there is just a better way to do it and im dumb

[balmy sky]

Why is this

What

Just the entire codeblock being shown in replies

[frigid bay]

lol

[flint bronze]

this is why I don't make Vencord plugins (I was going to make a "force show member safety page" plugin but the API behind it is locked and I can't be bothered finding members normally)

[trim cedar]

Never

[trim cedar]

std::expected :)

[trim cedar]

Maybe you should restructure generally

[trim cedar]

You'd probably be better of calling the discord functions instead of the API?

[flint bronze]

I was testing to see if it would work in theory

I wouldn't've needed to even call the functions to do it

All I wanted to do was always show a button/enable a menu without permissions

and the menu calls stuff that definitely needs the permissions :/

Back to looking at other stuff to improve :)

I will also confess that I have never submitted a PR to Vencord before

[spare sky]

hi
I'm coding a plugin
I'd like to know if it's possible to get a tampermonkey script, instead of injecting vencord.

[dull magnet]

pnpm buildWeb

[spare sky]

thx

[frigid bay]

I was thinking the same, I'm rly stuck on this and can't end up with anything I'm satisfied with, but this is good enough

[grand haven]

external plugins when

[viral roost]

never

[grand haven]

cri

external plugins plugin time it is

https://cdn.discordapp.com/emojis/722917812649197711.gif?size=48&name=PepeLaugh~1&quality=lossless

[vast karma]

That's probably not impossible tbh

The major obstacles, as I see it, are

• I don't think all apis are exposed at runtime (i.e. only via import, not in window.Vencord)
• Making a nice ui for it
• Fending off the angry vees

[jagged briar]

If we wanted to we probably could but obviously we don’t want to

That’s the whole point of vencord

[vast karma]

A plugin for loading plugins would be rather meaningless tbh
Such a plugin has approximately negative 47% chance of being accepted upstream, so you can't use it without building from source
And if you're already building from source, you don't need a plugin for loading stuff

[jagged briar]

Exactly

[spare sky]

Is it possible to launch a console script (web) from a vencord plugin?

[frigid bay]

;

[spare sky]

?

[frigid bay]

'\

[spare sky]

‎ ¯_(?)_/¯

[frigid bay]

;;;;m

[spare sky]

(_/)
( •.•)
/>?????

[jagged briar]

Elaborate

[frigid bay]

i think they just want to run js through a plugin

which i thought was the point of a plugin..

except for the regex patches

[vast karma]

Plugins are indeed capable of executing javascript

[spare sky]

Yep

[jagged briar]

If it’s user input good luck if not then use exec

Also this sounds like https://xyproblem.info

Oops

[spare sky]

How ?

[frigid bay]

.info i believe

[jagged briar]

Thanks lol

[frigid bay]

:p

[jagged briar]

Look up 2 messages

[spare sky]

Haaaaaaa

Ty !

[pure temple]

textreplace scripting will change everyone's life 🚀

[jagged briar]

True

[balmy sky]

Grammar corrector thingy

Idk

Im just too lazy to add an apostrophe

[frigid bay]

Im

[vast karma]

Apostrophe's are pointles's an'way

[jagged briar]

dont tell me you made a plugin for this

[frigid bay]

i assume its textreplace

[balmy sky]

I did

[frigid bay]

prob not

[jagged briar]

textreplace is your friend

[balmy sky]

Would textreplace even work for this

[jagged briar]

yes

[balmy sky]

What about capitalisation

Like if i type dON't

It does it correctly

You might be able to do that with textreplace regex idfk

[jagged briar]

you can totally do that

it's not hard

[balmy sky]

Welp too late now

[floral plaza]

i mean if you can use a case insensitive regex it will

[vast karma]

(\b[dD][oO][nN])([tT]\b) → $1'$2

Or yeah insensitive

[balmy sky]

I mean eh it's fine, if it works it works

[weak idol]

tessie w

[hushed bloom]

[raw portal]

do third party plugins you make get added to to the offical menu

like if u want or somthing

[vast karma]

They're in the same page in the settings, yes

[raw portal]

no i mean do they add them to the offical thingy

[vast karma]

If you make a pull request and it is accepted, I guess

[raw portal]

gotcha

also is there like a template i can use to try out

[vast karma]

Just check what all the other plugins do

[raw portal]

can i just xopy and paste e, like a third party one or are they made diffrently

[trim cedar]

the basic structure is the same

[amber basin]

wtf

trying to patch a website and

its webpack but without the webpack

??????????

[dull magnet]

yeah

that's when they don't use chunks

it's still webpack tho

[amber basin]

how would i go about webpack patching then

[dull magnet]

you can use prototype pollution to get access to the webpack instance

look at https://github.com/Vendicated/WebpackGrabber

[frigid bay]

if you look closely it says "grabber" here.... further proof of vencord being malware..

[vast karma]

I thought that was obvious

[amber basin]

webpackgrabber doesnt get it :c

i put a console.log in it and it looks like its not setting Function.m

[dull magnet]

send website

if wpgrabber doesn't get it then it's likely a different bundler

ill check soon once i get on pc

[amber basin]

its the vrchat website

https://vrchat.com/home

trying to fix a bug that crashes the ui

[vast karma]

Yep, that thing doesn't define any m

[balmy sky]

Which part of the code is responsible for zooming in/out the client? I'm trying to modify it so that the scaling is linear and smooth instead of snapping, but i'm having trouble finding the relevant code.

[amber basin]

its using some mangled form of webpack, but all the variables are renamed i think

[balmy sky]

Oh boy

[vast karma]

Isn't that handled by the browser

[balmy sky]

I don't think so? on desktop you can ctrl +/- to zoom

[frigid bay]

yeah but isnt desktop basically a web browser anyway

[balmy sky]

True

[frigid bay]

on web its handled by the browser

and it looks the same as on app

nearly 1:1 same

[vast karma]

Why would electron make their own zoom function when the browser it's wrapping already has one

[dull magnet]

their webpack is really interesting

[frigid bay]

i tried asking gpt 3.5 if it can make my code cleaner to see what it can do but all it did was change the naming convention and indentation 😭 this ai is so useless for code

"Your original code seems fine in terms of functionality." fuck offf

[dull magnet]

yes

ai suxx

so dumb

[thorn drift]

if anyone speaks github actions please help. it brokey. 😭

on: [push]

jobs: 
 build:
  runs-on: ubuntu-latest  
  steps: 

    - name: Automated XPI build process
      uses: TheDoctor0/zip-release@0.7.6
      uses: my/extension/directory
      with:

        filename: nightly-release.xpi
        type: 'zip'

        exclusions: '*.git* *.ps1* */gre-resources/* */other-tools-and-tests/* *.editorconfig'
        directory: .
        path: .```

[frigid bay]

ok i got it to do something when i told it exactly what i was trying to achieve

idk if its even valid code because it uses types im not familiar with

i should really learn the c++ stl...

[floral plaza]

your items arent on the same indentation

[frigid bay]

i wonder how much better gpt4 is, id like to try it but im definitely not gonna pay for it and neither do i write code that often so i wouldnt even use it that much

[floral plaza]

wait nvm comments

[thorn drift]

on: [push]

jobs: 
 build:
  runs-on: ubuntu-latest  
  steps: 

    - name: Automated XPI build process
      uses: TheDoctor0/zip-release@0.7.6
      with:
        filename: nightly-release.xpi
        type: 'zip'

        exclusions: '*.git* *.ps1* */gre-resources/* */other-tools-and-tests/* *.editorconfig'
        directory: .
        path: my/extension/directory```
this version is marginally less broken but still fails with a nothing to zip error

I tried 17 times lol

[dull magnet]

oh my god

remove the useless comments

terrible

[thorn drift]

1 useless comments removed later

but how do i fix it 😭

[floral plaza]

why aren't you checking out the repo

[thorn drift]

because I have no idea what I'm doing

[floral plaza]

have you read https://docs.github.com/en/actions/learn-github-actions/understanding-github-actions

[thorn drift]

IT WORKED OMG TYSM

[thorn drift]

i tried and then when i got to page 4 of me not understanding i gave up and went to stack overlfow with my errors

[dull magnet]

yeah how do u expect it to do anything without a checkout 😭

also i recommend using long commit hash instead of tags

TheDoctor0/zip-release@0.7.6

this is unsafe

[amber mantle]

horror

make vrchat website client mod

the time has come and so have i

[frigid bay]

ik this sounds like a big xyproblem but does anyone know how i could make a global variable in a header file that is only accessible from said file in c++

i tried the static keyword and unnamed namespaces

idk if its possible but if it is then @ me

[pliant pasture]

Have you tried using static variables inside a function?

Or you could also try at managing how global variables link up by splitting where you declare and define them into different files, It's a bit on the advanced side though

you can manually control the linkage of global variables by separating the declaration and definition of the variable in different files

[frigid bay]

yup I have, that's the first thing I tried, it works but causes something dumb that I'm trying to optimize, maybe a better compiler can solve it tho idk

I mean the static var in func

[mystic sable]

what is the error ur getting?

cause there's like 10 kajillion things u can do but what is the problem created by the one that works and why does other methods not work

stackoverflow rule appreciation moment

[frigid bay]

there is no error

I don't like the assembly generated by the static var it's pointless maybe clang does a better job at making it not shit tho I haven't tried

[mystic sable]

dude

[frigid bay]

Yo

[mystic sable]

then write the assembly

[frigid bay]

there should not be any assembly

global var does the job

[mystic sable]

why is there assembly then

also

what kind of c++ file is this

static is one thing but also assigning a variable as private

which is stupid ik but like

micro-optimization is the killer of code lol

and if static vars are creating issues in ur code then it could be an actual code problem not a var type problem

[frigid bay]

because it initializes stuff for no reason when it can just be a value in .rdata idk how else I can describe it

[mystic sable]

okay so this is a using c++ problem

since quite frankly it's a quirky language

[frigid bay]

it's probably a compiler problem idk I use msvc

i tried making it const static but it treats it like it isn't const

and does some bs

[mystic sable]

send me ur fucking code in a dm i'll check it for stupid shit

cause if it's ignoring const you might have bigger problems

[frigid bay]

It's a 2 line template function maybe the template is an issue

[mystic sable]

yes

correct

[frigid bay]

I'm on mobile I can't send shit

💔

[mystic sable]

bro writing c++ on mobile?!?

[frigid bay]

no I'm just not on my pc

lol

[mystic sable]

lol

yea templates are

weird

[frigid bay]

templates are silly but it should still just put the thing in rdata which is basically just read only stuff that doesn't need to be inited or anything

I just want to pass a pointer to a const 32 bit value

to a func

and have it not be dumb

that's all

[mystic sable]

c++ is not good with passing shit inherently

it's extremely hard typed

[frigid bay]

idk why it's like that, I don't care about small optimization THAT much but I wanted to see if I could make it a bit less dumb nonetheless :P

because it does not need to be doing whatever it's doing

[mystic sable]

idk look it up on the c++ creators q&a

[frigid bay]

like it's a very simple thing idk why it's so overcomplicated

[mystic sable]

probably has a reason

[pliant pasture]

I fucking love header files. Truly i need truly, I need more header files in my life like I need a hole in my head. There's nothing quite like the thrill of scrolling through hundreds of lines of declarations, searching for that one elusive function prototype. It's like a treasure hunt, but instead of treasure, it's a pointer to a function you can't remember writing. Ah, the sweet, sweet smell of a build breaking because of a missing semicolon in a header file you didn't even know you were using. Truly, header files are the gift that keeps on giving ❤️

[frigid bay]

also when I make it a global const variable it just like.. makes duplicates of the value? for some fucking reason?

[mystic sable]

what are you using to debug

[mystic sable]

i unironically enjoy header files

i'm the anti-java

[frigid bay]

I'm just dropping the binary into a static disassembler/decompiler LMAO I know it's dumb but I'm used to it

[pliant pasture]

[mystic sable]

bro

bro

bro

bro

[frigid bay]

oh I don't use it to debug don't worry

only for dumb shit like this

[mystic sable]

are u a college student

[frigid bay]

no

[mystic sable]

are you a student

of any kind

[frigid bay]

no

[mystic sable]

okay then just find a debugging program or plugin made for c++

[frigid bay]

i know how to debug

I don't wanna debug I wanted to analyze stuff statically to see if the compiler was doing dumb shit to a func I wrote

[mystic sable]

okay but a debugger can do that better than you looking at that shit

[frigid bay]

I wanna look at code not debug it

[mystic sable]

okay

[frigid bay]

in this specific case

[mystic sable]

do you understand what a debugger does

[frigid bay]

yes

[mystic sable]

what does it do

[frigid bay]

dude

what is this lmfao

[mystic sable]

i feel like you don't understand that a debugger can point you to exactly where the thing you want to figure out is happening

like looking under the hood

[frigid bay]

I don't want to run any code and neither is there a bug anywhere in the code

[mystic sable]

that's not the only reason to use a debugger!

debugging programs also allow you to easily see where shit is happening

like if your function is generating this stuff and you don't like it

you can look under the hood and see "oh c++'s blah blah blah file is creating this blah blah blah and the forums say blah blah blah avoids this"

[frigid bay]

by generating I don't mean it's doing something wrong

the result is as expected

[mystic sable]

of course but it's doing something you don't desire, correct?

[frigid bay]

in the generated asm it's doing stuff that I feel like should be optimized but is not

[mystic sable]

so you can see, when using a debugger, where the asm generation is doing what you don't want to do

debugging is a bad term

refactoring is a better term

because your code works, but you're optimizing it

debuggers can be both used to debugged but are even better for advanced refactoring if used optimally

[frigid bay]

yea but it's not even really related to my code, it's just that declaring a const static variable is doing dumb stuff that it doesn't have to be doing :p so idk if there's anything I can do about that

ik what causes it to be doing what it's doing I just don't know what to do about it

other than a global const variable

[mystic sable]

again using a debugger may show you the answer in allowing you to look up how to avoid the specific thing you're seeing from generation as the c++ insides

like "file **.c line *****"

[frigid bay]

well idk whats possibly avoidable in const static uint32_t test = 0xffff7fff; i might just be completely misunderstanding and being stupid if that's the case then I'm sorry

[mystic sable]

so when the compiler compiles that, it's converting a pointer to another value

a debugger will show you where that conversion is

[frigid bay]

no pointers in the code I sent

[mystic sable]

well besides the pointer, it will show you what file the asm is being generated in

[frigid bay]

if u mean the specific line of code I sent then there are no pointers there

[mystic sable]

yes but the compiler isn't magical with curtains close

it has to reference another file to understand the input

a debugger can show you where in another file the magic is happening

[frigid bay]

not sure how this relates to anything if I gotta be honest, this is all done at compile time anyway

[mystic sable]

😭 im too tired to explain better

[frigid bay]

me using unsigned long instead of uint32_t won't change anything :p

if that's not what u mean then idk

[mystic sable]

i mean the debugger can show you where ur asm code is being generated from as in where the conversion from ur typed function, as in the middle man

like ur code -> msvc (debugger shows u here) -> asm

[frigid bay]

I don't think any kind of conversion is related to the issue tho :P I'm only declaring a static const 32 bit variable then passing a pointer to it

I could try doing uh

const static unsigned long test = 0xffff7ffful; this should have no conversions, unless ur describing some other kind of conversion but I dont think anything else would be related to this anyway

passing a pointer shouldn't convert anything and actually using the variable doesn't cause the issue I think, I'm p sure it's just that the declaration causes some avoidable nonsense since it's const but there's some weird bs being done that would only matter if it wasn't const

idk I wanted to sleep more but I'll prob just get up maybe I can send some code and maybe it can clear things up

godbolt compiler explorer my beloved might be able to help maybe

cus I think I can show the issue in 2 lines of code

"issue" not even rly an issue just lack of optimization I guess

I could also just completely ignore this because in the end it makes basically zero difference but I'm a bit of a nerd about this kind of stuff so might as well try to optimize it :p

[mystic sable]

ignore it, get sleep, goodnight

[frigid bay]

Compiler.. From OHIO!! 🤣🤣🤣

[frigid bay]

I honestly unironically feel bad for testing your patience all cause of a dumb meaningless thing the compiler won't do I want to apologize

I appreciate you helping though

[trim cedar]

no

[trim cedar]

if you don't need to modify it make it (static) constexpr?

[trim cedar]

if you pass it by value? that's to be expected, if you don't want that make it a macro or constexpr

[trim cedar]

have fun debugging the compiler lmao (especially when it's a propriatary one like msvc lol)

[trim cedar]

also - what compiler are you using?

[frigid bay]

pointer not value

msvc

I'll show code and stuff soon

[frigid bay]

tbh i have not tried that but i thought constexpr only differs from const for compiletime stuff

the thing is not compile time

global variable works but its a bit weird

[trim cedar]

uhm

so the value is not fixed?

[frigid bay]

ill explain from scratch in a moment

[mystic sable]

lol ur good

[trim cedar]

Meaning you initialize it like that: static const auto something = some_func()?

or how are you gonna initialize a const global?

[frigid bay]

const static uint32_t a = 12859071;

does constexpr differ from const here? if its used in runtime

[trim cedar]

depends on the compiler, but if that is your actual line why would you not make it constexpr?

[frigid bay]

fair point

[trim cedar]

that can already be the issue lmao

[frigid bay]

yeah lol

ill see if i can show waht the "issue" is in godbolt

im yet to try clang still

[trim cedar]

that should always be the first approach to issues like this

[frigid bay]

yeah i was tired and went to bed

[trim cedar]

https://godbolt.org/z/P5vTbTEG1

[frigid bay]

and also kind of fixed it with that global var anyway

[trim cedar]

no copy on gcc

[frigid bay]

yeah thats not exactly the issue ill explain in a moment

its like that in msvc too (but for some reason theres multiple deadbeef copies in rdata, even if unused...? idk what thats about.. but oh well

[trim cedar]

on godbolt? it's only one

[frigid bay]

in my binary

ill show what i mean in a moment

could be template func dumbness

[trim cedar]

if it's not constexpr and in a header (and not inline) that's to be expected when you include the header from multiple places

[frigid bay]

hm true that could be it

[trim cedar]

remember

[frigid bay]

idk why theres unused ones tho

[trim cedar]

the compiler just copy pastes it (the header) into your source file when you include it

[frigid bay]

yea true

[trim cedar]

option a: the compiler is retarded (it's msvc so it might just be that)
option b: you're doing stuff where that needs to be done

[frigid bay]

🫨 i think literally making it constexpr instead of const fixed the issue i had, i feel really stupid idk why i hadnt tried that

ok so the issue i had is i had a const static variable in a func but it was being initialized for some reason instead of just being a value somewhere in .rdata

if that makes any sense? sry if im being dumb about this

[trim cedar]

in a func

[frigid bay]

yep

i know

but its const

[trim cedar]

That is different from other static initialization

[frigid bay]

yea

ik

[trim cedar]

Doesn't matter

[frigid bay]

im aware

[trim cedar]

It gets initialized the first time it's seen

[frigid bay]

yep

[trim cedar]

even if it's const

const doesn't matter here

The initialization could have side effects

[frigid bay]

ok i think i get it now actually

[trim cedar]

in cases like that simply make it constexpr, especially since it's just a primitive

[frigid bay]

yeah it was 5am and i half forgot what constexpr did i guess

i was confused because i just made it a constant that was known at compile time

so i wasnt sure why it was initialized

but i guess the compiler needs to be told that its a constexpr

which makes sense

i kinda forgot

now its doing exactly what i expected it to do

[trim cedar]

The "hard part" of C++ is to know how to express to the compiler that you want this one specific thing to happen

Which can get quite tricky sometimes, especially with all the rules there are

[frigid bay]

:P that thing with forceinline noexcept i mentioned some days ago still pisses me off bc its the only reason i entirely disabled c++ exceptions

[trim cedar]

not sure what that was to be honest xd

[frigid bay]

ah yea fair

it just refused to do what i wanted to do is my point

idk if it was me being stupid (the code is over 2 years old and lost now because version control is for losers) or if it was a msvc moment

[trim cedar]

I mean in the worst case there's always compiler specific attributes that can force the compiler to do something

[frigid bay]

ig but the thing is that i tried to do just that and it wouldnt do it, probably just me being dumb tho as i said, cant check now

either way thanks for reminding me to try constexpr lmao i told myself "no way that it makes a difference here" and just never tried it until now loll

lesson learned

[trim cedar]

The __forceinline keyword overrides the cost-benefit analysis and relies on the judgment of the programmer instead. Exercise caution when using __forceinline. Indiscriminate use of __forceinline can result in larger code with only marginal performance gains or, in some cases, even performance losses (because of the increased paging of a larger executable, for example).

https://learn.microsoft.com/en-us/cpp/cpp/inline-functions-cpp?view=msvc-170

(MSVC specific though!)

[frigid bay]

yea it just wouldnt inline it even tho i used that for some reason

if i remember it right

[trim cedar]

You can't force the compiler to inline a particular function, even with the __forceinline keyword

that probably explains it

[frigid bay]

The function and the caller use different types of exception handling (C++ exception handling in one, structured exception handling in the other).

arrgh

it s probably the caller

that was the issue

[trim cedar]

is the caller winapi?

[frigid bay]

thats what i was thinking but as i said i cant check now bc the code is gone

[trim cedar]

If the compiler can't inline a function declared with __forceinline, it generates a level 1 warning, except when:

Should check your warnings :P

[frigid bay]

huh, i swear i wouldve noticed if i got one

ugh lol oh well

im actually writing something very similar to that old project now

might see if its still an issue

maybe

idk i dont use anything that uses c++ exceptions right now anyway

so no need

[trim cedar]

may I ask what? Because forcing inlining is usually nothing you'd care too much about even in performance critical software

[frigid bay]

the func in question was a helper func for calling vmethods

[trim cedar]

ah, well I guess that would make sense there

[frigid bay]

yeah

i prob forgot a noexcept somewhere

that one time

[flint bronze]

bruh

Discord does NOT want me to make plugins (The functionality exists in the normal client but hidden away behind a keyboard shortcut)

[frigid bay]

Sorry to hear

[flint bronze]

i guess I can still patch that to be a smaller number

[trim cedar]

oh wow is that a notification mute shortcut?

[flint bronze]

Timeout member button

[trim cedar]

ah lol

[flint bronze]

I was gonna put a quick timeout button below it

but you can press control to do the same thing

[trim cedar]

could still be useful to have a plugin to patch out the key requirement

in case you need it a lot

[flint bronze]

i mostly just want to lower the 60 seconds down to something like 15 seconds

i mean overhaul that UI that pops up with a slider too

[viral roost]

discord doesnt let you go below 60s normally right

passing a duration you cant set through the client could be a bit but kinda doubt it now that i think about it

[flint bronze]

that is true

[flint bronze]

remember that it is a timestamp that is sent, not a duration
that will have some amount of inaccuracy to the duration

I will probably put a disclaimer

[pure temple]

couldn't you change your time xd

wait no probably not

[flint bronze]

yes you could

[shadow ruin]

guh

right

[frigid bay]

uh

[shadow ruin]

/run c #include <stdio.h> int main() { char s1[] = "guh"; s1[0] = 'b'; printf(s1); }

[worldly oxide]

Here is your c(10.2.0) output @shadow ruin

buh```

[shadow ruin]

ok so

buh

/run c #include <stdio.h> int main() { char s1* = "guh"; s1[0] = "b"; printf(s1); }

[worldly oxide]

@shadow ruin I received c(10.2.0) compile errors

file0.code.c: In function 'main':
file0.code.c:3:10: error: expected '=', ',', ';', 'asm' or '__attribute__' before '*' token
    3 |   char s1* = "guh";
      |          ^
file0.code.c:3:12: error: expected expression before '=' token
    3 |   char s1* = "guh";
      |            ^
file0.code.c:4:3: error: 's1' undeclared (first use in this function)
    4 |   s1[0] = "b";
      |   ^~
file0.code.c:4:3: note: each undeclared identifier is reported only once for each function it appears in
chmod: cannot access 'a.out': No such file or directory
/piston/packages/gcc/10.2.0/run: line 6: ./a.out: No such file or directory

[shadow ruin]

ye

why

[frigid bay]

char *s1

not

char s1*

[shadow ruin]

guh

/run c #include <stdio.h> int main() { char *s1 = "guh"; s1[0] = "b"; printf(s1); }

[worldly oxide]

@shadow ruin I received c(10.2.0) compile errors

file0.code.c: In function 'main':
file0.code.c:4:9: warning: assignment to 'char' from 'char *' makes integer from pointer without a cast [-Wint-conversion]
    4 |   s1[0] = "b";
      |         ^
/piston/packages/gcc/10.2.0/run: line 6: 31408 Segmentation fault      (core dumped) ./a.out "$@"

[shadow ruin]

ye

this was my question

[frigid bay]

ur setting the 1st character to the LSB of the address of "b"

[shadow ruin]

please talk in stupid language

[frigid bay]

you code bad write

is error bad unexpect behavior

[mild ermine]

Do the css imports need to be profile.github.io or does github.com work?

[flint bronze]

as long as its the raw link it should work iirc

[viral roost]

profile.github.io or raw.githubusercontent.com

not github.com

(usually, unless the theme def specifies that's what you should use)

[mild ermine]

Yeah I meant raw but was too lazy to type it on phone

[viral roost]

fair

[flint bronze]

does it not follow redirects to the raw subdomain due to csp or something

[viral roost]

the issue is not sending proper content type but vencord fixes that

[mild ermine]

I don't think it matters if it's https or http

It's just it can't not include it

Like it has to have the https://www. as www. will use Vencord//www.

[trim cedar]

you try change character of word by setting character to word

this no work

[shadow ruin]

oh that was just a typo

what i was trying to do is

/run c #include <stdio.h> int main() { char *s1 = "guh"; s1[0] = 'b'; printf(s1); }

[worldly oxide]

@shadow ruin I only received c(10.2.0) error output

/piston/packages/gcc/10.2.0/run: line 6: 32607 Segmentation fault      (core dumped) ./a.out "$@"

[trim cedar]

the string may reside in read only memory because it's actually a constant

/run ```c
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

char* strdup(const char* src)
{
char* rtn = malloc(strlen(src) + 1);
strcpy(rtn, src);
return rtn;
}

int main() {
char *s1 = strdup("guh");
s1[0] = 'b';
printf(s1);
free(s1);
}```

[worldly oxide]

Here is your c(10.2.0) output @trim cedar

buh```

[trim cedar]

bruh where is strdup

[shadow ruin]

/run bash man strdup

[worldly oxide]

@shadow ruin I only received bash(5.2.0) error output

file0.code: line 1: man: command not found

[shadow ruin]

WHAT

[trim cedar]

???

ah

it's not a standard c func

[trim cedar]

@shadow ruin there you go

gnus libc should have strdup though

you're normally only allowed to treat strings defined in code (not dynamically allocated) as const, however I think C has extensions where it is allowed to not do that - however as you can see that's not a good idea

[dull magnet]

C polyfills wtf

is this why autoconf and shit exists

[trim cedar]

not really I think

[pure temple]

something i have concerns with with writing a discord bot in C++
is security 😭

it couldn't be that bad, could it?

i imagine use after free can create vulerabilities, that's the main thing i might end up doing

yeah i should just use the crap out of shared pointers and the code will be totally safe

i guess this is why rust has the lifetime stuff

my bot "idea" is esentially dyno/carl/mee6/whatever generic thing but good

[flint bronze]

https://tenor.com/view/buzz-lightyear-gif-21327280

[frigid bay]

theres a compiler thing to detect buffer overflows

[flint bronze]

jk all the existing ones suck

[pure temple]

and as you can imagine people give bots like that admin (for no reason)

[flint bronze]

"hello please put this bot at the top of the role list it will be fine"

[pure temple]

imo for bot programming you should just use std::array or whatever

[pure temple]

imagine finding a bug in the parser and finding a way to write a command and run machine code to give yourself admin

C++ experience

[pure temple]

and try to be strict about stuff like not using raw pointers

i guess sometimes you need them to avoid copying (they are more explicit than references so maybe better)

actually if you're just using a value and not storing it anywhere else references are fine...

introducing multithreading...!

[trim cedar]

if you don't fuck up really bad then no

won't be worse than with any other language

[pure temple]

can you give some examples of screwing up

because i've still only been writing cursed C++ for a few years...

[trim cedar]

using C

[pure temple]

LOL

[trim cedar]

jk