#archive-library-discussion | discord.js - imagine an app | Page 1

merry python Feb 26, 2020, 11:04 AM

#

this channel is for serious discussions only

#

please refrain from memes

rigid trail Feb 26, 2020, 11:49 AM

#

alright, lets christen the channel by bringing up an issue with websocket event handlers, the gist of it is there is no distinction between handlers and actions, some handlers just call actions, some call actions and have their own logic, etc. the full issue is right here https://github.com/discordjs/discord.js/issues/3790, my question is how would you implement a solution, considering some actions are also called internally by the lib too. (This will be a v13 thing).

fringe temple Feb 26, 2020, 2:20 PM

#

This channel is specifically for discussions about the development of the library Discord.js. That means you can't ask for help using Discord.js in this channel, and talking about moderation/server administration isn't appropriate here either. Talking about issues/pull requests or features is appropriate, however!

wild flax Feb 26, 2020, 2:56 PM

#

rigid trail alright, lets christen the channel by bringing up an issue with websocket event ...

vernal atlas Feb 26, 2020, 6:41 PM

#

https://github.com/discordjs/discord.js/blob/9cb306c823216130e5bddc268ba3aee025a5b98f/src/client/voice/dispatcher/StreamDispatcher.js#L72 is it about time this is removed? as its only there for backwards compatibility and the comment above says it will probably be removed?

#

also on that, should the finish listener be .once instead of .on?

slate nacelle Feb 26, 2020, 6:47 PM

#

I'd say yes to both

undone ravine Feb 26, 2020, 6:50 PM

#

i'd say no to once instead of on

#

finish is an event from WritableStream, and typically does indicate that there will never be any more data written to it... at which point the object is essentially being disposed of anyways

#

and if somebody somehow manages to kludge in a way to reuse the stream after that, we'd still want it to do the same stuff on finish

#

not at all a valid use-case, but the listener is going away with the object anyways

vernal atlas Feb 26, 2020, 6:53 PM

#

hmm, ok

#

https://github.com/discordjs/discord.js/blob/master/src/client/voice/dispatcher/StreamDispatcher.js#L81-L97 here i can see that on each of the this.streams.* error listeners, they call the streamError event, which then calls this.destroy, but after emitting the error event which then calls streamError again and calls destroy a second time, would there be a problem leaving it as is?

#

also, https://github.com/discordjs/discord.js/blob/9cb306c823216130e5bddc268ba3aee025a5b98f/src/client/voice/dispatcher/StreamDispatcher.js#L235 here done seems to be called without actually checking if it is a function or not (possibly resulting in an error as done won't be a function), if you look https://github.com/discordjs/discord.js/blob/9cb306c823216130e5bddc268ba3aee025a5b98f/src/client/voice/dispatcher/BroadcastDispatcher.js#L19 here you can see the method is called without a 3rd parameter, i'm not sure how the writable streams and whatnot work, so i'm not sure exactly how to fix it, i thought id point this out incase anyone wants to fix it

#

ill open an issue about it later

spice lava Feb 26, 2020, 7:14 PM

#

I am extremly unsure about the possibility and if this is something the devs would support, but it would be an idea to actually use the typings for typescript users in terms of enums. So I personally as an example just cant remember the zws code so one possibility would be to have Sign.ZWS or similar within typescript enums.
Whats do you think

copper laurel Feb 26, 2020, 7:26 PM

#

v13 is highly likely to at least partially be a TS rewrite, probably fully, so stuff like that can be considered. The ZWS isn't really discord functionality though, so in terms of it being part of the library core that's a bit questionable. At the very least I'd say it could be a snippet in the guide like the emoji numbers workaround.

#

Something people can easily implement in their own code as a util

zenith monolith Feb 26, 2020, 7:27 PM

#

Will there be a streamEnd reason in v12 or is there already that im not aware of?

warped crater Feb 26, 2020, 7:31 PM

#

Hello, considering that this channel now exists it seems pointless to ask this in #archive-offtopic and expect a quick response I will go on ahead with my thing

I am totally confused about the disableEveryone -> disableMentions change, why exactly was this done.

I totally understand the purpose of disableMentions, but I think a very large portion of us only want to disable <@&265426791538229258> & <@&265427044882579456> to completely nullify any possible oopsies.
Should also be noted that disabling ALL mentions for us is inane and is not something we can easily do.

vernal atlas Feb 26, 2020, 7:32 PM

#

https://github.com/discordjs/discord.js/blob/9cb306c823216130e5bddc268ba3aee025a5b98f/src/client/voice/networking/VoiceWebSocket.js#L4 so apperently you don't have to use the EventEmitter property from the events module (bad wording but, you can do const EventEmitter = require("events") instead of const { EventEmitter } = require("events"), should i change this throughout the lib to use the first way or the second, as both ways seem to be used throughout (im doing a PR to fix little things like this)

wild flax Feb 26, 2020, 7:32 PM

#

No, we had a PR changing it to be the destructured one

#

The former is just a global namespace

vernal atlas Feb 26, 2020, 7:33 PM

#

aight, so ill make everything use the destructured?

wild flax Feb 26, 2020, 7:33 PM

#

Its not entirely consistent, but it doesnt have to be

#

But I guess

#

its a non-trivial change

#

Actually

#

It should be consistent already

#

https://github.com/discordjs/discord.js/commit/7b442623462f7718f65fc9f30cd2a3fb3463caa3 relevant PR

vernal atlas Feb 26, 2020, 7:37 PM

#

in some places it does const EventEmitter = ... and some places const { EventEmitter } = ...

#

so not consistent it seems?

fringe temple Feb 26, 2020, 8:03 PM

#

In response to streamEnd, no there will not, you should use the debug event to listen for related info instead

vernal atlas Feb 26, 2020, 8:19 PM

#

in some files, the EventEmitter.off method is used over EventEmitter.removeListener (which they are the exact same method it seems), for consistency, which one would be better?

#

pretty small change i know thinkLul , just doing a PR to improve consistency in certian places and figured i may as well include everything

frozen current Feb 27, 2020, 12:55 AM

#

Djs uses on to add listeners so I guess off to remove them would make sense. My personal opinions is addListener and removeListener would look better as it’s more descriptive if you wanna go around and change everything mmLol

#

@vernal atlas

vernal atlas Feb 27, 2020, 1:08 AM

#

hm

rigid trail Feb 27, 2020, 1:13 AM

#

I disagree, in the end its a lot of refactoring for really an extremely minor change, and chances are, if you are looking at source code you already know what .on and .off do

vernal atlas Feb 27, 2020, 1:15 AM

#

yea

copper laurel Feb 27, 2020, 3:45 AM

#

imo looking at the websocket event handlers and actions, there's definitely cleanup potential. There may still be a need to differentiate between handling when the event is from a websocket, and handling when its a lib call/rest related, but that can easily be done by a single class that has different methods. No need to separate the two

rigid trail Feb 27, 2020, 4:13 AM

#

imo a class with one or more methods sounds good

#

although there might be a way to just merge into one method and if possible have something resolve if its a raw packet or not

#

@copper laurel ^

copper laurel Feb 27, 2020, 4:34 AM

#

Maybe, but I don't think that improves maintability over separate methods

real jetty Feb 27, 2020, 1:20 PM

#

ive been working on local changes ages ago for rewriting the logic for actions, i pushed 2 days ago to clean my working tree for rebasing some other pr. so here's some of my insight about this..
u can easily distinguish between whether the rest called the handler or whether it was the ws itself via a simple param (https://github.com/izexi/discord.js/blob/packet-handler/src/client/websocket/handlers/index.js#L42 3rd param is set to true over on WebSocketManager#handlePacket()). the actual issue is with handling overlap from the ws & rest, theres 2 ways to get around this: either don't call the handler at all for rest calls and return a patched structure or call the handler and add some logic within the handler to ensure the same function doesn't get run twice which in most cases turns out in unexpected behavior (which is pretty much for the *_CREATE/UPDATE/DELETE events - https://github.com/izexi/discord.js/blob/packet-handler/src/client/websocket/handlers/index.js#L4-L18). the 1st solution is somewhat hacky and is currently used in some places atm (e.g: when GuildChannel#delete() is called it returns the channel itself which isn't always guaranteed to be .deleted or removed from the ChannelStore) so i went with the 2nd and simply just stored the structure that has been "handled" and then return that when the handler gets called again from ws/rest (note that the order isn't always consistent) similarly to how its currently done on the current GuildDelete action.
another major concern is the stability of all this, hence y i started off by setting out some tests for all this https://github.com/izexi/discord.js/blob/packet-handler/test/packetHandler.test.js
im planning to allow more flexibility for the PacketHandler class so that it can be interacted with externally allowing developers to inject their own handlers, so thats a todo along with some cleanup which i will prolly finish after my exams

vernal atlas Feb 27, 2020, 6:28 PM

#

https://github.com/discordjs/discord.js/blob/653784b56445b015a1584de1b693e57fc9069050/src/sharding/ShardingManager.js#L168 eslint keeps telling me this is a race condition Thonk

#

im assuming it must have passed that check when that change was merged, so what's changed there

#

📎 unknown.png

#

looking at git blame, it seems it didn't pass the checks (its quite old so its not available anymore to see exactly whats wrong) but shouldn't it have fired off errors in other PRs in that file?

vernal atlas Feb 27, 2020, 6:48 PM

#

theres another one, https://github.com/discordjs/discord.js/blob/653784b56445b015a1584de1b693e57fc9069050/src/structures/GuildMember.js#L308 very interesting

worthy coral Feb 28, 2020, 3:42 AM

#

Was looking at commits to upgrade my client to and I noticed that as of 9cb306c8, disableEveryone was replaced with disableMentions.
https://github.com/discordjs/discord.js/blob/653784b56445b015a1584de1b693e57fc9069050/src/structures/APIMessage.js#L95
For some clients, this functionality might be intended and helpful and I understand that this fixes exploitation of clients using the disableEveryone option but still being able to mention everyone and here via RTL override char but completely replacing all at symbols which might contain mentions to users and the developer having the intent to mention a user but not be able to mention everyone and here is very limiting. I do, however, like the idea of it being a client option.
I do have a proposed solution to make the two work in tandem which would be to either reintroduce the option itself or change ClientOptions#disableMentions to be of type "all" | "everyone"

vernal atlas Feb 28, 2020, 3:51 AM

#

that sounds alright, but it can't have the same implentation as before

#

as you would have to find every character discord strips and strip it

#

the whole reason of disableMentions was because disableEveryone (or as you proposed above, disableMentions set to "everyone") was unreliable

worthy coral Feb 28, 2020, 3:54 AM

#

Like I said, I’m aware it fixes the RTLO char issue but if someone were to reimplement it, they could just strip the RTLO char after at symbols

vernal atlas Feb 28, 2020, 3:55 AM

#

yeah but its not only that character

#

from what i gather, there are quite a few characters that discord strip from the messge which aren't all documented

#

making it unreliable

worthy coral Feb 28, 2020, 3:56 AM

#

The way the current implementation does it is by adding a ZWSP after all at symbols which does work so that implementation would be a viable fix

vernal atlas Feb 28, 2020, 3:56 AM

#

yeah exactly

worthy coral Feb 28, 2020, 3:58 AM

#

I see where you’re getting at now. Regular expressions

#

to which I say

📎 unknown.png

vernal atlas Feb 28, 2020, 4:11 AM

#

$>js const regex = /@.+(everyone)/; const text = "@vernal atlas say hello to everyone"; text.replace(regex, "@\u200b$1");

pearl glacierBOT Feb 28, 2020, 4:11 AM

#

<@everyone

vernal atlas Feb 28, 2020, 4:12 AM

#

needs a bit of work kappapuff

#

maybe test a regex on a string like that first

wild flax Feb 28, 2020, 5:54 AM

#

Yeah I have no intentions of accepting something that’s only working 50 percent of the time

#

Neither am I okay with adding a disclaimer like “this only works as long as someone doesn’t want to fuck you over” lol

worthy coral Feb 28, 2020, 6:20 AM

#

Working on it

worthy coral Feb 28, 2020, 7:34 AM

#

Ended up making it more reliable and be future proof if any other unicode characters are discovered to become a vulnerability

Relative regular expression if you want to test it. If anything breaks the intended functionality, let me know

/@([^<> ]*?(?:everyone|here))/g

Below are tests of both methods which I modified the source to allow either method to be run.
Below the output is how the client would look like when it's constructed and also an image of the test source
https://cdn.discordapp.com/attachments/682818889993814078/682851488552321070/unknown.png
https://cdn.discordapp.com/attachments/682818889993814078/682851620488478721/unknown.png
https://cdn.discordapp.com/attachments/682818889993814078/682851723756568618/unknown.png
https://cdn.discordapp.com/attachments/682818889993814078/682851784099758101/unknown.png
https://cdn.discordapp.com/attachments/682818889993814078/682852007744110614/unknown.png

#

Yikes. My client appears to be broken rendering the embedded images

#

The "all" disableMentions uses the method in the latest commit

#

@vernal atlas @wild flax if you two were curious ^

grave pollen Feb 28, 2020, 8:05 AM

#

disableMentions: "false",?

worthy coral Feb 28, 2020, 8:05 AM

#

?

wild flax Feb 28, 2020, 8:41 AM

#

Is this including the test cases of the original PR?

worthy coral Feb 28, 2020, 8:42 AM

#

No. I can test those as well really quickly

#

Test 2 failed. I'll look for a fix

worthy coral Feb 28, 2020, 9:21 AM

#

Ended up having to change the regular expression to

/@([^<>@ ]*?(?:everyone|here))/gsmu

And all of the tests passed. Both disabling everyone/here and all mentions ended up producing the exact same output

📎 unknown.png

worthy coral Feb 28, 2020, 9:53 AM

#

While all of the tests did pass, I found another issue which was to put the RTLO anywhere inside of the everyone or here which I am going to look for a fix now

ornate topaz Feb 28, 2020, 11:54 AM

#

at this point i'm going to again suggest doing this from the other side

#

instead of assuming every string is safe and try to search for word everyone in it to put zws, treat every @ character as unsafe unless you can use it to form a valid user mention

#

makes it a bit easier, as user/role mentions cannot have letters, so pretty much any try to make everyone ping disguised as user mention will fail, as basic /<@!?\d{17,19}>/ already fails

wild flax Feb 28, 2020, 11:57 AM

#

how does that play out with the guild id

#

just an additional check?

ornate topaz Feb 28, 2020, 11:58 AM

#

is it possioble to mention everyone role with guild id?

wild flax Feb 28, 2020, 11:58 AM

#

I mean that is the everyone role

#

the guild id

ornate topaz Feb 28, 2020, 11:59 AM

#

if it is possibel to perform actual everyone ping using guild id and role mention, probably check would be needed

#

if it doesn't work as actual everyone ping, there is no point in checking for guild id being used

#

i think i posted screenshot in #archive-offtopic

wild flax Feb 28, 2020, 12:00 PM

#

ah its actually producing @@everyone

#

strange

#

well why dont you just PR it then

#

here it will just get lost

#

if you have a working implementation

ornate topaz Feb 28, 2020, 12:01 PM

#

okay, gonna sit at it in an hour or so

worthy coral Feb 28, 2020, 12:53 PM

#

I believe I got a good enough implementation, now

#

https://cdn.discordapp.com/attachments/682818889993814078/682933842956255237/unknown.png
Good thing replace can accept a function

vernal atlas Feb 28, 2020, 1:40 PM

#

i don't think that actually mentions people tho

ornate topaz Feb 28, 2020, 1:44 PM

#

looking into source code (and on screenshots) it should

#

it doesn't put zws if this regex matches, at least that's what i get from the PR

📎 unknown.png

worthy coral Feb 28, 2020, 1:45 PM

#

I think they were referring to @<@&265426791538229258>

ornate topaz Feb 28, 2020, 1:45 PM

#

https://canary.discordapp.com/channels/222078108977594368/682166281826598932/682919870433591297

wild flax Feb 28, 2020, 1:46 PM

#

yeah it doesnt work

#

📎 unknown.png

#

normal, escaped, role, user (guildid), user (guildid)

warped crater Feb 28, 2020, 2:37 PM

#

tried the escape too yesterday when I was tryna find a workaround

#

I ended up just underlining the @, its subtle; not ideal but it does what I need it to

vernal atlas Feb 28, 2020, 4:58 PM

#

should something be done about these race conditions i mentioned https://canary.discordapp.com/channels/222078108977594368/682166281826598932/682655314193285170

wild flax Feb 28, 2020, 5:01 PM

#

that jump link somehow does not work for me

#

monkaS

frozen current Feb 28, 2020, 5:01 PM

#

https://github.com/discordjs/discord.js/blob/653784b56445b015a1584de1b693e57fc9069050/src/sharding/ShardingManager.js#L168

GitHub

discordjs/discord.js

A powerful JavaScript library for interacting with the Discord API - discordjs/discord.js

vernal atlas Feb 28, 2020, 5:10 PM

#

interesting

#

i can't even scroll up above that point

📎 unknown.png

vernal atlas Feb 28, 2020, 5:39 PM

#

https://github.com/discordjs/discord.js/blob/3d0c1df19d66bbb83e2fe73a9377bd6a8428f515/src/structures/GuildChannel.js#L207 should i make a PR to change this to accept the overwrites as the 1st parameter, and the reason as the 2nd? since as it stands now, this is just an alias to GuildChannel.edit, and, at least to me, it makes more sense to just allow the overwrites and reason

spice lava Feb 28, 2020, 5:40 PM

#

Yeah seems actually smart to do that

snow crypt Feb 28, 2020, 8:05 PM

#

What if Message#awaitReactions and TextBasedChannel#awaitMessages moved to the managers?
Like message.reactions.await and channel.messages.await

#

same for collectors, message.reactions.createCollector

rigid trail Feb 28, 2020, 8:06 PM

#

Personally i dont believe that naming it .await is a good naming scheme

#

You also run into the issue that its not for the purpose of a manager

#

await.. does not make api calls, therefore its out of the scope of the manager

snow crypt Feb 28, 2020, 8:07 PM

#

maybe it should be?

rigid trail Feb 28, 2020, 8:07 PM

#

That would completely bork the purpose of a manager

snow crypt Feb 28, 2020, 8:07 PM

#

yes, the purpose of managers (originally) was to separate api from cache

rigid trail Feb 28, 2020, 8:07 PM

#

And it still will be

#

The whole purpose of a manager is to be able to distinguish what uses the api

#

If you start adding these things it completely clouds the purpose of it

snow crypt Feb 28, 2020, 8:09 PM

#

i dont know, kinda makes sense for everything reaction or message to be in one place

#

for the same reason datastores were originally introduced

burnt cradle Feb 28, 2020, 8:09 PM

#

And the same reason stores were removed

rigid trail Feb 28, 2020, 8:09 PM

#

exactly

#

Honestly, if things like these are allowed to be implemented, managers will become the same crapshoot that datastores where ¯_(ツ)_/¯

#

plus think about it, does it interact at all with the cache of messages? Nope, does it interact with the api to do something to the cache? Nope

vernal atlas Feb 28, 2020, 8:46 PM

#

yeah i agree it should stay as is

vernal atlas Feb 28, 2020, 9:55 PM

#

https://github.com/discordjs/discord.js/blob/master/typings/index.d.ts#L2619-L2630 shouldn't it be defined somewhere in there what properties exactly are missing when something is partialized, i believe the [K in keyof Omit<T, 'id' | 'partial'>]: T[K] | null; is meant to do that but it doesn't seem to, typescript should be throwing an error here because content could be null?

📎 unknown.png

#

and also, wouldn't [K in keyof Omit<T, 'id' | 'partial'>]: T[K] | null; mark things such ascreatedAt as possibly being null, when infact they can't

zenith monolith Feb 29, 2020, 10:06 AM

#

Will there be a dispatcher end reason in master?

snow crypt Feb 29, 2020, 4:15 PM

#

What is reply-prefix for? Also, if it is needed in the main lib, why isn't it merged into master?

ornate topaz Feb 29, 2020, 4:16 PM

#

https://github.com/discordjs/discord.js/pull/2910, from the button directly next to branch selector

analog oyster Feb 29, 2020, 4:50 PM

#

~~I don't think this is showing as intended https://i.imgur.com/yUYhInu.png https://i.imgur.com/qUOSHM8.png~~

umbral aspen Feb 29, 2020, 4:51 PM

#

What do you mean? That looks fine

analog oyster Feb 29, 2020, 4:52 PM

#

'none' vs none'

#

~~just noticed I came to the wrong channel~~

ornate root Feb 29, 2020, 5:31 PM

#

I think this should be deprecated
https://discord.js.org/#/docs/main/11.6.0/class/Client?scrollTo=e-userNoteUpdate

#

since only userbots update a note

ornate topaz Feb 29, 2020, 5:36 PM

#

none of 3 user-account-only events have a note about deprecation, my guess is that there is nothing emitted for them if you have bot account. emitting them doesn't make much sense either.
additionally, all are removed on master, along with all user-account-only methods

slate nacelle Feb 29, 2020, 5:53 PM

#

I think it's too late for that now

snow crypt Feb 29, 2020, 5:55 PM

#

not sure the docgen will do something about @depracated on @events

vernal atlas Feb 29, 2020, 6:12 PM

#

i seem to be having an issue with the VoiceChannel.join (ClientVoiceManager.joinChannel) not resolving, can anyone else reproduce?

📎 unknown.png

#

can't recall exactly what commit im using, but i updated a couple nights ago and i don't think there has been any changes to the voice since then

wild flax Feb 29, 2020, 6:13 PM

#

thats an ever lasting problem

ornate topaz Feb 29, 2020, 6:13 PM

#

(there was one literally minutes ago)

wild flax Feb 29, 2020, 6:13 PM

#

and the one minutes ago yeah

vernal atlas Feb 29, 2020, 6:17 PM

#

ah

vernal atlas Feb 29, 2020, 7:05 PM

#

anyway, https://discordapp.com/channels/222078108977594368/682166281826598932/683069729912389675

real jetty Mar 1, 2020, 1:44 AM

#

Question about 11.6.0, why would bots be able to recieve urgent system messages?

#

According to the help article (https://support.discordapp.com/hc/en-us/articles/360036118732-Discord-System-Messages), the only reasons why an urgent message would be sent is for user-related issues such as payment problems, support ticket updates, and ToS violations

Discord

Discord System Messages

You got mail! ✉️
What are Discord System Messages?
Discord System Messages are official messages that we send to alert you of an account issue. These Messages can range from a variety of Payment is...

#

Also, why are message flags applicable to Message#edit?

#

Another question, GuildMember#manageable seems poorly documented. I dont understand what would constitute a member being "manageable"
maybe an example would help

📎 unknown.png

unique axle Mar 1, 2020, 1:58 AM

#

for that last one: i think that message is pretty clear if you know how discord works
role hierarchy dictates if you are able to action a member - be that nickname them, kick/ban them etc.
if the highest role of the target (instance this is called on) is higher or equal than the logged in users (bots) highest role they are not actionable/manageable

real jetty Mar 1, 2020, 1:59 AM

#

So, manageable then becomes a shortcut for
targetMember.highestRole.position > message.guild.me.highestRole.position

undone ravine Mar 1, 2020, 2:00 AM

#

and an owner check

unique axle Mar 1, 2020, 2:00 AM

#

Also, why are message flags applicable to Message#edit?
do you mean not applicable to Message#edit?
flags are not applied via a patch request, that Message#edit constitutes

undone ravine Mar 1, 2020, 2:00 AM

#

i do agree that the description for manageable could be improved

unique axle Mar 1, 2020, 2:01 AM

#

any suggestion then?

undone ravine Mar 1, 2020, 2:01 AM

#

we shouldn't make the assumption that users of the lib are 100% familiar with the ins and outs of discord

unique axle Mar 1, 2020, 2:01 AM

#

you don't really need to be with the description as-is, it really is just that, manageable in terms of role hierarchy
what exactly that is can be as much of a blackbox as you want imho

real jetty Mar 1, 2020, 2:02 AM

#

No, @unique axle I mean applicable.
If you follow the documentation from Message#edit, it accepts two parameters: content, and MessageEditOptions.
MessageEditOptions includes the ability to set flags

undone ravine Mar 1, 2020, 2:02 AM

#

probably just an addition to it elaborating on what "manageable" means

real jetty Mar 1, 2020, 2:02 AM

#

📎 unknown.png

#

This is why I'm confused, one would think that's read only. And even if, n/a

unique axle Mar 1, 2020, 2:04 AM

#

oh, that's stable, i was looking at master ablobcatsipsweats
it is the interface embed supression is based on
https://github.com/discordjs/discord.js/blob/stable/src/structures/Message.js#L567

real jetty Mar 1, 2020, 2:05 AM

#

Ok, so then, why include all of the flags when it seems that only embed suppression can be applied?

#

To account for those that use bitwise operators? Or bitfields? or..

unique axle Mar 1, 2020, 2:06 AM

#

why provide that interface at all, when on master it isn't public facing and you can just call Message#suppress... anyways
is the question i'm having now, probably an oversight potatodetective

real jetty Mar 1, 2020, 2:07 AM

#

Interesting.
Seems then it's been "fixed" in master, so I wont bother with this any longer

unique axle Mar 1, 2020, 3:33 AM

#

So in summary:
Q: Why is urgent system message in MessageFlags
A: It's documented in the public API documentation https://discordapp.com/developers/docs/resources/channel#message-object-message-flags
WIth discord.js trying to be API complete it should be compatible with it.

Q: GuildMember#manageable is unclear
A: Thanks for the suggestion, you can track its status here: https://github.com/discordjs/discord.js/pull/3870

Q: Flags being part of Message#edit despite their informational nature:
A: V12 removes this behavior and just allows Message#suppressEmbeds to be called without direct ability to pass flags into Message#edit

clever crypt Mar 1, 2020, 7:50 PM

#

I wanted to go about editing all appropriate tags for the release of v12 but I'm semi-confused by which version is which. Additionally, I don't have permissions to edit tags which aren't mine. So below is a list of all tags that need changing.

~~documentation~~,~~ stable~~, ~~master~~, ~~11.5-dev~~

feel free to ping me in #archive-offtopic if more need to be added

edit: thanks space!

slate nacelle Mar 1, 2020, 7:56 PM

#

Should be updated (or deleted in case of 11.5-dev) blobthumbsup

tawny warren Mar 1, 2020, 8:55 PM

#

other than the cache and addBlankField are there any major changes in v12 if i was already using it before now?

clever crypt Mar 1, 2020, 8:55 PM

#

All the changes are listed at https://discordjs.guide/additional-info/changes-in-v12.html

Discord.js Guide

A guide made by the community of discord.js for its users.

wide palm Mar 1, 2020, 11:09 PM

#

The v12 update announcement said that node needed to be updated to v12, but the update guide has said (up til and including now) that node would need to be updated to v10.2
Which is correct?

rigid trail Mar 1, 2020, 11:10 PM

#

v12 is correct, guide will be updated

wide palm Mar 1, 2020, 11:10 PM

#

Ok, thank you!

white grove Mar 2, 2020, 8:03 PM

#

Is there any big reason to upgrade to djs v12 if you don't specifically need any of the new features?

warped crater Mar 2, 2020, 8:35 PM

#

yes

#

v11 is going to become deprecated & eventually EOL

#

unless its a huge hassle you should rewrite when you can yeptune

white grove Mar 2, 2020, 9:24 PM

#

Was afraid of that, thanks

spice lava Mar 2, 2020, 10:59 PM

#

to be fair the rewrite is not as big as I thought too. I changed not listening to raw event now and using partials and just changing shit around to use caches etc. Took me less than 2 hours for the whole project (which included unit test adjustment)

zinc folio Mar 3, 2020, 6:05 AM

#

yeah, the effort to rewrite code with this update was thankfully minimal and painless, thanks djsdevs 👍

opaque vessel Mar 3, 2020, 12:18 PM

#

This may be some sort of minor feature request, but can .displayAvatarURL() use .png as a default instead of .webp? On iOS, we can't see it because apparently Discord doesn't support it? I have to specify that option every time as well as take into consideration animated profile pictures because of it for iOS users

📎 image0.png

small flame Mar 3, 2020, 12:24 PM

#

you can do that with the dynamic property

#

you set a default format with the format option and then set dynamic to true so it defaults to gifs for animated one's

#

but for the library i would rather tend to say no since that brings opinions into it, you can ofc create an default
options object and reuse that through your application, which i would recommend

opaque vessel Mar 3, 2020, 12:32 PM

#

I don't understand what you mean, the first half is what I am already doing because of it but right now discord.js defaults to using .webp?

small flame Mar 3, 2020, 12:33 PM

#

yes and as i said you should continue doing that and d.js should not change it default behavior because of an bug on ios since thats opinionated

rigid trail Mar 3, 2020, 12:33 PM

#

I disagree with the point of “that brings opinions into it”, while its true, the opinion of most people is png is generally favored over webp, so this would be a nice QOL change

opaque vessel Mar 3, 2020, 12:33 PM

#

It's not a bug, it's just unsupported

small flame Mar 3, 2020, 12:34 PM

#

the opinion of most people is png is generally favored over webp this is not at all true, you should not assume what people think imo

wild flax Mar 3, 2020, 12:34 PM

#

iOS supports webP just fine, its discord on iOS

opaque vessel Mar 3, 2020, 12:34 PM

#

Yeah, that

small flame Mar 3, 2020, 12:34 PM

#

it is a bug with discord

#

which djs shouldn't monkey patch imo

opaque vessel Mar 3, 2020, 12:34 PM

#

But the problem is that if this was to truly work for everyone, for now it would be better using .png? Cause everywhere in my code I always have to specify those options

small flame Mar 3, 2020, 12:35 PM

#

you can write an getter which sets the default options, you can extend the structure for that

wild flax Mar 3, 2020, 12:35 PM

#

We can't just change it back and forth all the time

#

So for now is a bad metric

opaque vessel Mar 3, 2020, 12:35 PM

#

I don't see why you'd need to change it back

wild flax Mar 3, 2020, 12:36 PM

#

Tinier filesize, less bandwidth

opaque vessel Mar 3, 2020, 12:36 PM

#

Oh I see... well thank you for replying everyone

small flame Mar 3, 2020, 12:36 PM

#

i dont see why d.js would need to change it in the first place when you clearly can just make your own options and extend the Structure with a custom method to use your own "default" options Eyes

opaque vessel Mar 3, 2020, 12:37 PM

#

Sounds unnecessary

wild flax Mar 3, 2020, 12:37 PM

#

Thats an opinion

small flame Mar 3, 2020, 12:40 PM

#

well it might be unnecessary but it solves your problem you mentioned Cause everywhere in my code I always have to specify those options

#

with a custom method which uses these options you dont anymore

zealous parrot Mar 4, 2020, 4:22 AM

#

Why did you guys have to break discord.js like that

#

Like bruh

rigid trail Mar 4, 2020, 4:23 AM

#

offtopic

zealous parrot Mar 4, 2020, 4:23 AM

#

How

rigid trail Mar 4, 2020, 4:23 AM

#

this is about serious discussion, not just "wtf bruh"

#

if you want to discuss it seriously bring up serious points about why its bad

zealous parrot Mar 4, 2020, 4:24 AM

#

Yes discord.js is broke, that is my response. It just doesn’t work, I’ve heard many people say that the devs changed the way that data is stored

#

And it caused it to not work

rigid trail Mar 4, 2020, 4:24 AM

#

yep, managers

#

per semver, 11 => 12 breaks code

opal mauve Mar 4, 2020, 4:25 AM

#

there's an extensive article on the official guide about migrating. if you're frustrated with the changes, I suggest you take a look

real jetty Mar 4, 2020, 5:35 AM

#

@rigid trail, that's why I already use master version

tired kindle Mar 4, 2020, 7:19 AM

#

I think when you major version it's intended to specify the library naming or functionality changed in a backwards incompatible way

snow crypt Mar 4, 2020, 10:32 AM

#

literally impossible in this case

#

managers, fine - just add collection methods that call the ones on their cache

#

but url functions like iconURL would be impossible because it would have to be callable AND a string

#

actually, it looks possible https://discordapp.com/channels/222078108977594368/222086648706498562/684711852193611833

#

but not ideal

#

and really, just replace anything that ends with URL with ....URL()

#

anyway, i migrated to master from stable when v12 was master, and it was not that difficult

#

wow

unique axle Mar 4, 2020, 11:06 AM

#

pinmorning Regarding breaking changes in version 12 and updating:

We have an extensive guide on updating at https://discordjs.guide/additional-info/changes-in-v12.html?v=12
Alternatively you can revert the update by changing the entry in your package.json to ^11.6.1 (the last version of v11) and running npm update discord.js
Or un- and re-install it with npm rm discord.js, npm i discord.js@11.6.1

Note: Major versions are prone to breaking changes as per semver: https://semver.org/#spec-item-8
Note: The ^ here signifies "compatible with" https://docs.npmjs.com/files/package.json and prevents breaking updates from being applied accidentally

small flame Mar 4, 2020, 12:28 PM

#

@unique axle do me a favor and pin that

vagrant holly Mar 4, 2020, 12:32 PM

#

👋 Regarding https://github.com/discordjs/discord.js/issues/3892 - what should be the approach here?
Should a PR be made to d.js to gracefully handle empty names/fields, or should this be reported to Discord directly as their API doing something wrong?
(Currently causing my bot to boot loop and getting banned from the API due to https://github.com/discordjs/discord.js/issues/3890)

small flame Mar 4, 2020, 12:33 PM

#

funny enough this seems to be only broken on v12 after the embed change, the link (https://www.deadstock.ca/products/rototo-loose-pile-socks-yellow) works fine on some commits before that change (on my prod bot)

#

i would say d.js should not validate the fields when receiving at all should it?

#

i dont really see a reason since it comes from discord AreYouSweating

vagrant holly Mar 4, 2020, 12:36 PM

#

Right, makes sense. MessageEmbed is also what's used by library users to make embeds though, right?
So I guess a flag would need to be added to the constructor to indicate it should skip validation?

small flame Mar 4, 2020, 12:37 PM

#

yea i think so, discord might do stuff which breaks d.js internal checks at any time so i dont think checking against received embeds makes much sense,

#

i guess a flag would work

vagrant holly Mar 4, 2020, 12:56 PM

#

What's super annoying is I can't reproduce this locally

small flame Mar 4, 2020, 12:57 PM

#

hmm?

vagrant holly Mar 4, 2020, 12:57 PM

#

Posting that link with a test bot on 12.0.1 doesn't seem to crash it

small flame Mar 4, 2020, 12:57 PM

#

oh

#

wait can you check the embed type for me

#

is it rich or link

vagrant holly Mar 4, 2020, 12:57 PM

#

What's the best way to check that

small flame Mar 4, 2020, 12:58 PM

#

just check the embed type in the embed of the message you receive

vagrant holly Mar 4, 2020, 12:59 PM

#

article

small flame Mar 4, 2020, 1:01 PM

#

oh?

vagrant holly Mar 4, 2020, 1:01 PM

#

Yup

small flame Mar 4, 2020, 1:01 PM

#

the embed you logged had the type rich tho

vagrant holly Mar 4, 2020, 1:01 PM

#

Indeed it does

small flame Mar 4, 2020, 1:01 PM

#

thats interresting

snow crypt Mar 4, 2020, 1:01 PM

#

it is rich

#

i pointed out in genral

vagrant holly Mar 4, 2020, 1:01 PM

#

I've no idea where the message is, somewhere in one of the 4k servers the bot that's crashing is

snow crypt Mar 4, 2020, 1:01 PM

#

check this https://discordapp.com/channels/222078108977594368/222197033908436994/684743996957917367

small flame Mar 4, 2020, 1:02 PM

#

yea but we should still not validate fields when receiving

vagrant holly Mar 4, 2020, 1:02 PM

#

Yeah, I'm about to raise the PR for that

small flame Mar 4, 2020, 1:02 PM

#

since as you just saw it might have different specs

#

yes 👍

vagrant holly Mar 4, 2020, 1:02 PM

#

Just can't confirm it's actually fixed that issue

small flame Mar 4, 2020, 1:02 PM

#

well

#

it should since that error comes from d.js field checking

heavy dove Mar 4, 2020, 1:54 PM

#

Anyone using vscode getting errors after installing v12

small flame Mar 4, 2020, 1:54 PM

#

vscode has nothing to do with the library

vernal atlas Mar 4, 2020, 5:33 PM

#

https://github.com/discordjs/discord.js/blob/bfcd4befc9a6d56c4c7db5c736bd7f31524f097c/src/structures/MessageEmbed.js#L199 isn't that technically inaccurate? because you can set an embed timestamp to any date

slate nacelle Mar 4, 2020, 5:37 PM

#

I'd say so, yes.

vernal atlas Mar 4, 2020, 5:43 PM

#

what would be a better description for it? having trouble coming up with one

vagrant holly Mar 4, 2020, 5:44 PM

#

The date displayed on the embed

#

?

slate nacelle Mar 4, 2020, 5:45 PM

#

Something like that, maybe?

Now that I see it, the name createdAt seems to be a bit of a misnomer. too.
Just date would probably have been better, but it's too late for that now.

vernal atlas Mar 4, 2020, 5:45 PM

#

yea

#

Just date would probably have been better, but it's too late for that now.

should i make a PR to just change the description then?

slate nacelle Mar 4, 2020, 5:49 PM

#

Yep

vagrant holly Mar 4, 2020, 5:56 PM

#

@slate nacelle You're SpaceEEC on GH, right?

#

So, we're removing the whole docstrings for constructor & setup due to skipValidation being a private param

#

However, normalizeField and _normalizeFields also have that param

#

But we really don't want to remove those docstrings

slate nacelle Mar 4, 2020, 5:58 PM

#

We do not need those changes anymore.

vagrant holly Mar 4, 2020, 5:59 PM

#

I replied to your review on that bit, I don't think it's sensible to skip normalizeFields completely

copper laurel Mar 4, 2020, 7:43 PM

#

I'm pretty sure this issue comes from my last PR on this where I introduced normalizeFields into the constructor

#

So my bad on that - it was intended to address the fact that people would pass in partial fields via the constructor data, but not fail until they werent sent to the API

#

However that's clearly impacted the incoming embeds so reverting it seems fine

silk gust Mar 5, 2020, 5:06 PM

#

can someone make a pr about the default avatars
user.avatarURL() =? nothing => Default Avatar URL

opaque vessel Mar 5, 2020, 5:08 PM

#

Use .displayAvatarURL()

tough geode Mar 5, 2020, 5:11 PM

#

I'm pretty sure it's like as it is on a purpose, it might be useful to know if someone has the default avatar or a set one. and ^, just use User#displayAvatarURL

drifting knot Mar 5, 2020, 6:52 PM

#

is this a place to discuss d.js-next

#

if so, i have bunch of minor improvements to stuff like CI on it, should i make a single big PR for it or many PRs that only change a single thing?

wild flax Mar 5, 2020, 7:05 PM

#

latter

drifting knot Mar 5, 2020, 7:05 PM

#

cool

#

pr spam time

#

also ur eslint GH action is using src dir by default and borking the ci

wild flax Mar 5, 2020, 7:06 PM

#

it won't use the action anymore

drifting knot Mar 5, 2020, 7:06 PM

#

ok

#

relevant lerna docs for linting many packages https://github.com/lerna/lerna/blob/master/FAQ.md#ci-setup

GitHub

lerna/lerna

:dragon: A tool for managing JavaScript projects with multiple packages. - lerna/lerna

wild flax Mar 5, 2020, 7:06 PM

#

yes I am aware

drifting knot Mar 5, 2020, 7:06 PM

#

cool

wild flax Mar 5, 2020, 7:06 PM

#

but we don't have any code yet, so its not really relevant

drifting knot Mar 5, 2020, 7:06 PM

#

got it

wild flax Mar 5, 2020, 7:17 PM

#

i mean you can fix it if you want @drifting knot

drifting knot Mar 5, 2020, 7:18 PM

#

i can try later

#

is it fine if the script is just raw ass eslint <files> and no pretty github actions integration

wild flax Mar 5, 2020, 7:21 PM

#

yup

#

that was the plan anyway

drifting knot Mar 5, 2020, 7:26 PM

#

cool

analog oyster Mar 5, 2020, 8:43 PM

#

Crawl, .eslintrc.json could be removed from the file root and added to package.json as eslintConfig then

wild flax Mar 5, 2020, 8:44 PM

#

no because it defines too many rules

#

we dont have our own shared config

#

and well likely not make one with the current rules

analog oyster Mar 5, 2020, 8:44 PM

#

ok

drifting knot Mar 5, 2020, 11:27 PM

#

@wild flax does it even make sense to have eslint running in the cron job?

#

also what is the plan for eslint config, make our own or use a premade one? (i like the XO ts config)

real jetty Mar 6, 2020, 12:23 AM

#

not 100% sure if this fits here correctly:
Is there seriously any disadvantage of removing all users from cache? If someone messages the bot or writes a message in the channel, the bot receives the user info so it works without caching somehow.

So why not simply "uncaching" all users?

sick raft Mar 6, 2020, 1:03 AM

#

var emb = new Discord.MessageEmbed()
        .setColor("#ff0d00")
        .setTitle("Title")
        .addField("Field: ", "Text", false)
        .attachFiles(["img1.png"])
        .setImage('attachment://img1.png');
        msg.channel.send(emb);

Is this intended behaviour or a bug? the path of setImage does not work when there are spaces. (example: if the image was named "image one.png" the image would still be attached to the message, but it will appear outside, above the embed)

drifting knot Mar 6, 2020, 3:26 AM

#

you want to be in #archive-djs-v12-deprecated @sick raft

#

as for you ishidres

#

caching is good if you need to query a bunch of users/channels/whatever but haven't had a presence update or smth come in to populate the data

sick raft Mar 6, 2020, 3:27 AM

#

on the contrary, that's where I came from, I've been told to ask if this is a bug in discord or not

drifting knot Mar 6, 2020, 3:27 AM

#

lmao

rigid trail Mar 6, 2020, 3:27 AM

#

caching is essential, its faster than fetching, and it minimizes api calls

drifting knot Mar 6, 2020, 3:28 AM

#

read their message burger

#

wait nvm im illiterate

rigid trail Mar 6, 2020, 3:28 AM

#

the lib also relies heavily on some caches

drifting knot Mar 6, 2020, 3:28 AM

#

yeah you need to cache stuff unless you want to get ratelimited by dapi

#

if you want to restrict caching their are options for it in the Client

copper laurel Mar 6, 2020, 3:40 AM

#

@sick raft very unlikely to be a bug - if there's spaces in the URL it would be resolved as image%20one.jpg and wont be a match

#

Besides, even if t is a Discord bug, theres nothing we can do about it so it still isn't really a library discussion

sick raft Mar 6, 2020, 3:51 AM

#

lol I see

#

but attachFiles works like a directory, not url lol

#

but ok

copper laurel Mar 6, 2020, 3:57 AM

#

It attaches it to be referenced by a URL

wild flax Mar 6, 2020, 5:47 AM

#

@drifting knot we already.. have one?

#

lol

drifting knot Mar 6, 2020, 5:48 AM

#

eslint config?

wild flax Mar 6, 2020, 5:48 AM

#

yes

#

we also have prettier already

#

and we are not using spaces

drifting knot Mar 6, 2020, 5:48 AM

#

hmmmm

#

prettier config is copied from discord.js repo

wild flax Mar 6, 2020, 5:49 AM

#

its already in the repo

#

same as the eslint config

drifting knot Mar 6, 2020, 5:49 AM

#

hmmmmmmmmmmmmmmmmmmmm

#

where

wild flax Mar 6, 2020, 5:49 AM

#

we only use prettier on lint staged

drifting knot Mar 6, 2020, 5:49 AM

#

ok

wild flax Mar 6, 2020, 5:49 AM

#

https://github.com/discordjs/discord.js-next/blob/master/package.json#L48-L62

drifting knot Mar 6, 2020, 5:50 AM

#

damn i guess im illiterate

wild flax Mar 6, 2020, 5:50 AM

#

so theres no reason to prettier the md files

#

nor the json files

drifting knot Mar 6, 2020, 5:53 AM

#

surely there is

#

why only use prettier on some of the files

wild flax Mar 6, 2020, 5:54 AM

#

because just as an example, the stuff prettier does to certain md files is not appreciated

drifting knot Mar 6, 2020, 5:56 AM

#

fair

#

im on mobile rn and can revert those changes tomorrow

#

or if you'd like you can do it

wild flax Mar 6, 2020, 5:56 AM

#

On the issue templates for example it weirdly merged the html comment and I had to manually fix it

#

Makes go a bit "eh"

drifting knot Mar 6, 2020, 5:57 AM

#

weird

vagrant holly Mar 6, 2020, 11:09 AM

#

👋

#

Is addFields or spliceFields ever used by the lib internally, or only by users?

#

The suggested simple fix in https://github.com/discordjs/discord.js/pull/3894 would mean these no longer skip validation unless I add duplicate logic in these methods.

warped crater Mar 6, 2020, 11:13 AM

#

if something is marked as @private in the jsdoc it isnt meant to be used externally

#

private things wont even show up on the docs by default

#

besides that, unsure about internal use for those two

vagrant holly Mar 6, 2020, 11:13 AM

#

I'll just add logic to be safe

#

¯_(ツ)_/¯

snow crypt Mar 6, 2020, 12:41 PM

#

i dont think they are

#

ill scan source

#

spliceFields

📎 unknown.png

#

addField and addFields

📎 unknown.png

#

@vagrant holly just embed and typings

vagrant holly Mar 6, 2020, 12:44 PM

#

yeah, that was the conclusion I reached as well

#

I ended up adding the logic in anyway, who knows what the lib might do in the future, might as well have those methods patched

drifting knot Mar 6, 2020, 6:21 PM

#

the collection module looks like it's the easiest thing to bring over to djs-next since it's already written in TS
does it need any refactoring or is it basically as easy as copy-pasting the file over?

another thing i'm wondering about is how unit/integration/e2e tests will be done.
will they be written in TS as well, or use the current method of JS testing with assert

wild flax Mar 6, 2020, 6:23 PM

#

Yeah, it "seems" like an easy PR

#

but no, I'll do that manually

#

And tests will be written in TS too

drifting knot Mar 6, 2020, 6:23 PM

#

cool

snow crypt Mar 6, 2020, 6:28 PM

#

what test lib?

wild flax Mar 6, 2020, 6:30 PM

#

jest

drifting knot Mar 6, 2020, 6:36 PM

#

rly

#

but the globals 😡

wild flax Mar 6, 2020, 6:45 PM

#

?

#

are you saying ava is better lol

drifting knot Mar 6, 2020, 6:55 PM

#

i might be

#

or node tap

snow crypt Mar 6, 2020, 7:14 PM

#

how do they even work

wild flax Mar 6, 2020, 7:14 PM

#

ava is sadly bad, and so is XO

drifting knot Mar 6, 2020, 7:33 PM

#

RIP

#

i like xo and ava

drifting knot Mar 6, 2020, 8:09 PM

#

more questions:

what service will we use for code coverage tracking? (ex. coveralls, codecov.io)
what is the types package for? typings for the dapi and/or types shared between d.js projects?

warped crater Mar 6, 2020, 8:10 PM

#

dapi

drifting knot Mar 6, 2020, 8:14 PM

#

thoughts on this https://github.com/itslukej/discord.d.ts

GitHub

itslukej/discord.d.ts

Typings for the Discord gateway. Contribute to itslukej/discord.d.ts development by creating an account on GitHub.

#

last time i posted it shay said "bad" and never explained why

warped crater Mar 6, 2020, 8:17 PM

#

shay is often not serious

#

theres not much to comment on that repo honestly

#

appel's spec-tacles has dapi types too

#

and so do I

drifting knot Mar 6, 2020, 8:18 PM

#

ok

#

can we use fake opaque types for maximum type safety

warped crater Mar 6, 2020, 8:19 PM

#

hell, looking at that repo it looks like a ripoff of spectacles cut into more files

drifting knot Mar 6, 2020, 8:20 PM

#

i mean its typings for the same api

warped crater Mar 6, 2020, 8:20 PM

#

I already asked Crawl about this, he said they'd just partly port appel's probably

drifting knot Mar 6, 2020, 8:20 PM

#

ok cool

#

where are appel's types?

#

nvm found them https://github.com/spec-tacles/types.js

GitHub

spec-tacles/types.js

Typings for Discord objects. Contribute to spec-tacles/types.js development by creating an account on GitHub.

real jetty Mar 8, 2020, 8:57 AM

#

Hey, so I am trying to implement throwing errors for empty embed names or fields

#

I edited addFields to this:

#

    if(!fields[0].name) throw new Error('EMBED_FIELD_NAME');
    if(!fields[0].value) throw new Error('EMBED_FIELD_VALUE');
    this.fields.push(...this.constructor.normalizeFields(fields));
    return this;
  }```

#

Then I realized that it just covers the first embed

#

If anyone would like to take over go ahead!

copper laurel Mar 8, 2020, 9:01 AM

#

So, addFields calls normalizeFields which calls normaliseField

#

Which does throw a range error if Util.resolveString returns a falsey value

real jetty Mar 8, 2020, 9:02 AM

#

It does not throw it, i've tested

copper laurel Mar 8, 2020, 9:02 AM

#

Problem is, undefined will resolve to "undefined" which is a valid string, not falsey

real jetty Mar 8, 2020, 9:02 AM

#

Yep

copper laurel Mar 8, 2020, 9:02 AM

#

So maybe in that function, throw before resolving and after? idk

real jetty Mar 8, 2020, 9:04 AM

#

Ye before resolving we can throw error

tired kindle Mar 8, 2020, 9:06 AM

#

actually, shouldn't the resolver take care of it? (I'm unsure about d.js design but I think the resolver itself should throw, as ResolveStringNotNullOrUndefined or so)

copper laurel Mar 8, 2020, 9:07 AM

#

Not sure I agree

#

undefined can be resolved to a string, as such

#

So it's doing its job

tired kindle Mar 8, 2020, 9:08 AM

#

I mean, implement a new resolver for the field values where it checks if it's undefined or not and then calls toString

copper laurel Mar 8, 2020, 9:08 AM

#

That's really what normalizeField already is

real jetty Mar 8, 2020, 9:08 AM

#

Im PRing a fix

tired kindle Mar 8, 2020, 9:08 AM

#

oh

real jetty Mar 8, 2020, 9:11 AM

#

Question:

#

- [ ] does this make the checkbox ticked or not

tough geode Mar 8, 2020, 9:13 AM

#

no
[x] - ticked checkbox
[ ] - non-ticked checkbox

real jetty Mar 8, 2020, 9:14 AM

#

Ok done

vernal atlas Mar 9, 2020, 6:01 PM

#

@copper laurel did you find out a way to fix the type guarding on partials in your PR?

copper laurel Mar 9, 2020, 6:01 PM

#

Was never specifically in my PR, doesn't seem to work in v12 either

vernal atlas Mar 9, 2020, 6:02 PM

#

hm

copper laurel Mar 9, 2020, 6:03 PM

#

It should just be the boolean literals true and false, but it isn't working

vernal atlas Mar 9, 2020, 6:03 PM

#

well best way i came up with is another union type for partials, which is each channel partialized,

type PartialChannelTypes = Partialize<Channel> | Partialize<DMChannel> etc```

copper laurel Mar 9, 2020, 6:03 PM

#

Hmm

vernal atlas Mar 9, 2020, 6:04 PM

#

i tried doing ChannelTypes | Partialize<ChannelTypes> but that just acted the same as before

copper laurel Mar 9, 2020, 6:04 PM

#

That's also what I tried

#

Will look into it later

#

Did you see my replies on your last set of comments?

vernal atlas Mar 9, 2020, 6:04 PM

#

yea

vernal atlas Mar 9, 2020, 6:37 PM

#

@copper laurel i cracked it sorta

#

type PartializeChannel<T extends ChannelTypes> = {
    id: string;
    partial: true;
        fetch(): Promise<T>;
  } & {
    [K in keyof Omit<T, 'id' | 'partial'>]: T | null;
  };```

#

then PartializeChannel<ChannelTypes>

copper laurel Mar 9, 2020, 6:37 PM

#

Ayyy I like that, ty

vernal atlas Mar 9, 2020, 6:38 PM

#

👍

copper laurel Mar 9, 2020, 6:39 PM

#

Hadn't had much chance to look into that part myself, I'll test that shortly

real jetty Mar 9, 2020, 7:23 PM

#

Regarding https://discordapp.com/channels/222078108977594368/223867697312694272/686654139559575698
Why is .endReason() a method? Why doesn't it get removed? Or implemted? Or become a getter?

copper laurel Mar 9, 2020, 7:53 PM

#

@vernal atlas How is your partialize different to whats already there?

type Partialize<T> = {
    id: string;
    partial: true;
    fetch(): Promise<T>;
  } & {
    [K in keyof Omit<T, 'id' | 'partial'>]: T[K] | null;
  };

  interface PartialChannel extends Partialize<Channel> {}

#

Other than changing what gets passed in as the generic

vernal atlas Mar 9, 2020, 7:54 PM

#

T extends ChannelTypes

#

whereas before it was just any

copper laurel Mar 9, 2020, 7:56 PM

#

I dont understand why that fixes it but okay lmao

slate nacelle Mar 9, 2020, 7:56 PM

#

Isn't that "just" a constraint?

copper laurel Mar 9, 2020, 7:56 PM

#

Thats why Im asking, yeah

vernal atlas Mar 9, 2020, 7:56 PM

#

🤷 not sure but it does fix the issue

slate nacelle Mar 9, 2020, 7:56 PM

#

Just TS things I guess ✨

vernal atlas Mar 9, 2020, 7:56 PM

#

also haven't tested but it might fix another issue with partials i pointed out

copper laurel Mar 9, 2020, 7:56 PM

#

So ```ts
type Partialize<T> = {
id: string;
partial: true;
fetch(): Promise<T>;
} & {
[K in keyof Omit<T, 'id' | 'partial'>]: T[K] | null;
};

interface PartialChannel extends Partialize<ChannelTypes> {}
``` doesnt work, but changing it to T extends ChannelTypes does?

#

(well, not changing, but making another)

vernal atlas Mar 9, 2020, 7:57 PM

#

https://discordapp.com/channels/222078108977594368/682166281826598932/683069729912389675

#

yes

copper laurel Mar 9, 2020, 7:57 PM

#

Yeah thats a bit of a larger question that implies there should be different Partializers for everything

#

...TS is now compiling fine without adding that additional constraint

#

I have no idea why I thought it didnt work for me before

vernal atlas Mar 9, 2020, 8:06 PM

#

i tried testing it without the T extends ChannelTypes and it was yelling at me

copper laurel Mar 9, 2020, 8:07 PM

#

Yeah it yelled at me too, or so I thought

#

But now I'm testing it without that, and it isnt yelling at me

#

Just using regular Partialize<ChannelTypes>

#

I also had to add partial: false to all the other channel types though

#

Noticed that was missing

vernal atlas Mar 9, 2020, 8:08 PM

#

hm

copper laurel Mar 9, 2020, 8:12 PM

#

Regarding your proposed change to Invite stuff

#

I think that's really out of the scope of my PR

vernal atlas Mar 9, 2020, 8:13 PM

#

can a DMChannel be updated? aka emitted in the channelUpdate event?

#

pretty sure it can't right?

copper laurel Mar 9, 2020, 8:13 PM

#

Since currently, GuildChannel#createInvite() is where that method is defined, so my typings are correct as it currently stands

#

I honestly dont know but I'd rather not block support for that in typings

#

The old event emit Channel so I'm keeping it equivalent, not trying to change things

vernal atlas Mar 9, 2020, 8:14 PM

#

the type guarding with Partialize<ChannelTypes> isn't working here for me?

📎 unknown.png

copper laurel Mar 9, 2020, 8:15 PM

#

It is for me though

📎 unknown.png

vernal atlas Mar 9, 2020, 8:16 PM

#

oh thats because you fetch it

copper laurel Mar 9, 2020, 8:16 PM

#

Well... yeah?

vernal atlas Mar 9, 2020, 8:16 PM

#

and fetch returns the full object

copper laurel Mar 9, 2020, 8:16 PM

#

Ahh i see

#

One sec

vernal atlas Mar 9, 2020, 8:17 PM

#

and you can still send messages to a partial channel etc

copper laurel Mar 9, 2020, 8:17 PM

#

Not according to typings

#

Okay, here's the way that didnt work

#

And doesnt work with T extends Channel either

#

if(!channel.partial) {
        if(channel.type === "text") {
            channel.send();
        }

        if(channel.type === "voice") {
            channel.join()
        }
    }

#

oh wait

#

explicit false check does work

#

Doesnt work

📎 unknown.png

vernal atlas Mar 9, 2020, 8:19 PM

#

i get an error because partial doesn't exist on CategoryChannel

copper laurel Mar 9, 2020, 8:19 PM

#

Works!

📎 unknown.png

#

Yeah it does exist in my typings

vernal atlas Mar 9, 2020, 8:19 PM

#

ah

copper laurel Mar 9, 2020, 8:20 PM

#

Needs to be false, not falsey, thanks Javascript

vernal atlas Mar 9, 2020, 8:20 PM

#

hm?

📎 unknown.png

warped crater Mar 9, 2020, 8:20 PM

#

huh, how does that happen

copper laurel Mar 9, 2020, 8:20 PM

#

My typings will define that

#

Its only a property if it gets partialized currently

#

So I'm adding public readonly partial: false to them

warped crater Mar 9, 2020, 8:21 PM

#

ah

vernal atlas Mar 9, 2020, 8:22 PM

#

shouldn't it be partial?: true, since if its partial, it will be true, if not it will be undefined

warped crater Mar 9, 2020, 8:22 PM

#

I tried to do this myself but failed miserably

copper laurel Mar 9, 2020, 8:22 PM

#

But I'm defining it to be false

vernal atlas Mar 9, 2020, 8:23 PM

#

oh i see

copper laurel Mar 9, 2020, 8:23 PM

#

You started it on the right track

vernal atlas Mar 9, 2020, 8:23 PM

#

👍

#

also, regarding this [K in keyof Omit<T, 'id' | 'partial'>]: T[K] | null; part of the Partialize type, this makes everything nullable right?

#

wouldn't that include functions? which aren't nullable

#

hm one sec

copper laurel Mar 9, 2020, 8:27 PM

#

This is really starting to get out of scope of my PR

vernal atlas Mar 9, 2020, 8:28 PM

#

ok im lost here now lmfao

#

[K in keyof Omit<T, 'id' | 'partial'>]: T | null;

T | null is meant to be T[K] | null

warped crater Mar 9, 2020, 8:28 PM

#

hoold on Sugen

vernal atlas Mar 9, 2020, 8:28 PM

#

and T | null is what's in my PartializeChannel

copper laurel Mar 9, 2020, 8:28 PM

#

It is T[K] null?

warped crater Mar 9, 2020, 8:28 PM

#

what is T over there, im on mobile

copper laurel Mar 9, 2020, 8:29 PM

#

Yeah but Im not using that

vernal atlas Mar 9, 2020, 8:29 PM

#

hold up lemme explain it fully

warped crater Mar 9, 2020, 8:29 PM

#

no I get what you're trying to do

#

im just asking what T extends

copper laurel Mar 9, 2020, 8:29 PM

#

In theory - anything

#

In this particular use case - channels

warped crater Mar 9, 2020, 8:30 PM

#

okay but what does he actually pass into the generic

vernal atlas Mar 9, 2020, 8:30 PM

#

    type PartializeChannel<T> = {
    id: string;
    partial: true;
        fetch(): Promise<T>;
  } & {
    [K in keyof Omit<T, 'id' | 'partial'>]: T | null;
  };```
the typeguard works fine with this

#

with T not extending ChannelType anymore

warped crater Mar 9, 2020, 8:30 PM

#

and you're passing channel classes? meguhappy

vernal atlas Mar 9, 2020, 8:30 PM

#

and instead T | null

warped crater Mar 9, 2020, 8:30 PM

#

the Omit there wont work correctly

vernal atlas Mar 9, 2020, 8:30 PM

#

instead of T[K] | null

warped crater Mar 9, 2020, 8:31 PM

#

nevermind, I get it now, yup, that looks good

#

or does it

vernal atlas Mar 9, 2020, 8:31 PM

#

T is a channel class tho, meaning that everything there would be a Channel

warped crater Mar 9, 2020, 8:31 PM

#

why T | null

vernal atlas Mar 9, 2020, 8:31 PM

#

exactly

#

but everything isn't a channel it seems

warped crater Mar 9, 2020, 8:31 PM

#

yeah meguFace

#

T[K] should be it

#

but you end up making methods nullable NotLikeThis

vernal atlas Mar 9, 2020, 8:33 PM

#

yeah you can fix that ez by doing T[K] extends Function ? T[K] : (T[K] | null) right

warped crater Mar 9, 2020, 8:33 PM

#

as far as I know, no

#

maybe that syntax works

#

but for starters extends -> is :+1:

vernal atlas Mar 9, 2020, 8:34 PM

#

hol up i think i got it

warped crater Mar 9, 2020, 8:34 PM

#

similar to how Array#isArray is typed as (value: any): value is any[]

copper laurel Mar 9, 2020, 8:43 PM

#

The typeguard is already working fine though

#

If you want to make less props nullable thats fine

vernal atlas Mar 9, 2020, 8:50 PM

#

@copper laurel don't you need to actually define partial in the constructor tho

#

otherwise the typings would be incorrect

copper laurel Mar 9, 2020, 8:52 PM

#

So maybe it should be partial?: true

vernal atlas Mar 9, 2020, 8:52 PM

#

yeah thats what i was saying above

copper laurel Mar 9, 2020, 8:52 PM

#

idk this is too hard lmao

vernal atlas Mar 9, 2020, 8:52 PM

#

also would it be better to put it on the Channel class

#

instead of each class individually

copper laurel Mar 9, 2020, 8:53 PM

#

Valuable learning experience for djs-next

vernal atlas Mar 9, 2020, 8:53 PM

#

as they all extend Channel

copper laurel Mar 9, 2020, 8:53 PM

#

Didn't work when I did that

#

idk why

#

Will try again, might just be VSCs fault

vernal atlas Mar 9, 2020, 8:54 PM

#

see,what i did here

  type PartializeChannel<T> = {
    id: string;
    partial: true;
        fetch(): Promise<T>;
  } & {
    [K in keyof Omit<T, 'id' | 'partial'>]: T | null;
  };```
works, but its technically incorrect, as it defines every property as being a ChannelType, but if you combine it with ChannelTypes, `ChannelTypes | PartializeChannel<ChannelTypes>` it appears to have everything with the correct property values / return values

#

oops accidental enter

#

~~lemme edit~~ edited

copper laurel Mar 9, 2020, 8:55 PM

#

I still don't at all understand why we need a separate partialize

#

Because as far as my testing goes we don't

vernal atlas Mar 9, 2020, 8:56 PM

#

yeah not saying we do just that what i pointed out earlier was a bit incorrect

copper laurel Mar 9, 2020, 8:56 PM

#

It should be T[K] | null right?

vernal atlas Mar 9, 2020, 8:56 PM

#

yea

copper laurel Mar 9, 2020, 8:56 PM

#

Then yeah

#

We just need partial defined on the other channel types

#

At least as an optional property I think

vernal atlas Mar 9, 2020, 8:57 PM

#

ig but the typings still need to show that the methods work on PartialChannels

copper laurel Mar 9, 2020, 8:58 PM

#

Out of scope of what I was aiming for in this PR though

#

Unless I widen that

#

I'm mobile, lemme get to my desk lol

vernal atlas Mar 9, 2020, 9:00 PM

#

hard to explain but i think it sort of is in the scope of the PR

copper laurel Mar 9, 2020, 9:07 PM

#

So

#

Where are you suggesting ChannelTypes | PartializeChannel<ChannelTypes> belongs?

vernal atlas Mar 9, 2020, 9:08 PM

#

i was suggesting that, but there needs to be another way

real jetty Mar 9, 2020, 9:11 PM

#

hello is there a difference between updateoverwrites and createoverwrites they look like the same to me so why 2 for 1?

vernal atlas Mar 9, 2020, 9:11 PM

#

current state of the PR, type guard seems not to work

📎 unknown.png

copper laurel Mar 9, 2020, 9:11 PM

#

Have you pulled down my latest commits?

vernal atlas Mar 9, 2020, 9:12 PM

#

ye

copper laurel Mar 9, 2020, 9:13 PM

#

So, is your concern here that methods dont exist on partials?

vernal atlas Mar 9, 2020, 9:13 PM

#

yea

#

i suppose it could be out of scope if you put it that way

#

both methods and properties don't exist for each partial channel type

copper laurel Mar 9, 2020, 9:15 PM

#

Im 50/50 - it wasnt my intention to change how partials work

vernal atlas Mar 9, 2020, 9:15 PM

#

ok

copper laurel Mar 9, 2020, 9:15 PM

#

Just to typeguard based on the existing implementation

#

Which currently, if not partial, can use channel

vernal atlas Mar 9, 2020, 9:16 PM

#

maybe ill do a PR for fixing the partials ig

copper laurel Mar 9, 2020, 9:24 PM

#

@vernal atlas ayy I tracked down why this shit doesnt work properly

#

Channels don't have a getter to determine when they're partial or not

📎 unknown.png

vernal atlas Mar 9, 2020, 9:25 PM

#

ah

copper laurel Mar 9, 2020, 9:27 PM

#

is the assumption that other channel types are never partial?

#

Yeah according to docs

Currently, the only type of channel that can be uncached is a DM channel, there is no reason why guild channels should not be cached.

vernal atlas Mar 9, 2020, 9:30 PM

#

so should channel events just not even emit a generic partial

copper laurel Mar 9, 2020, 9:30 PM

#

DMChannels

#

Can still be partial

vernal atlas Mar 9, 2020, 9:32 PM

#

yeah, so instead of emitting a PartialChannel, it should emit a PartialDMChannel

copper laurel Mar 9, 2020, 9:46 PM

#

I'm leaving that alone for now

vernal atlas Mar 9, 2020, 10:51 PM

#

everything with an id has createdAt right?

rigid trail Mar 9, 2020, 10:51 PM

#

it should yes, since you can get a timestamp from the snowflake

vernal atlas Mar 9, 2020, 10:51 PM

#

yea

copper laurel Mar 9, 2020, 10:55 PM

#

The docs on partials say this though

The only guaranteed data a partial structure can store is its ID. All other properties/methods should be considered invalid/defunct while accessing a partial structure.

vernal atlas Mar 9, 2020, 10:56 PM

#

yeah, that should be changed because of the ID, and some methods do work normally

#

like sending a message to a channel

#

i don't see what the channel being partial could effect there?

#

maybe theres a better example but you get my point

copper laurel Mar 9, 2020, 10:57 PM

#

I think like

tough geode Mar 13, 2020, 2:22 PM

#

oh, right

snow crypt Mar 13, 2020, 2:23 PM

#

https://github.com/discordjs/discord.js/blob/master/src/client/Client.js#L134

slate nacelle Mar 13, 2020, 2:57 PM

#

@tough geode another reason would be that the id of the client user is not necessarily the id of the application.
This is only the case for very old applications however.

ornate topaz Mar 13, 2020, 2:58 PM

#

(Nadeko bot would be one example)

vernal atlas Mar 13, 2020, 5:41 PM

#

i can't seem to get TS to recognise that messages is meant to be an array?

#

using the normal send overloads, not from my PR

spice lava Mar 13, 2020, 5:46 PM

#

you can cast it <Channel>.send().then(msg => (msg as Message[]).length ...)

vernal atlas Mar 13, 2020, 5:50 PM

#

should have explained but, i made a PR to remove some duplicate types as parameters for the send overloads, but then crawl commented saying that would break it, but i can see the overloads arent even working as they should in the first place @spice lava

#

so, not trying to fix an issue on my end, but with the send overloads

#

i think i found the problem, the overloads with only options need to be defined first, i think its something to do with the fact StringResolvable is any

snow crypt Mar 13, 2020, 6:33 PM

#

i pointed that out earlier

#

because send takes StringResolvable it would not autocomplete or type check

vernal atlas Mar 14, 2020, 2:57 AM

#

ah

solemn kite Mar 14, 2020, 6:37 AM

#

are intents fully implemented yet?

#

I imagine the new cache way is to make way for the intents system instead because cache is gonna be redundant at some time

#

... right? cause I assume its gonna be because of the intents you're unable to really make sure that the dats you have in the cache is gonna be there so its gonna be deprecated soon?

copper laurel Mar 14, 2020, 7:35 AM

#

Uhh, no?

#

Just specify all the intents so that we can cache stuff

#

Intents are available but not yet required

#

But there are some issues where if intents arent specified cache will be missing, yes

ornate topaz Mar 14, 2020, 1:24 PM

#

Even if missing, it will definitely not be redundant though. That would mean that bot doesn't cache literally anything at all, while you can just use the intent to get data for cache back again (assuming your bot will be allowed to use all intents)

analog oyster Mar 14, 2020, 4:35 PM

#

https://i.imgur.com/7cVhGLj.png

#

'none' | 'all' | 'everyone' aren't types PES_Hands

real jetty Mar 14, 2020, 4:38 PM

#

No the typings are correct, they are valid types in TS

#

You just didn't use it correctly

#

What is the code you tried?

analog oyster Mar 14, 2020, 4:39 PM

#

https://i.imgur.com/FzLzffh.png

#

what can I even do about this

real jetty Mar 14, 2020, 4:40 PM

#

Yeah like that, now you're encountering "type widening"

#

You can do disableMentions: 'everyone' as 'everyone'

analog oyster Mar 14, 2020, 4:40 PM

#

oh, thank you

#

sorry for coming to the wrong channel ig

copper laurel Mar 14, 2020, 9:23 PM

#

Or define CLIENT_OPTIONS as ClientOptions

#

Might also work /superlate

thorny cosmos Mar 15, 2020, 11:03 AM

#

Ok, I just came across something, <Guild>.verificationLevel returns a VerificationLevel typedef, which has a type of String

#

However, when I see the typeof this thing, it says it returns a number actually

#

same goes for the ExplicitContentFilterLevel and so... I don't know if this was done with intention for showing the string type in the respective methods which takes a parameter as a string | number, I think it would be better if the numbers are shown along with the Level alongside in the TypeDef and make the Types as string | number

real jetty Mar 15, 2020, 2:53 PM

#

version 12 of discord.js is a poop

opal mauve Mar 15, 2020, 2:53 PM

#

please don't randomly complain here. this channel is for serious discussion about the library.

vernal atlas Mar 15, 2020, 3:10 PM

#

@thorny cosmos oh, i think that came from a PR i made, maybe a merge conflict or smth changed it without realising

#

https://github.com/discordjs/discord.js/pull/3742/files#diff-e5c54069adfa0d32480eb3cc9faeee20R256

#

interesting

#

hm

#

@thorny cosmos i assume you are talking about

#

https://github.com/discordjs/discord.js/blob/69d69f25b964b6a2bb13bd086696abec17782ca3/src/structures/Guild.js#L254 this line?

#

because it looks as it should to me

#

perhaps your version is out of date

thorny cosmos Mar 15, 2020, 3:18 PM

#

Oh, you are right@vernal atlas , I realised i am using an older commit of v12, well, will update then

vernal atlas Mar 15, 2020, 3:19 PM

#

👍

green sentinel Mar 15, 2020, 5:59 PM

#

I think Channel should have a partial field, since when you use client.on("channelUpdate",...) you get a Channel | PartialChannel.

#

I use typescript so to check if the channel is partial I either have to just channel.fetch() or do something like

(channel as PartialChannel).partial ? channel.fetch() : channel as Channel;

vivid field Mar 15, 2020, 7:33 PM

#

DMChannels have a property showing if it's a partial or not, other channels are never partials as guild channels are always cached

raven juniper Mar 15, 2020, 7:34 PM

#

you can remove guild channels from cache

green sentinel Mar 15, 2020, 7:41 PM

#

I think it just makes sense to put a partial property on a value that can possibly be partial. Call me crazy.. lolol

copper laurel Mar 16, 2020, 12:20 AM

#

Theres already a PR in place for this

#

GuildChannel#partial will always return false
DMChannel#partial will do the appropriate check
Typings will be improved so directly handling the Channel class shouldnt happen, since its supposed to be abstract

tough geode Mar 16, 2020, 11:46 AM

#

https://github.com/discordjs/discord.js/blob/9c8aaf1bbcc220299e65cadb6ac54e1a7f208576/src/client/Client.js#L389 shouldn't this check for only 'all' | 'everyone' | 'none' instead of the type string?

opaque vessel Mar 16, 2020, 11:53 AM

#

Probably

#

It's specified in the typings as such

unique axle Mar 16, 2020, 5:53 PM

#

this channel is for library discussion not support

split steeple Mar 16, 2020, 5:54 PM

#

sry

finite rover Mar 16, 2020, 10:51 PM

#

is there any reason djs doesn't use ms precision ratelimit?

tired kindle Mar 17, 2020, 12:11 AM

#

~~you can't~~ you don't receive milliseconds precision from the API

#

https://discordapp.com/developers/docs/topics/rate-limits#header-format

X-RateLimit-Reset:
Epoch time (seconds since 00:00:00 UTC on January 1, 1970) at which the rate limit resets

#

it's in seconds, from the Discord API

opal mauve Mar 17, 2020, 12:19 AM

#

@finite rover discord.js does use ms precision

finite rover Mar 17, 2020, 1:14 AM

#

it's in s unless you tell it to give you ms

#

@opal mauve would you mind showing me the relevant line?

#

i ctrl-f ed the repo to find where the x-ratelimit-precision header was and I couldn't find it but maybe i missed smthing

opal mauve Mar 17, 2020, 1:16 AM

#

@finite rover oh, my bad. it's included in @tacit crypt's REST PR which is still in draft form.

#

https://github.com/discordjs/discord.js/pull/3443

finite rover Mar 17, 2020, 1:17 AM

#

yeah exactly what i found as well

#

so any reason this hasn't been introduced sooner?

#

it shouldn't imply much change apart from sending the additional header right?

tacit crypt Mar 17, 2020, 1:18 AM

#

It wasn't really needed?

opal mauve Mar 17, 2020, 1:19 AM

#

the PR that's adding it isn't finished yet

tacit crypt Mar 17, 2020, 1:19 AM

#

I mean you're not gaining SO much speed from it

finite rover Mar 17, 2020, 1:19 AM

#

that's fair, honestly I'm purely asking for information here

tacit crypt Mar 17, 2020, 1:19 AM

#

Maybe I'll actually finish that PR one day

#

After rewriting the entire bucket system...again

finite rover Mar 17, 2020, 1:20 AM

#

I was not even talking about the PR that seems to introduce much more than just ms precision

#

I was mostly wondering if adding ms precision would needmore changes than it seems

#

and that maybe I don't know about

tacit crypt Mar 17, 2020, 1:21 AM

#

I mean, to keep it simple

Can we use it? Absolutely
Should we? Probably?
Is it that important we use it? Not really
Is it gonna need massive changes? Nop, one liner. Testing tho... Yike

finite rover Mar 17, 2020, 1:21 AM

#

Alright

#

thanks for the answer

#

That's what I thought, I wanted confirmation thank you

#

Also yes I guess it doesn't matter that much theshrug

real jetty Mar 17, 2020, 7:33 AM

#

I have been thinking. But in the message reaction remove event, when fetching cuz it's a partial. this.emoji could be a guild emoji what could trigger the error. But why does message reaction add event doesn't have that problem? think

real jetty Mar 17, 2020, 8:12 AM

#

@real jetty I'll be opening a pr today that fixes this issue with an explanation to y all this mess happens

#

Okie

real jetty Mar 17, 2020, 4:23 PM

#

There is a lot of text and stuff but basically you use only messageReaction.fetch() where it would also fetch the message automatically? (if the message was also partial) (In the new solution such to speak)

#

yh thats pretty much wat it was doing before, now it just uses Message#_patch() to end up with the same result

#

u would only use MessageReaction#fetch() if u care about the count prop otherwise u prolly just wanna fetch the message depending on wat ur doing

alpine pier Mar 18, 2020, 3:11 PM

#

I might be stupid but why is TextChannel.members not a GuildMemberManager like Guild.members but a normal Collection instead?

rigid trail Mar 18, 2020, 3:12 PM

#

because its a getter, which means it would have to be a pseudomanager, and it would also be redundant with the guild's member manager, overall its just not really worth it, at least in my opinion

alpine pier Mar 18, 2020, 3:13 PM

#

I see. Thank you meowmeltthumbsup

vernal atlas Mar 18, 2020, 4:05 PM

#

would it make sense that the 'newMessage' in messageUpdate technically couldn't be a partial?

#

listening to the 'raw' event, i see that the full object seems to be sent by the websocket, which is enough to construct a full message right

slate nacelle Mar 18, 2020, 4:07 PM

#

More of an edge edge case:
If you receive an embed update for a partial (or uncached) message, you would only have relevant ids and embeds.

vernal atlas Mar 18, 2020, 4:07 PM

#

ah

real jetty Mar 18, 2020, 7:40 PM

#

yo is there an API for the discordjs.org website that returns info of the docs in JSON, kind of like how NodeJS docs has a "View as JSON" format?

slate nacelle Mar 18, 2020, 7:48 PM

#

There is the docs branch with the relevant json files you could use: https://github.com/discordjs/discord.js/tree/docs

real jetty Mar 18, 2020, 7:57 PM

#

thanks

frozen bridge Mar 18, 2020, 9:55 PM

#

haiya! I was wondering how long it takes issues to be looked at and fixed.

rigid trail Mar 18, 2020, 9:59 PM

#

it really depends on how complex, how big of a change it is, and if there are people who bother to fix it ¯_(ツ)_/¯

copper laurel Mar 18, 2020, 10:11 PM

#

Discussing it here/general (if its a feature/change) or in support (if its not actually an issue with the lib) would speed things along

#

Since it gets more eyes looking at the problem

ornate topaz Mar 19, 2020, 9:05 AM

#

Actually, is adding information about needed permission to setChannel desription (https://github.com/discordjs/discord.js/pull/3959/files#diff-3598194ff2d32819aea8eaf51aeb1b09R157) necessary?
This would make it only method that has it mentioned.

unique axle Mar 19, 2020, 9:42 AM

#

My stand on that: either everywhere or nowhere
In the past we've always come to the conclusion that discord.js should not document upstream permissions, as they weren't even clearly documented to begin with. Discord has now fully documented permissions on the API documentation per endpoint.

However since we recently introduced the permission and hierarchy matters this might have changed.
If we really want to start documenting permissions I'd personally be in favour of adding a link to the used API endpoint rather than just the permissions as part of the method description. If that is even needed/wanted however is another point of discussion

frozen bridge Mar 19, 2020, 4:38 PM

#

@copper laurel It's a big issue thats causing bigger bots not to be able to start:
https://github.com/discordjs/discord.js/issues/3956

#

Staff have seen it

#

just wondering how long it will take to be fixed

#

as I can't keep a shard offline for too long

wild flax Mar 19, 2020, 4:41 PM

#

There is really not enough information to even figure out where its starting or coming from currently

#

Despite the high priority it has

frozen bridge Mar 19, 2020, 4:46 PM

#

Any way I can help ya get info or something

wild flax Mar 19, 2020, 4:49 PM

#

cc @slate nacelle maybe you have a better general idea where this could come from and know which information we'd need

#

cc @tacit crypt

tacit crypt Mar 19, 2020, 4:53 PM

#

I..

For one, can you get me debug logs from the shards, especially those that error?

manager.spawn(20); // Start 4 Shards

This spawns 20 shards, not 4, try changing that?

#

As for errors like Error: Something took too long to do I have NO idea where it's coming from. Check your code to ensure it doesn't have many hanging loops..

slate nacelle Mar 19, 2020, 5:00 PM

#

My thesis is that this is caused by a guild not being available (presence limit reached or discord outage, probably #1).

In detail:

Error: Something took too long to do.
This is a Client (shard) error.

Thrown on v11, once a timeout elapsed before all guilds were ready.
(Nothing you can do about this on v11 sadly.)

That was taken care of with v12 by readying the client after a (different) timeout instead of erroring.
(You'll have unavailable guild(s) then though, but that's probably not an issue, d.js-wise)

Error: Shard 1's Client took too long to become ready.
This is a ShardingManager error.

This timeout can be disabled by setting the spawnTimeout to -1 or Infinity, may not be the fanciest solution, but should work.
https://discord.js.org/#/docs/main/stable/class/ShardingManager?scrollTo=spawn
(Nothing you can do about this on v11 sadly.)

frozen bridge Mar 19, 2020, 5:03 PM

#

@tacit crypt Same issue with 4, used 20 to make sure that only 1/20 of the servers had the bot down

tacit crypt Mar 19, 2020, 5:03 PM

#

Can you check the debug log for each shard?

frozen bridge Mar 19, 2020, 5:03 PM

#

I tried debug, nothing was wrong

#

I can start it with debug and send ya the logs 1 sec

tacit crypt Mar 19, 2020, 5:06 PM

#

Please do

frozen bridge Mar 19, 2020, 5:06 PM

#

The issue persisted in v12

#

Sharding Manager ➜ Launched shard 1
Shard 1 ➜ Using gateway wss://gateway.discord.gg/?v=6&encoding=json
Shard 1 ➜ [ws] [connection] Connecting to wss://gateway.discord.gg/?v=6&encoding=json
Shard 1 ➜ [ws] [connection] Identifying as a new session
Shard 1 ➜ [ws] [connection] Connected to gateway wss://gateway.discord.gg/?v=6&encoding=json
Shard 1 ➜ [ws] [connection] Setting a heartbeat interval for 41250ms

tacit crypt Mar 19, 2020, 5:07 PM

#

thats v11

#

i'm asking for v12

frozen bridge Mar 19, 2020, 5:07 PM

#

o

#

1 sec

#

Hmm it seems to be working now

#

If I get an offline guild ill try again

tacit crypt Mar 19, 2020, 5:18 PM

#

owo 👍

copper laurel Mar 19, 2020, 6:05 PM

#

Sorry, why was I pinged here? Is this relevant to me? I don't understand the sharding at all lol

frozen bridge Mar 19, 2020, 8:02 PM

#

Nope not anymore

#

@tacit crypt Yep it seems to work now, it just takes like a minute longer than the other shards. What could be causing this? (Nothing specifically singles this shard out in the code)

tacit crypt Mar 19, 2020, 8:05 PM

#

Because we wait for all guilds to come in. However, due to unavailable guilds, we wait 15 seconds. Aka

Say you have like 1000 guilds, and 5 are unavailable. We set a timer for 15s to mark the shard as ready regardless if all guilds came. We refresh that timer every time a guild comes. Once all but the last unavailable guilds came, the timeout takes over.

clever crypt Mar 19, 2020, 9:00 PM

#

I keep having problems with one unavailable guild - 7100 members.

Sometimes it doesn't come through until 60 seconds after coming ready, sometimes it's perfectly fine.

fallen arrow Mar 19, 2020, 9:02 PM

#

and this 15s timer is present in v12 but not v11, correct?

#

the solution i came up with is to manually run client.ws.connection.triggerReady(), not sure if its the best workaround

tacit crypt Mar 19, 2020, 9:11 PM

#

I keep having problems with one unavailable guild - 7100 members.

Sometimes it doesn't come through until 60 seconds after coming ready, sometimes it's perfectly fine.
@clever crypt blame discord

warped crater Mar 19, 2020, 10:03 PM

#

^ literally nothing you can do about it

#

after Discord's ready event they begin sending you guilds in the form of GUILD_CREATE packets

#

all d.js does is process them

glass hamlet Mar 20, 2020, 4:45 AM

#

Thanks @fallen arrow , that workaround is working partially for me. Gonna have to work with it for now

lofty birch Mar 20, 2020, 8:58 AM

#

So am I correct in thinking that the ready event is emitted when the bot connects, and the shard ready event is emitted when all of the guilds have been sent?

fallen arrow Mar 20, 2020, 9:18 AM

#

should be the opposite afaik

lofty birch Mar 20, 2020, 10:13 AM

#

Ah ok

real jetty Mar 20, 2020, 1:36 PM

#

https://discord.js.org/#/docs/main/stable/class/User?scrollTo=toString

When concatenated with a string, this automatically returns the user's mention instead of the User object.
That's not true, when doing someString + user, the valueOf method is used, resulting in the userID.
https://github.com/discordjs/discord.js/blob/stable/src/structures/Base.js#L37-L39
Doing somestring ${user} or user.toString() does work as expected, but still, I think the word "concatenated" isn't the right word for it.

old kelp Mar 20, 2020, 3:57 PM

#

I'm having a problem understanding the version 11.6.3 and 12.00

#

like what are they both

#

and as I know, master is the version 11.6.3, right?

frozen bridge Mar 20, 2020, 4:42 PM

#

@wild flax wyd ya reopen?

wild flax Mar 20, 2020, 4:42 PM

#

I commented.

frozen bridge Mar 20, 2020, 4:42 PM

#

o

#

yea just saw

#

One of the devs above thought that there was nothing we could do about it?

#

nope I misread

fallen arrow Mar 20, 2020, 5:22 PM

#

the only thing that could be done about it is to give it the same timer that v12 uses

real jetty Mar 20, 2020, 5:52 PM

#

return { guild };
What is the point in having the above over this:
return guild;

deft holly Mar 20, 2020, 5:54 PM

#

this is #archive-library-discussion 👀

real jetty Mar 20, 2020, 5:55 PM

#

Yea that is a line in the library

deft holly Mar 20, 2020, 5:56 PM

#

that's a question about the library code, but the answer would be it returns an object (shorthand for { guild: guild }), as to why, depends on context

real jetty Mar 20, 2020, 9:17 PM

#

message.channel.bulkDelete(2, true)```
works but return this error

(node:2068) UnhandledPromiseRejectionWarning: DiscordAPIError: 405: Method Not Allowed

#

 
  name: 'DiscordAPIError',
  message: '405: Method Not Allowed',
  path: '/api/v7/channels/%3C',
  code: 0,
  method: 'POST' }```

#

(Taken from #archive-djs-v12-deprecated )

#

Whats weird is from where %3C came from

raven juniper Mar 20, 2020, 9:39 PM

#

Without knowing the context, that sounds like it's in a DM

slate nacelle Mar 20, 2020, 9:44 PM

#

%3C is an encoded <
This method does not even hit that endpoint (POST /channels/:id)
(It impossibly could result in that error)
Bug reports should go to our issue tracker instead
(This does not include enough info though, see point 2)

clever crypt Mar 21, 2020, 2:17 AM

#

Is version 7 even available? https://discordapp.com/developers/docs/reference#api-versioning

vernal atlas Mar 21, 2020, 2:25 AM

#

doesnt seem to throw an error

#

but then again doing v9999 doesnt throw an error either

#

GWrjkKappaLUL

ornate topaz Mar 21, 2020, 2:53 AM

#

some people say v7 is "out" for multiple years now, but discord can't justify releasing it

tacit crypt Mar 21, 2020, 11:51 AM

#

v7 of rest is like v6 of rest but with better errors

vivid field Mar 21, 2020, 3:41 PM

#

Probably a rather nooby question, but why doesn't TextBasedChannel extend a regular Channel?

warped crater Mar 21, 2020, 3:42 PM

#

its an "interface"

#

and it's being implemented into whatever channels are "text based"

tough geode Mar 21, 2020, 3:45 PM

#

/\, it's never constructed, whatever it's implemented into in the lib extends Channel

warped crater Mar 21, 2020, 3:45 PM

#

in short, it's a class but its not used the "normal" way

real jetty Mar 21, 2020, 4:25 PM

#

guild.channels.create(name, options) and guild.roles.create({ data, reason }), I'd say it's better if GuildChannelManager#create would only accept one argument

opaque vessel Mar 21, 2020, 4:39 PM

#

Huh? It already does, the options are optional and the type defaults to a text channel

real jetty Mar 21, 2020, 4:44 PM

#

No I mean like guild.channels.create({ name: 'somevoice', type: 'voice', permissionOverwrites })

#

I'd say it's more consistent with other methods

raven juniper Mar 21, 2020, 4:49 PM

#

The object is for optional parameters, the name is not optional for channels, it is for roles (apparently)

#

You see the same format in emojis.create(), first parameters are required, and then an object for optional ones

real jetty Mar 21, 2020, 4:51 PM

#

Ahh so that's the logic behind it

ornate topaz Mar 21, 2020, 4:54 PM

#

regarding roles, discord defaults to new role name, additionally, it's name every role you create in app had at one point, since you can rename it after you create it

thorny copper Mar 21, 2020, 8:06 PM

#

I tried to update my bot to v12. It's horrible how Typescript's support has become worse than v11

#

Some examples:

#

https://cdn.discordapp.com/attachments/479323925726101514/690988941184073768/unknown.png

#

https://cdn.discordapp.com/attachments/664599917117046794/691011284262453278/unknown.png

clever crypt Mar 21, 2020, 8:20 PM

#

Manager#cache is a collection, which is an extended Map.
Typings for that would be for (const [id, channel] of server.channels.cache).
If you wish to loop over just the channels suffix the .cache access with .values().

thorny copper Mar 21, 2020, 8:20 PM

#

https://cdn.discordapp.com/attachments/664599917117046794/691013481318711296/unknown.png

clever crypt Mar 21, 2020, 8:20 PM

#

This is more of a support question that a complaint/suggestion/nit about the package.

thorny copper Mar 21, 2020, 8:20 PM

#

with .values()

#

I did find a solution tho

#

(server.channels.cache.array() as GuildChannel[])

clever crypt Mar 21, 2020, 8:22 PM

#

I have no problem with this code

for (const channel of <guild>.channels.cache.values()) {
    console.log(channel.permissionOverwrites);
}
``` *edited for your use case

thorny copper Mar 21, 2020, 8:23 PM

#

I don't think there is, but TS grabs it as a "any" type

clever crypt Mar 21, 2020, 8:23 PM

#

More or less a problem with your code editor

thorny copper Mar 21, 2020, 8:23 PM

#

Another fantasy

#

And this one is really from DJS

#

https://cdn.discordapp.com/attachments/484807050766057483/690985999282602034/unknown.png

#

fetch method doesn't take any arg, right?

clever crypt Mar 21, 2020, 8:24 PM

#

Correct

thorny copper Mar 21, 2020, 8:24 PM

#

https://cdn.discordapp.com/attachments/484807050766057483/690986061794639953/unknown.png

umbral aspen Mar 21, 2020, 8:24 PM

#

If you're calling fetch on the message, then yeah.

clever crypt Mar 21, 2020, 8:24 PM

#

What is message?

thorny copper Mar 21, 2020, 8:25 PM

#

client.on("message", (message) =>

umbral aspen Mar 21, 2020, 8:25 PM

#

Also wouldn't #archive-djs-v12-deprecated or #archive-updating-to-v13 be more suitable for this? Thonk

thorny copper Mar 21, 2020, 8:25 PM

#

@umbral aspen Actually I complaint more than I ask for help :x

real jetty Mar 21, 2020, 8:25 PM

#

For the collection loop, it should work fine on es6, how does your tsconfig look?

umbral aspen Mar 21, 2020, 8:26 PM

#

You seem to be asking questions that belong in the variety of subjects discussed / asked (about) in support center.

thorny copper Mar 21, 2020, 8:26 PM

#

Erf, I'm going to #archive-djs-v12-deprecated then

#

Sorry

spice lava Mar 21, 2020, 9:20 PM

#

We have GuildMember#bannable would there be a place for a general method like GuildMember.canban(GuildMember) for example. Or is this too high level same as addBlankField was removed

Edit: or would comparing roles positions by highest roles of each member be the preferred method without a custom implementation

rigid trail Mar 21, 2020, 9:29 PM

#

i really do not see a use for this

#

why dont you just check .bannable on the member Thonk

#

the last thing the lib should do is make wrappers for already kind of iffy permission getters blobsweats

spice lava Mar 21, 2020, 9:30 PM

#

As that would only count for the bot, as said this would be quite high level

rigid trail Mar 21, 2020, 9:30 PM

#

you want to see if a member can ban another? lol

#

again, dont see a use for it

#

its too specific of a method to be justified to be added, especially when it has no direct correlation with things the bot is doing

spice lava Mar 21, 2020, 9:31 PM

#

That's why I asked. Alright!

sly pollen Mar 22, 2020, 5:07 PM

#

Does djs use cluster module for sharding or no?

#

And does it scale to the number of cores available?

opal mauve Mar 22, 2020, 5:08 PM

#

the node cluster module isn't particularly useful for sharding

#

d.js uses normal child processes

sly pollen Mar 22, 2020, 5:15 PM

#

any reason why you guys wanted to go with normal child processes?

#

one thing i noticed about internal sharding is worker threads

#

would a build of several process that have several worker threads(shards) be a bad idea?

#

it's like a mix of internal sharding and process sharding except that you can get the most out of these processes?

opal mauve Mar 22, 2020, 5:17 PM

#

we went with child processes because we didn't need the functionality of the cluster module

#

worker thread based sharding was merged in https://github.com/discordjs/discord.js/pull/2908

sly pollen Mar 22, 2020, 5:20 PM

#

ah i see, but what about a mix between process sharding and threads sharding?

opal mauve Mar 22, 2020, 5:21 PM

#

unsure what advantage that would bring

sly pollen Mar 22, 2020, 5:28 PM

#

Instead of having 10 child processes with their own independent instances and their own memory being shards
We get to have 3 processes with 4 worker threads for each. These worker threads being shards (and every 4 worker threads do share the same memory)
instead of 10 child processes being shards, we get 12 workers being shards and only 3 processes

#

thats what i meant

undone ravine Mar 22, 2020, 6:11 PM

#

@sly pollen workers don't share memory with the main thread

copper laurel Mar 22, 2020, 10:55 PM

#

Does anyone know why the message event shares typings with messageDelete | messageReactionRemoveAll?

public on(
  event: 'message' | 'messageDelete' | 'messageReactionRemoveAll',
  listener: (message: Message | PartialMessage) => void,
): this;

When would client.on("message") emit a PartialMessage?

#

I can't think of a reason, so I feel like it should have its own typings

real jetty Mar 22, 2020, 11:23 PM

#

yh it should just be typed as Message. it never will be a partial within the message event, the author and content fields will always be present there so Message#partial will always be false

frozen bridge Mar 23, 2020, 11:10 PM

#

Haiya! I'm wondering if this is a bug or not... in djs 12.0.2, setting the position of a category to 0 puts it at the bottom of the channels, while setting it to 1 puts it under the channel in position 0

#

im using guild.channels.create

slate nacelle Mar 23, 2020, 11:13 PM

#

The position is passed to discord as-is: https://github.com/discordjs/discord.js/blob/stable/src/managers/GuildChannelManager.js#L121

frozen bridge Mar 23, 2020, 11:28 PM

#

Then why is setting the position to 0 having the category created at the bottom of the channel list?

#

guild.channels.create("Covid-19", {"type": "category", permissionOverwrites: [{id: guild.id, allow: "VIEW_CHANNEL", deny: "CONNECT"}], "position": 0})

rigid trail Mar 23, 2020, 11:40 PM

#

this is not at all the place for that question

copper laurel Mar 23, 2020, 11:40 PM

#

This is not a support channel

rigid trail Mar 25, 2020, 7:56 PM

#

you mean client#voiceStateUpdate?

#

it includes a voice state which has guild info..

#

not the right channel for this

#

please use a support channel

clever crypt Mar 25, 2020, 7:58 PM

#

?gotosupport

rich iglooBOT Mar 25, 2020, 7:58 PM

#

You are in #archive-library-discussion right now. Go to a support channel instead:

rigid trail Mar 25, 2020, 7:58 PM

#

on stable its a VoiceState for both params and on v11 its a guildmember so im not quite sure what you are suggesting

#

anyways its kind of offtopic since at this point its a code issue, not a library issue

copper laurel Mar 25, 2020, 10:44 PM

#

reaction.count isn't showing in docs on 12.0.2 for some reason

#

Possibly because // eslint-disable-next-line eqeqeq has been added between the JSDoc and the property?

slate nacelle Mar 25, 2020, 10:46 PM

#

~~I think there was a PR for that~~
This was fixed with https://github.com/discordjs/discord.js/pull/3980

uneven dawn Mar 26, 2020, 5:31 AM

#

I'm not sure the best practice on this, but instead of having a generic type like GuildChannel that's defined to extend Channel , shouldn't there be a GuildTextChannel that extends TextChannel etc.?

This way you don't have to cast everything that comes as GuildChannel to the text/voice/category/news/store equivalents

rigid trail Mar 26, 2020, 5:42 AM

#

so:

GuildTextChannel ---- TextChannel ---- GuildChannel ---- Channel```? what about DmChannels, those simply extend Channel and implement TextBasedChannel, similar to TextChannel which extends GuildChannel and implements TextBasedChannel

#

you will still have to cast, some methods can return any channel, say grabbing a channel from the GuildChannelManager, it can be text/voice/category, etc, so i fail to see how that solves anything, it seems to just add more clutter to inheritance

copper laurel Mar 26, 2020, 10:22 AM

#

Channel typings will definitely be improved in v13 with proper typescript

snow crypt Mar 26, 2020, 11:54 AM

#

I think messages could implement a guild typeguard

wild flax Mar 26, 2020, 11:54 AM

#

Like the non-working channel typeguard?

snow crypt Mar 26, 2020, 11:55 AM

#

I really don't know, but something to tell TS that if message.guild is present, message.channel is a TextChannel

plucky jay Mar 26, 2020, 11:55 AM

#

I don't think a typeguard like that is possible with TS

#

But maybe a GuildMessage or something?

snow crypt Mar 26, 2020, 11:56 AM

#

was about to say

wild flax Mar 26, 2020, 11:56 AM

#

That falls under what borger said right at the top there

snow crypt Mar 26, 2020, 11:56 AM

#

type GuildMessage = Message & {
  guild: Guild;
  channel: TextChannel;
  member: GuildMember;
};

type DMMessage = Message & {
  guild: null;
  channel: DMChannel; // can be partial(?)
  member: null;
};
``` and rest

#

some methods can return any channel

#

@wild flax why would that fail for message?

wild flax Mar 26, 2020, 11:59 AM

#

oh

snow crypt Mar 26, 2020, 12:01 PM

#

so would that work?

#

also, the channel typeguard works?

#

slate nacelle Mar 26, 2020, 12:04 PM

#

That part works, yes.
But breaks when attempting to use Promise#then on an union of Promises.
https://github.com/discordjs/discord.js/pull/3978

plucky jay Mar 26, 2020, 12:12 PM

#

@snow crypt But if you use message.guild instead, of message.channel.type === 'text', it won't work

snow crypt Mar 26, 2020, 12:12 PM

#

I know, that is why I say something like this could be used https://discordapp.com/channels/222078108977594368/682166281826598932/692703619455254579

#

type GuildMessage = Discord.Message & {
  guild: Discord.Guild;
  channel: Discord.TextChannel;
  member: Discord.GuildMember;
};

type DMMessage = Discord.Message & {
  guild: null;
  channel: Discord.DMChannel; // can be partial(?)
  member: null;
};

type AnyMessage = DMMessage | GuildMessage;

// @ts-ignore
const m : AnyMessage = get();
if (m.guild) {
  m.channel.permissionsFor
}

#

this works

tough geode Mar 29, 2020, 8:48 AM

#

why is the return type of CategoryChannel#children marked as nullable in docs (?Collection<Snowflake, GuildChannel>)? it's not nullable in typings, and it should not be, because it just returns Collection#filter(...), which returns Collection<K, V>, not nullable.
Docs link: https://discord.js.org/#/docs/main/stable/class/CategoryChannel?scrollTo=children

umbral aspen Mar 29, 2020, 8:59 AM

#

I think it should be nullable since a category doesn't necessarily need to have any children channels. In other words, a category can be empty.
This perhaps should be updated to it not being marked as nullable, or updating it so it is nullable and that it returns a collection with entries present in it or null.

tough geode Mar 29, 2020, 9:03 AM

#

categories can be empty, but .children is an empty collection for empty categories, not null, therefore it's misleading in my opinion.

thorny cosmos Mar 29, 2020, 9:06 AM

#

Yes, 1s3k3b is right, the jsdocs for .children is kind of wrong, since collections extend maps, they can give an empty collection object if no key value pair exists, here is what I just evaled on an empty category
Also the typings on Collection states that Collection#filter() returns a Collection, not ?Collection, so it makes sense for why should it not return ?Collection<Snowflake, GuildChannel>

snow crypt Mar 29, 2020, 9:18 AM

#

speaking of CategoryChannel, the setParent method is documented (because of GuildChannel), but it shouldn't be there at all

#

it looks like they tried to document it with a warning, but it doesn't show up https://github.com/discordjs/discord.js/blob/stable/src/structures/CategoryChannel.js#L19

tough geode Mar 29, 2020, 9:19 AM

#

well, it's inherited, but it has a warning

snow crypt Mar 29, 2020, 9:19 AM

#

it doesn't have

#

https://discord.js.org/#/docs/main/master/class/CategoryChannel?scrollTo=setParent

#

same thing for lockPermission

#

It can be removed by just deleting from the child's prototype (https://discordapp.com/channels/222078108977594368/222086648706498562/693752580664459295)

unique axle Mar 29, 2020, 9:32 AM

#

CategoryChannel#setParent should be there because the API technically allows it, despite throwing an error at this moment in time
Hence why instead the warning is introduced

snow crypt Mar 29, 2020, 9:32 AM

#

and what would it do?

unique axle Mar 29, 2020, 9:33 AM

#

as i just said, at this moment in time you can not tier categories like a file system, however the endpoint is there, so we decided to support it

snow crypt Mar 29, 2020, 9:46 AM

#

but it does practically nothing?

unique axle Mar 29, 2020, 9:55 AM

#

throwing an error at this moment in time
would not describe that as "nothing" per se

snow crypt Mar 29, 2020, 10:09 AM

#

i'm talking discord

#

like, you can set a topic to a voice channel with djs, but it really does nothing if you look at it practically

unique axle Mar 29, 2020, 10:11 AM

#

i fail to understand what you mean, the API throws an error, the request is rejected

snow crypt Mar 29, 2020, 10:11 AM

#

the api throws an error when trying to set a parent?

#

does the api throw or is it just the lib?

slate nacelle Mar 29, 2020, 10:15 AM

#

Depends on the error class you catch:
DiscordAPIError -> Discord threw it
DiscordjsError -> discord.js threw it

unique axle Mar 29, 2020, 10:16 AM

#

as said, API error, to be more precise:

DiscordAPIError: Invalid Form Body
parent_id: Categories cannot have subcategories

snow crypt Mar 29, 2020, 10:36 AM

#

well isn't this going with the head in the wall?

#

making a request that is going to fail is very strange to me

wild flax Mar 29, 2020, 10:39 AM

#

its not worth it to specifically handle that use case or somehow remove it from the documentation despite its inheritance

snow crypt Mar 29, 2020, 10:41 AM

#

but the library is there to interact with discord

#

could also split it to have a base GuildChannel, and a ParentedGuildChannel to have setParent on it, then have voice and text inherit from it, and categories would inherit from the base GuildChannel with no setParent

ornate topaz Mar 29, 2020, 2:29 PM

#

but the library is there to interact with discord
Well, it's a wrapper for Discord's API, so theoretically, if CategoryChannel#setParent is documented in API, there is no reason to not have it. It might reject today, but what if tomorrow discord will release subcategories?
It's a wrapper for whole API, not wrapper for some of its functions

copper laurel Mar 29, 2020, 10:53 PM

#

It exists because the API doesn't actually have different endpoints for different channel types the way Discord.js does. There isnt even a "set parent" API endpoint. Its literally just Modify Channel, specifying a parent_id
https://discordapp.com/developers/docs/resources/channel#modify-channel

setParent() is already just a helper function around an edit call really. The most logical place to share this inheritence is GuildChannel otherwise we'd be duplicating code across Text/Voice/News/Store channels

digital hawk Mar 30, 2020, 7:26 AM

#

can someone explain this change? https://github.com/discordjs/discord.js/commit/2be9ebaad2edca24f87f527397d4f2330f107a89#diff-4419bad7f25fd13106e6fe7516df7617

#

it doesnt play well with custom events and my event loader

#

i think having a [event: string]: any[] at the end of ClientEvents could potentially fix it

copper laurel Mar 30, 2020, 8:54 AM

#

Makes sense to me - I think we've effectively locked it down too far

slate nacelle Mar 30, 2020, 8:59 AM

#

You can augment the interface adding your own events there:

declare module 'discord.js' {
    interface ClientEvents {
        foo: [Map<string, any[]>];
    }
}

(Or that catch-all clause, too.)

snow crypt Mar 30, 2020, 9:00 AM

#

I think it should be event: string | symbol in the all-catch

#

now how do i make a pr again

slate nacelle Mar 30, 2020, 9:02 AM

#

This is intentional: https://github.com/discordjs/discord.js/pull/3944#discussion_r393221258

snow crypt Mar 30, 2020, 9:04 AM

#

so I guess we're not adding it back?

slate nacelle Mar 30, 2020, 9:04 AM

#

I'd be in favor to not do that.

snow crypt Mar 30, 2020, 9:05 AM

#

I'll do a pr with that note

#

@slate nacelle https://github.com/discordjs/discord.js/pull/4008

wild flax Mar 30, 2020, 9:32 AM

#

This doesn't make a lot of sense

#

why would you not extend the interface?

#

Why would you want to just allow any symbol or string?

snow crypt Mar 30, 2020, 9:33 AM

#

as written in the pull, i'm just waiting for feedback

wild flax Mar 30, 2020, 9:34 AM

#

You got feedback from 2 devs

snow crypt Mar 30, 2020, 9:34 AM

#

why not more ¯_(ツ)_/¯

vivid field Mar 30, 2020, 12:49 PM

#

Probably more of an api question, but is there a reason why guildMemberRemove gives you a member object whereas guildBanAdd gives you a user object?

rigid trail Mar 30, 2020, 12:49 PM

#

hackbans

#

you can ban a user by id which is not in the guild (see: Guild#members#ban())

snow crypt Mar 30, 2020, 1:17 PM

#

I would not be surprised if member.ban calls member.guild.members.ban

#

aaand... it does

rigid trail Mar 30, 2020, 1:18 PM

#

why would it not

digital hawk Mar 30, 2020, 4:03 PM

#

@slate nacelle I went ahead and just made my own interface in my bot which extends ClientEvents but I guess adding it in a d.ts works too

snow crypt Mar 30, 2020, 6:57 PM

#

so do i close the pr?

#

yep

tender surge Apr 2, 2020, 6:27 AM

#

Wish intellisense would still work if you allowed anything but yea if you allow anything now you either don't get autocompletion or you can't get the typings that are returned from each event

#

Also if you extend now you can easily type what's returned from your custom event instead of asserting

snow crypt Apr 5, 2020, 5:11 PM

#

Would v11 break if moved to node-fetch?

warped crater Apr 5, 2020, 5:27 PM

#

otherwise it would have been done already

tacit crypt Apr 5, 2020, 5:39 PM

#

Would it break? No.
Would it be breaking? Yes

undone ravine Apr 5, 2020, 5:42 PM

#

It wouldn't be breaking

tacit crypt Apr 5, 2020, 5:47 PM

#

breaking from a semver point, no?

rigid trail Apr 5, 2020, 5:49 PM

#

if nothing changes on the public API then i dont see how it would be semver major

undone ravine Apr 5, 2020, 5:49 PM

#

No, unless there are points where raw return values are given to a user from snekfetch/node-fetch

#

Changing dependencies is not a breaking change

slate nacelle Apr 5, 2020, 5:51 PM

#

We expose the raw errors from snekfetch when encountering 5xxs and the retry limit was reached:
https://github.com/discordjs/discord.js/blob/v11/src/client/rest/RequestHandlers/Sequential.js#L76
https://github.com/discordjs/discord.js/blob/v11/src/client/rest/RequestHandlers/Burst.js#L46

snow crypt Apr 5, 2020, 5:53 PM

#

if nothing changes on the public API then i dont see how it would be semver major
that's why i'm asking

rigid trail Apr 5, 2020, 5:56 PM

#

i mean, i believe there is little to be gained from switching to node fetch but a significant amount to lose if done improperly, also, v11 isnt really in development as much as v12, as long as snekfetch works i dont see a need

real jetty Apr 6, 2020, 7:51 PM

#

Would there be any way of adding in where an API call failed in the code? As in, more detailed stack traces

#

So instead of it coming from the library it would come from a method called from code the user had written

rigid trail Apr 6, 2020, 7:54 PM

#

you mean unhandled promise rejections? not really no, as async code is run in parallel to the stack elsewhere, therefore stack traces are hard to make and would be huge

exotic flicker Apr 6, 2020, 7:54 PM

#

v11-dev had a way to show accurate stack traces, but afaik node always had this missery with async stack traces

real jetty Apr 6, 2020, 7:57 PM

#

then what changed between v11-dev and now that required that code to be changed?

#

well, ok probably a lot in all fairness but

slate nacelle Apr 6, 2020, 7:59 PM

#

There is a "round 2" for this in #3443.
A bit of the relevant code, but it's all over the place, really:
https://github.com/discordjs/discord.js/pull/3443/files#diff-b06340a028069dac07f7b735c4e096a5R12
https://github.com/discordjs/discord.js/pull/3443/files#diff-162f9af6d66ef5705d4310712d982f4bR85
Comment linking "round 1":
https://github.com/discordjs/discord.js/pull/3443#issuecomment-523510644

exotic flicker Apr 6, 2020, 8:02 PM

#

discords information on missing permissions / missing access is also lacking (like what permission is missing)
tho this is a feature request
https://github.com/discord/discord-api-docs/issues/1467

tacit crypt Apr 6, 2020, 8:34 PM

#

My implementation in that PR you linked @slate nacelle would only work on v8 sadly (which would be fine if we didn't have browser builds); it's to be changed

slate nacelle Apr 6, 2020, 8:34 PM

#

RIP

#

So basically like appells comment said.

tacit crypt Apr 7, 2020, 12:49 PM

#

Well, yes but actually no

tacit crypt Apr 8, 2020, 3:54 PM

#

TypeError: timeout.refresh is not a function
    at GABClient.handler (C:\Users\GAwesomeBot\Development\node_modules\discord.js\src\stores\GuildMemberStore.js:210:17)
    at GABClient.emit (events.js:323:22)
    at GABClient.EventEmitter.emit (domain.js:482:12)
    at Object.module.exports [as GUILD_MEMBERS_CHUNK] (C:\Users\GAwesomeBot\Development\node_modules\discord.js\src\client\websocket\handlers\GUILD_MEMBERS_CHUNK.js:18:10)
    at WebSocketManager.handlePacket (C:\Users\GAwesomeBot\Development\node_modules\discord.js\src\client\websocket\WebSocketManager.js:391:31)
    at WebSocketShard.onPacket (C:\Users\GAwesomeBot\Development\node_modules\discord.js\src\client\websocket\WebSocketShard.js:437:22)
    at WebSocketShard.onMessage (C:\Users\GAwesomeBot\Development\node_modules\discord.js\src\client\websocket\WebSocketShard.js:295:10)
    at WebSocket.onMessage (C:\Users\GAwesomeBot\Development\node_modules\ws\lib\event-target.js:120:16)
    at WebSocket.emit (events.js:311:20)
    at WebSocket.EventEmitter.emit (domain.js:482:12)``` If anyone wants to have a look, I'm a bit busy ![sad](https://cdn.discordapp.com/emojis/585885410257928194.webp?size=128 "sad")

rigid trail Apr 8, 2020, 3:56 PM

#

GuildMemberStore 🤔

tacit crypt Apr 8, 2020, 3:57 PM

#

yes, i'm not on latest cause we're rewriting soon

rigid trail Apr 8, 2020, 3:58 PM

#

what node version are u running? cause timeout.refresh was added in v10

tacit crypt Apr 8, 2020, 4:49 PM

#

nodejs 12.something

#

OH