Ability to check if DMs are disabled

[TehIvan]

I would prefer if I could check if the bot is capable of dm’ing the bot - rather than catch the error and handle it.

[DaliborTrampota]

Yes, this is pretty much needed. Right now you gotta send test message if it sends, delete the message in some cases so 2 api calls for nothing

[8auu]

Why do you have to send two messages? Just send the one message and handle the error if it doesn't send..

[DaliborTrampota]

Why do you have to send two messages? Just send the one message and handle the error if it doesn't send..

Didnt say send to messages but send it and delete it + I said in some cases. Sometimes I just need to know if I can dm the user later

[Zoddo]

This would lead to two API calls: one to check if you can DM a user, and one to actually send the DM.

Currently, you have both features in a single API call: you send the DM, and if you got a 403, that means you are not allowed to send DM.

I don't see how this proposition would be useful unless you have a very specific use case where you want to check if you can DM a user without actually sending a DM.

[SinisterRectus]

The permission check is in the request to send a message, not in the request to open the channel.

[Zoddo]

Currently, you have both features in a single API call: you send the DM, and if you got a 403, that means you are not allowed to send DM.

Incorrect
First, you open the DMS
Second, you send the DM

Yes, you are right. But that doesn't change the issue: the op wants to do 3 api calls for something that can currently be done with only 2 calls.

[ghost]

er, no. something that would solve this in just 2 API calls is returning 400 on DM channel create if you cannot DM an user

[Zoddo]

er, no. something that would solve this in just 2 API calls is returning 400 on DM channel create if you cannot DM an user

Hmm, yes this is a good suggestion 😃

(but then people will complain it's still a 4xx error that could trigger a cloudflare ban 😕)

[ghost]

400 doesn't trigger a cloudflare ban - just 401, 403, or 429

[8auu]

Well then at that point you’d also be rate limited for sending too many test dms and deleting them, and then sending another message…

…

On Wed, 1 Sep 2021 at 7:18 pm, Simon Beal ***@***.***> wrote: Discord gives you an API ban if you get too many 403s. Whilst my bot doesn't normally send many DMs, it wouldn't be too awful to have a user bot farm killing off bots by getting them to DM you and blocking them on repeat — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#3747 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AM6DLUSXMSCSWBY5P2QV7HDT7ZU6RANCNFSM5DDF35AQ> .

[muddyfish]

Not if you have enough accounts, like certain user farms..

[8auu]

Discord gives you an API ban if you get too many 403s.

if you were sending 2 DMs per user to avoid getting a 403, you'd be sending 2x as many DMs, as well as having to delete the original DM? I don't see how this helps at all?

[muddyfish]

The feature request is an endpoint to detect if you can DM a user without actually DMing them, so I don't see how what you're saying is relevant

[TehIvan]

^^

[TehIvan]

If you’re concerned about api spam, then I would like to suggest mass-checking dms.
This is very useful - for example in a server where dms are required for them to be able to use the server properly and login to captcha.