Thanks! I don’t really plan to be though 😄 there’s so much more to web dev that I’d like to cover but I’m building up to it

Thank you!

Glad to hear that! Thanks for stopping by to comment!

😄🙌

Hello! 👋

soft delete can implemented just by using deletedAt as null and adding this filter in each query

Great! Good luck on your project!

@@mariusespejo thank you💪🏿

Thank you 🙂

Thanks for the idea!

You’re welcome!

I haven’t tried movies although for music I think it sounds amazing. It’s meant to be a studio monitor so it lets me hear all the details. I’m not an audio expert at all but that does mean it’s meant to be somewhat “neutral”, and I’m not sure what’s ideal for movies. For my use case if it’s not obvious I use it to monitor my mic to make sure it’s not recording things it shouldn’t, and to generally hear its quality in realtime as well as in post production

Sorry been a while since I wrote the code but I think you’re right, probably ideal to clearTimeout, I think I intended to do that that’s why we have the timeout variable defined. Basically we’re just trying to debounce there. Good eye!

@@mariusespejo I was going to comment the same thing.

Thank you, that could be a good video topic for the new year (how to approach learning things)! Will definitely consider it

@@mariusespejo Dying to watch it

Yeah that could be a fun tutorial, will think about it!

Yeah I think that would be fun to build! I’ll see if I can plan for that, thanks for the suggestion

Glad you think so, thanks!!

I’ll consider it, I don’t think there’s really much to it though, just store the rules as JSON and manage it. Casl can initialize an ability from json

Yeah man, glad you like it!

Take a look at socket.io, I believe it has built-in support for rooms

@@mariusespejo can you please make a video in this topic?

Thanks for feedback. Definitely still trying to find a good balance for things such as pacing and writing vs pasting code

There’s a thing called swaggger-socket but that’s only if you were attempting to do REST over websockets, which is probably uncommon. Other than that I’m not sure there’s any common standard for how to structure websockets so there likely isn’t much swagger-like solutions. You could always make your own custom documentation though

@@mariusespejo thanks

Sorry it’s just in the video for now, you’ll find it’s actually not a lot of code to write

Thanks for the suggestion, will keep it in mind!

I’m sure there are more details to your use case but my gut reaction is that: if you’re already using websocket, meaning you have a persistent connection, why wouldn’t you just send the data directly? You still have one source of truth which is the server who now says “hey all you clients this is now the current data!” And clients could react and update immediately. However your other idea of simply publishing a trigger, it still means that each client still needs to spend additional time to fetch data which sounds like could have been available to them at the same time they got the trigger. However, there is nothing inherently wrong with that approach. If it’s because you’re just trying to reuse APIs, that makes sense to me

@@mariusespejo Hmm, that is a very good point. I just have to make sure that all data transformation occurs upfront within the service's method before returning it. Then it would be as you said with the server being the source of truth. Thanks.

Monokai pro on this one I think!

Hey! I’m sorry but the code isn’t on github

@@mariusespejo Okay, thank you anyway :)

The underlying websocket library you use like socket.io usually will take care of keeping track of which clients are connected. You just need to build a way for your server to identify who the user is on that client, e.g. basically auth. Put those two together and you’ve got your list of connected users.

Hey sorry not on github at the moment, I recommend taking a look at the official docs for the most up to date guidance

My issue was related with my M2 chip and my node version. Is already working fine. Thank you for your answer.@@mariusespejo

if anyone get the error like me, give it a key like this should work

I have some videos on auth. The fundamentals are all the same, you need a way to identify the user (e.g. perhaps upon connection require a login event, or do login via REST prior to allowing a connection). Once you can identify which user is dispatching an event, then it’s up to you how to do authorization, you can pull the user record and check they have permissions etc. Generally with nest you can utilize guards to do those checks. Few videos about that in the channel

To some degree yes, here’s an example: stackoverflow.com/questions/58670553/nestjs-gateway-websocket-how-to-send-jwt-access-token-through-socket-emit However I don’t think you can add a guard to the entire gateway as a whole, and notice in github open ticket for handleConnection, can’t put a guard on it: github.com/nestjs/nest/issues/882

You can dispatch the current list of messages to a client when they connect, or you can use a separate API call to fetch it initially. Various ways to go about it

I believe I’m using cascadia code at the moment, I’m not sure haven’t changed it in a while

Using websockets is already one of the ways to directly push events from the server, what else are you looking for exactly?

@@mariusespejo to learn more as to understand the alternative way too.. I just heard sse is also good fot push event and compare to web socket, sse is good to go - i don't know what is it so I am here to know! If possible do make a tutorial one🙏🏽

Ah! I actually didn’t even know that was a thing, today I learn from you, thanks! I’ll look into it and maybe form opinions about it first. But at quick glance, it looks like it’s not bidirectional so the use case is limited to communication only coming from server. Also there is a hard limit of 6 connections per browser, unless it’s on HTTP/2 … that sounds pretty limiting. I could see this being useful for push notifications though!

I have a few videos on authorization already, it’s pretty much the same principles. E.g. see the CASL videos that I’ve made

Sure thanks for feedback, I’ll think about it. My content is already designed to be something you can follow along with from start to finish though. There are a lot of people that simply want the code and not to watch and you just won’t get the same value out of the content that way in my opinion.

It’s important to look past the “it’s just another chat app” and actually learn and understand the underlying fundamentals required to engineer a more specific problem, e.g. if you’re putting together your own real-time feature. Your use case however is about integration with a 3rd party. Any external api generally should include usage docs, you should rely on that not tutorials. If it doesn’t have docs… reach out to the devs

You can! Graphql subscriptions use websockets under the hood. So it’s just a different interface basically but same idea. I have an example of that in my teamseas video if you’re interested

@@mariusespejo yes please i am migrating to fully graphql with graphql-ws

It’s in the channel!

@@mariusespejo Is this the same as Server-Sent Events?

Paul, sorry misunderstood the question. No it’s not the same as server-sent, which is not bidirectional like websockets are

Thanks for checking in, everything is good! I am working on a new video, just had a lot on my plate this past month

but you’re the only one who has complained so far lol have a great day

There are timestamps in the description…? what checkpoints are you looking for?

@@mariusespejo We need to check in between how server is working. Postman testing in between so that people can code along. As of now we I say I want to code along with you I would just get to know my codes works at the end of the video