The same thing can be said about client side scripts and the reason why I have't even bothered to try it. They're nice for login screens but as soon as you grow a bit you start to get into management hell and everything turns into an ugly clusterf*.
Mostly. It helped a lot in preventing a condition which happened with the official plugin where your script grew too big to tame.
I'm currently in the process of creating a libuv(https://github.com/libuv/libuv) module. However, this might take some time because there's no playing around with Squirrel and threads and also with their differences on memory management. There might be some issues here and there until I understand how libuv works. But I hope everything will go as smooth as possible. I've just started on it tonight and so far I've only got the basic layout of how I'll proceed with the whole implementation. Here's the libuv timer that I just exposed a few moments ago to test whether the approach that I'm taking is functional:
One thing I don't like is that it is just as hard and hacky to implement a Discord bot here as the official plugin.
Let's hope the remaining things go as smooth as this basic prototype did so far.
I'm not even sure how to respond to that. I haven't reached that stage. But I'll take it as a good sign.
I'm glad the plugin is stable enough for a public server. Some people asked me about that and I thought my answer to them could be biased. So that's why you see this topic here.
Even though the plugin itself is just a bunch of hackish pieces of code glued together as I figured out what I needed during the development. And as I was learning stuff. At least it works. And decently enough it seems.
I've started working on a 2.0 rewrite that would keep the same API so nothing would break, but the actual implementation would be better. However, as time wen't on, I got into other things and I stopped working on it. And I doubt I'll do it anytime in the near future. Vcmp is just not worth the effort anymore.
Anyway, thanks for the input.