New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Open Discussion] - Create a core cog repo under Cog-Creators #3618
Comments
As you've mentioned, this is a very radical change and I'm gonna say a no with a full stop on it. Downloader is not designed for this kind of load or behavior, and we've discussed the potential issues and bottlenecks regarding this subject within Discord. However this requires a fundamental rewrite on how Downloader works and on how we distribute Red as a software. As example we have 56 open PRs at the time of writing this post. >90% is directed towards a cog. Asking people to close their PRs and move over to another repository, with another workflow is simply, in my opinion, unacceptable from an open source community. Other challenges arise too such as CI Workflows, contributing guidelines, repository setup and management. The overhead is simply too big to be managed sanely. More pressing concerns have also been raised in Discord with updating a individual cog. As of right now when you install Red, you can be guaranteed that it works. (3rd party cogs excluded, hence why their 3rd party in the first place, each person has their own implementation on an issue after all) Some cogs within Red require other cogs to function correctly, think Modlog that relies on Mod, and Economy on Bank. Hotfixes and minor releases are the answer here, and as I've already said in our Discord channels. Having a clear policy on when and where to do them is the correct way forward here. We already have the infrastructure in effect for them and they require no radical rework on how we work as an open source community. This is gonna be a hard no from me, it requires too much maintenance to keep working and stable and only adds to our already large workload for an open source piece of software. |
I agree with what Kowlin wrote. I'd like to also point out that a bunch of issues will inevitably be opened on the wrong repo if this happens. It could turn out to be much harder to communicate the difference to the end user. Something I would like however is the possibility to choose versions of the core cogs. Without the seperation that's not going to be possible though and the arguments in favor of keeping it as is seem more important overall to me. |
Unsure what downloader rewrite would be needed here. But the idea is that the bundled cogs would be updated on releases as they are now, this repo would just have fixes/updates before they make into the bundled cogs, the user could choose to use in case they need fixes that haven't made into Red yet. but since cogs won't be delivered via "carrier pigeon" closing this as any attempt to discuss this seems pointless. |
Summary
I've made a suggestion internally, but it would be nice to get the opinion from the community and org members alike.
Points for discussion
What was proposed
Pros
[p]streamalert
commands should accept links[p]streamalert
commands should accept links #3480[p]cog pin
Cons
Disclaimer: I'm strongly in favor of going this route, I believe it would be beneficial to Red users and such view may be clouding potential Cons.
The text was updated successfully, but these errors were encountered: