-
Notifications
You must be signed in to change notification settings - Fork 201
ci: harden workflows with pinned actions, timeouts, and concurrency controls #1412
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
base: main
Are you sure you want to change the base?
Conversation
…ern cross install - Add timeout-minutes to all jobs to prevent hung workflows - Add concurrency groups to cancel redundant runs - Add fetch-depth: 1 for faster shallow clones - Replace manual cross download with taiki-e/install-action - Keep all action SHAs pinned for security
…ts (native,cross)
…-cache, install-action)
|
@niStee Are you using LLMs to write/translate/check your comments? |
Yes, I use them – for me it’s just a smarter spell‑checker; if the style bothers you, I can try to make future replies shorter and less formal. |
|
The style isn't the problem. It's the length and verbosity. Take this comment:
I would've written this as:
The information you're putting in the comment is mostly duplicated or noisy:
That's just what I said but in more words. What benefit do you think it gives to write it out like this?
I can see you changing the title, and I can see you opening a new PR; link them if you so wish, but don't duplicate all this information.
You seem to understand this. I asked you to split it up, so I assume you know I understand it as well. Why would you need to clarify this? It looks like you're not putting any effort into your comments, just letting LLMs generate them. Which is quite disrespectful; you're not putting the effort into writing an extensive comment, instead letting an LLM do that, but you expect me to read the entire thing? You're just wasting my time at that point. (and if you believe it's good to have long verbose communication, perhaps even interpreting that verbose communication with an LLM, then I disagree, and I do not want that type of communication here) I really don't mind long comments, but they have to be useful. What LLMs tend to do is generate text that looks great on the surface, but when you actually read it, turns out to be full of obvious information, duplication, and useless styling. Which (again) just wastes people's time! A great way to make sure your comment is actually useful is to write it yourself. You'll notice if what you're writing isn't useful information. And if you want to pass it through a spell-checker or an LLM then fine (even though I don't really care, I don't mind bad English at all), but don't post any text that you wouldn't have written yourself. If I am interpreting this wrong, let me know. It's not my intention to "accuse" you of anything, but I feel like I have to be a bit harsh to keep out any bad faith. I cannot keep reviewing your stuff if you don't improve your communication. |
What does this PR do?
Hardens existing CI and release workflows with security and reliability best practices.
Changes:
Security hardening:
Reliability improvements:
Maintenance improvements:
crosscurl install withtaiki-e/install-actionfor easier Renovate updatesNotes: