-
Notifications
You must be signed in to change notification settings - Fork 22.5k
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
fix(ci): recycle the weekly spelling check issue #35653
Conversation
If we do this we should be able to run this far more often—what about once per day? Or even once after each new commit? |
Once a day sounds good because content goes live at 24-hour frequency. So whether you fix typos after every commit or once a day is the same thing. If we could run this 2-3 hours before and fix typos before the prod deployment, that would be sufficient. |
I don't think we should assume anyone could get to typos at any time. Thinking about it, I think weekly is fine since we don't want to encourage people to send typo fixes every day. |
Also, we hardly get 3-4 typos per week. |
I think in this PR:
|
The current workflow would work like this:
Note that there is no way to block the issue from being closed. If someone tags the issue in PR then the issue will be closed after the PR is merged. Working on the timestamp and making files linkable in a list.
We don't have to do this, as the closed issue will be the indicator of "all clear." |
I don't think we should make people ever close that issue. Closing and reopening generates a lot of noise in the history and makes it barely scrollable. I think we should keep it like a dashboard. Think of something like the Renovate dependency dashboard. |
If we run the workflow weekly, then how would you mark the issue as addressed? Also, autoclosing from PR is the best mechanism to tell other contributors that this has been fixed, so there is no need to work on it. In a year we'll have less than 48 open-close logs under the issue. Next year we'll create a new issue to track this. |
I meant: the issue may be accidentally closed. That's fine. But we shouldn't include "close the issue every week" as part of the workflow.
By having a PR called "20240831 typo fixes" that references the issue, so when you are on the issue you see what was the last typo fix PR and whether it's merged.
I think that's quite arbitrary and we'd rather keep the issue evergreen. |
So you mean to use |
You have a point, but then IMO recycling the issue doesn't provide real benefits over new issues every week, if we are going to have this kind of useless history piling up anyway. The goal was to (a) most importantly, reduce our footprint generated on GitHub (b) change the labels (c) know what the latest status is. None of these really necessitate a recycled issue. |
I've added the following two features:
|
I can see this being the case if the issue becomes very noisy. Aside from opening/closing, linking the same issue from PRs or other issues will also contribute to noise in the issue history. That said, I'm not against trying this PR out and we can course-correct if we don't like it. edit: Onkar's test issue is a good example of what it looks like with a lot of open/closing activity: OnkarRuikar#29 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm +1 on trying this out. There's a comment about line and colnumber in the logs, but it's not blocking.
We are interested in only the opening comment. In a year, we will get less than 12x4 open-close logs. We have way bigger PR pages with tons of review comments. At the beginning of the next year, we'll start fresh by using a new issue number. 😀 |
Recently, we implemented a weekly spelling check bot. It creates an issue every Monday to report typos and new words. We do not wish to flood the repo with such issues over a long period of time.
The PR updates the bot workflow to reuse the existing(35634) issue. The new code does so by updating the issue's opening comment and reopening the issue.
Testing
I tested the code in my fork.