Editorial bot cannot work across review and pre-review threads

[RichardLitt]

On the closed pre-review thread, I attempted to remove a reviewer. Editorialbot said it succeeded, but that reviewer was still on the newly opened review thread.

prereview
review

I subsequently removed them again on the review thread. However, there is a bug here: EditorialBot shouldn't have told me success in the prereview thread, or it should have edited the review thread too.

This may not come up much. I am just testing the system because I'm not familiar with it yet.

[danielskatz]

Once the review issue is opened, any changes in the pre-review issue only affect that closed issue

[RichardLitt]

Sure. But I didn't know that, and editorial bot didn't give me a warning about it, and then said that it had done the job - but it hadn't, because it only has a scoped ability to work in that thread. The message "Removed as reviewer" is inaccurate.

[danielskatz]

I guess it's not inaccurate to me, but because I have a different mental model of how JOSS's infrastructure works with GitHub. So we could better illustrate the model, by saying what the effect of this is when there's already a review issue open, or just disallow this command in that situation. I'm not sure what we would be best, and I guess I'm not a good judge either. I'm open to whatever fix you think is best.

[RichardLitt]

Either would be good solutions. I think a response saying "All commands are disabled as this thread is currently closed" would work, but it may run into issues if for any reason the thread is reopened. Does that ever happen in JOSS (if so, a warning reminding the user that another thread is already opened may be smart)?

[danielskatz]

I don't think a pre-preview issue is ever intentionally used after a review issue is opened.

[RichardLitt]

Cool.

Just to say: this is not an urgent bug. I'm just trying to make it easier for the next reviewer who tries this, four years down the line.