Conversation
anonx3247
force-pushed
the
human
branch
2 times, most recently
from
0b47708 to
79a89a6
Compare
| @@ -0,0 +1,20 @@ | |||
| PRAGMA foreign_keys=OFF;--> statement-breakpoint | |||
| ); | ||
| } | ||
|
|
||
| const publicationRes = await publication.submitReview("human", { |
There was a problem hiding this comment.
What happens if I submit a review after the publication was published?
There was a problem hiding this comment.
Well it works. I wondered what would occur If I gave a strong reject on an already published one, and it stays published.
There was a problem hiding this comment.
Yet we would not want that right?
There was a problem hiding this comment.
Theres a quick fix for that, simply by checking the status before adding the review and updating accrodingly
|
I think we want to have 2 modes of operations. One with human reviews one without it so that things are explicity. So the run command should take a new argument -h to specify that a human review is expected. This should be fed to the RunConfig and the logic of reviews should be addapted 🙏 |
|
Is this ready for re-review? |
Yes, should be ready |
spolu
left a comment
spolu
left a comment
There was a problem hiding this comment.
Let's tweak the config. It's cleaner and allows >1 human reviewer which makes sense.
| .description("Add a human review to a publication") | ||
| .requiredOption("-e, --experiment <experiment>", "Experiment") | ||
| .requiredOption("-c, --content <content>", "Review Content") | ||
| .requiredOption("-g, --grade <grade>", "Review Content") |
There was a problem hiding this comment.
Also we should check the validity of he grade?
| @@ -1,3 +1,4 @@ | |||
| export type RunConfig = { | |||
| reviewers: number; | |||
| humanReview?: boolean; | |||
There was a problem hiding this comment.
For the sake of completeness and more clarity:
reviewers: {
agents: number,
humans: number,
}
|
Last interaction here is my comment right? |
2 participants
This has a DB migration so that the
authorcolumn inreviewsbecomes nullable.#122