On Mon, Oct 31, 2022 at 8:18 AM Jehan-Guillaume de Rorthais
<jgdr@dalibo.com> wrote:
> However, I'm not strictly sure who is responsible to set these statuses. The
> reviewer? The author? The commiter? The CF manager? I bet on the reviewer, but
> it seems weird a random reviewer can reject a patch on its own behalf.
Here's my current understanding (jump in and correct as needed):
Needs Review: If a patch is Waiting on Author, and then the author
responds to the requested feedback and would like additional review,
they can bump the patch back to this state. This can also be done by a
reviewer, or by the CFM, if the author forgets.
Waiting on Author: This is set by a reviewer when they believe a
response is necessary for the process to continue for a patch. Some
people set it immediately upon sending a request; others wait a few
days to keep the administrative overhead down. A CFM might put a patch
into this state if a reviewer forgets and the thread has been hanging
open for a while. (They should probably ping the thread at the same
time.)
Ready for Committer: A reviewer (or a CFM) puts a patch into this
state once they think the patchset is ready. Authors typically should
not put their own patches into this state unless there is already
general agreement on the list that it should be there.
Rejected: This status doesn't actually happen very often due to its
"final" nature. An individual reviewer should usually not decide this
unilaterally; propose rejection and wait for general agreement, or
wait for a CFM or a committer to come along.
Returned with Feedback: A CFM will typically set this at the end of a
CF. An author may preemptively do it as well, to "pause" review for
the entry while they work on it for a future CF.
Moved to Next CF: A CFM does this at the end of a CF, or an author
does it voluntarily.
Withdrawn: An author does this voluntarily to their own entry.
Committed: The committer or CFM does this.
--Jacob