"The author of a revision can not be a reviewer" when author and reviewer are different
Categories
(Conduit :: Phabricator, defect)
Tracking
(Not tracked)
People
(Reporter: tnikkel, Unassigned)
Details
ERR-CONDUIT-CORE: Validation errors:
- The author of a revision can not be a reviewer.
(Run with--tracefor a full exception trace.)
CalledProcessError: Command '['/Users/tim/.mozbuild/moz-phab/arcanist/bin/arc', 'diff', '--base', 'arc:this', '--allow-untracked', '--no-amend', '--no-ansi', '--message-file', '/var/folders/mb/163lw65j4cl9w2g4ywd9l3y40000gn/T/tmp_PnRAx', '--create']' returned non-zero exit status 1
The author of the changeset in question in my hg repo is not me. I set the reviewer to myself however. So I think phab is getting confused and thinking I'm the author of the changeset just because I'm the one running moz-phab.
Pretty sure you could reproduce by creating a rev in your hg repo and changing the author to someone else and setting the reviewer to be yourself.
|
||
Comment 1•6 years ago
|
||
Revisions in HG are different from those in Phabricator.
The changeset author is used only for landing the revision.
Phabricator will treat you as an author as you're uploading it.
|
Reporter | |
Comment 2•6 years ago
|
||
What am I supposed to do? How do I land and review this patch (in a future world where there is no inbound)?
|
|
Reporter | |
Comment 3•6 years ago
|
||
Hello? Where can I get an answer to this issue? If the goal is still to get rid of landings outside of phab/lando then we need to have some sort of answer to this.
|
|
||
Comment 4•6 years ago
|
||
I will bring this up in the meeting tomorrow, but the best if the reviewer wouldn't submit the diff to the Phabricator.
It is against Phabricator's logic.
looks like this is bug 1538151, which indicates this is a phabriactor bug.
i've tentatively marked that bug as blocking the switch to lando to ensure it gets attention.
Description
•