Jacob Kopczynski

That's an improvement but I'd say it's ~20% as good as the requested feature. It is SOP in some of my organizations to add 3-10 people as reviewers effectively as...

https://gerrit-review.googlesource.com/Documentation/rest-api-changes.html#get-assignee and https://gerrit-review.googlesource.com/Documentation/rest-api-changes.html#get-past-assignees are the relevant REST endpoints. https://chromium-review.googlesource.com/q/project:chromiumos%252Fplatform%252Fec is a project which uses this workflow frequently

https://chromium-review.googlesource.com/q/is:assigned has all Chromium/ChromiumOS CLs which currently have an assignee.

I _could_ do that, but it would be dumb, because I _do_ actually need the objects to be valid.

That's not actually relevant. It doesn't matter that there's a workaround for the undocumented requirements. The problem is that the requirements are undocumented.

This issue is still present.

``` before add_command => sub { my ($self, $command) = @_; die if !$self->collecting_row(); }; ``` Relevant snippet, from [here](https://github.com/jsnell/terra-mystica/blob/52947e7c6a6e419e04ed9556cf4d3f20f5605c88/src/ledger.pm#L71)

I, too, would appreciate this feature. I did something hacky with `docker commit` recently, and being able to check how exactly the files changed in `dive ` would be very...

No test failures seem to be a result of anything in this PR. They fail on main.

I don't think it's reasonable to call this an edge case.