Use a squash merge in Bors

So I've run into some issues with Github.

If you have any required builds which aren't triggered by the staging branch, Github will disallow the merge. This can be worked around but it can be annoying.

More concernedly, if you have any required reviewers, Github will block the merge. You can use the Bors toml file to require reviewers but Bors doesn't support the CODEOWNERS file. This means that you can't require reviews from certain people or teams.

One possible work around, is to have bors squash the PR in-place. I'm not really a fan of it because I don't think Bors should be modifying a user's branch. The other alternative is to implement CODEOWNERS support in Bors.

Any thoughts?

Neither am I.

This is probably the better option, but I don't think it's necessary for the MVP. It also might be good for people even when they use regular merge mode, improving the error messages that bors emits, so it's really a separate feature.

1 Like

@notriddle

I've created an RFC to support CODEOWNERS files for Bors. Bors support for CODEOWNERS

2 Likes

I've edited your RFC to have a more useful summary.

Also, before I can approved this, could you please use the normal license disclaimer? It intentionally includes both Apache 2 and CC-BY-NC-SA because the former is used for the bors codebase and stuff, while the latter is actually the default license used for content on this forum. It seems kind of silly that you edited them on both of your RFCs, when I'm pretty sure the dual license is an implied fact of the ToS for the site (which says that you're granting a CC-BY-NC-SA license) plus your license disclaimer (which says that you're granting an Apache license). But I don't like ambiguity, so let's make it explicit.

> I allow this RFC document to be modified and redistributed under the terms of the [Apache-2.0 license](http://www.apache.org/licenses/LICENSE-2.0), or the [CC-BY-NC-SA 3.0 license](http://creativecommons.org/licenses/by-nc-sa/3.0/deed.en_US), at your option.
1 Like

Everyone else seems okay with this as-is, so this RFC is entering Final Comment Period. Disposition: accept.

I'd really like you, @MarkRobinson, to fix the license boilerplate before this RFC is accepted. I don't want to be too picky, but it is kind of important that we get everything done by the book.

1 Like

@notriddle Sure, I'll change the license back to the default. In the boiler-plate I thought it was an option (choose A or B).

2 Likes

Yeah, sorry about that. It is an option, but it's an option for the recipient of the RFC, not one for the author. It's a dual license.

1 Like