librelist archives

« back to archive

process

process

From:
Michael Augustin
Date:
2011-11-30 @ 00:14
Hi all,

I'm very impressed of the idea of hg-review. I read all the
documentation about it, but some aspects are still unclear.

What happens to the changes after they got signed off? Is there any
approve button and removal from the list?

What happens if the changeset is not signoff so it must be changed
again. Is this second commit bound to the first one?

What happens to the changeset in the developers repository if the change
was sign off on the review board but including rebasing or merging?

Must the change also be pushed to a well known repository or is the push
in the review-repo enough?

Kind regards,
Michael

Re: [hg.review] process

From:
Haszlakiewicz, Eric
Date:
2011-12-01 @ 21:51
> -----Original Message-----
> From: hg.review@librelist.com [mailto:hg.review@librelist.com] On
> 
> I'm very impressed of the idea of hg-review. I read all the
> documentation about it, but some aspects are still unclear.
> 
> What happens to the changes after they got signed off? Is there any
> approve button and removal from the list?

Nope.

> What happens if the changeset is not signoff so it must be changed
> again. Is this second commit bound to the first one?

Nope.  It's occurred to me that this would be nice, but it's definitely 
not implemented.

> What happens to the changeset in the developers repository if the
> change
> was sign off on the review board but including rebasing or merging?
> 
> Must the change also be pushed to a well known repository or is the
> push
> in the review-repo enough?

I'm not sure what you mean here.  If a source code changeset never makes 
it out of a developer's clone then no-one else will be able to see it to 
review it.  Even if someone else clones the developer's clone and pushes 
the hg-review comments up to a central hg-review repository, I don't think
you'll be able to see those comments if your source code clone is based on
a central repository that doesn't have the changeset that got reviewed.

Rebasing will probably break the connection between the review comments 
and the source code changeset.  In order for things to work the changes 
you're reviewing need to be "published" with a stable changeset id.

eric