Pull request workflow - changes post review and multiple heads

[J Pai]

Hi Sean,

Yes, this is hosted on bitbucket which is where we do the pull request
reviews.

Is the changeset evolution discussed/explained somewhere, so that I can
follow it? Would you also know if this is something that would available in
the near future?

-Jaikiran

On Tuesday, July 7, 2015, Sean Farley <sean at farley.io> wrote:
>
> Jaikiran Pai <jai.forums2013 at gmail.com> writes:
>
>> Hello everyone,
>>
>> We have a workflow within our team which is as follows:
>>
>> - There's an "upstream" repo "foo" which is shared by the team. Only
>> select members have push access to it
>> - There are N developers in the team, each one working against a forked
>> copy of the upstream repo, under their own account.
>> - Each of them commit to their own repos and issue a pull request once
>> they have their feature/bug-fix ready for review.
>> - A reviewer goes through the commit "A" in the pull request and
>> suggests any changes that are needed
>> - The developer then fixes those issues in the pull request and adds a
>> new commit "B" to the pull request.
>>
>> What we ideally want to do is "fold" A and B into one commit and have it
>> pushed to the "upstream". It's straightforward to ask the developer to
>> fold those 2 commits and *force* push the changesets to this own repo
>> (which by the way isn't shared). The force push is needed since the fold
>> ends up creating a new head. Of course, the (redundant) head can then be
>> manually closed via a "hg commit --close-branch" but is there a better
>> way of dealing with this? Perhaps auto-closing the redundant head while
>> dealing with the fold?
>>
>> The end goal we are trying to achieve is having the commits from most of
>> the developers come through via pull requests and once the changes are
>> reviewed and approved, have them committed as one (in most cases)
>> logical commits.
>
> This is one of the goals of changeset evolution, so in a future world,
> this use-case will be taken care of. Out of curiosity, is this repo on
> Bitbucket?
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mercurial-scm.org/pipermail/mercurial/attachments/20150707/6f30f174/attachment-0002.html>