Comment 11 for bug 492145

2009/12/8 Aaron Bentley <email address hidden>:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Martin Pool wrote:
>> Sorry, "subscribe" was inaccurate.  I meant distinguishing "you have
>> been requested to review" because ~bzr-core is asked to review all of
>> them
>
> That isn't really what happens.  ~bzr-core is the review team for
> lp:bzr, which means that it is the *default* reviewer for all proposed
> merges into lp:bzr, but it doesn't mean that it is asked to review all
> branches.  Even when it is, this doesn't cause its members to receive
> any email.
>
> Separately, ~bzr-core is subscribed to code review email on lp:bzr,
> which means that its members receive email about code review involving
> lp:bzr, whether or not they have been requested to review.

So this means they get mail starting "$person has proposed merging ..."?

>> vs "you personally have been requested to review"
>
> This is the only case where this message is sent.

>>, and also
>> distinguishing resubmitted from original mps.
>
> Oh, I thought you were talking about re-requesting a review.

I'm asking that the mail should make clear:

 * who has been asked to review it: you personally, a team to which
you belong, or something else
 * whether this is a totally new mp, or a request for a review on a
new mp, a request for additional review of something already reviewed
by other people, a request for a re-review of something I myself
looked at before, a new diff of an existing mp, or a resubmission
replacing an existing mp
 * any other variables relevant to understanding the context of the review

I am not asserting that all those cases are actually possible in
Launchpad today. However, they are all at least conceivable.

>> If there was a separate cover letter field, we would need to work out
>> how that meshes with the intended commit message.
>
> I don't think they're similar.  They have different formats and target
> audiences.

... but they're covering similar information?

>> Perhaps it should
>> be cast as being supplementary to the commit message and also empty.
>
> I think that the cover letter is essential, while the commit message is
> not.  Why do you think it should be supplementary?

Eventually, if the mp is accepted, it will have a commit message - at
least it will be put into the commit of the merge, even if Launchpad
never knows about it. So at least at that conceptual level, the
commit message is the key thing.

Things appropriate for a cover letter that might not get into the commit message

 * who you already spoke to
 * how it relates to other attempts
 * why in detail it was done this way
 * how you tested it

--
Martin <http://launchpad.net/~mbp/>