www-infrastructure-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jukka Zitting <jukka.zitt...@gmail.com>
Subject Re: @apache.org commit address requirement (Was: Git hosting is go)
Date Thu, 15 Dec 2011 12:47:54 GMT
Hi,

On Thu, Dec 15, 2011 at 10:23 AM, Simon Pepping <sampepping@gmail.com> wrote:
> With a workflow in which one pushes commits unmodified to
> the a.o repo, what is the meaning of the Committer entry?
> It says that someone committed it to some repo and nobody
> rewrote it since then.

Exactly.

> Either it should be rewritten when a commit is committed to
> the a.o repo, or it should be considered as meaningless.

Why? In the ref-updates.log we already have a separate (more reliable)
record of who pushed each commit to the canonical a.o repository.

Consider for example a case of someone submitting a patch to Jira. If
then some other contributor (not a committer of the project) takes
that patch, git commits it, and comments on Jira: "I have tested and
committed the above patch. Please pull the change as commit X from my
repository at Y".

In such a case the %ce field would contain valid information that's
distinct from the %ae field and that would need to be overwritten with
the current pre_receive checks. And should someone then take that
commit X and submit some further improvements based on it, they'd then
need to explicitly rebase their changes if the original commit X gets
rewritten.

So the way I see it, the requirement to overwrite the %ce field of
incoming commits has both social and technical drawbacks. As said in
the other message, it's fine if a project thinks these drawbacks are
acceptable for the benefits mentioned by Paul, but I wouldn't mandate
them on all projects.

BR,

Jukka Zitting

Mime
View raw message