Re: RLS Design

On 09/20/2014 12:38 AM, Stephen Frost wrote:

> I would not (nor do I feel that I did..) have committed it over a
> specific request to not do so from another committer.  I had been hoping
> that there would be another review coming from somewhere, but there is
> always a trade-off between waiting longer to get a review ahead of a
> commit and having it committed and then available more easily for others
> to work with, review, and generally moving forward.

Y'know what helps with that? Publishing clean git branches for
non-trivial work, rather than just lobbing patches around.

I'm finding the reliance on a patch based workflow increasingly
frustrating for complex work, and wonder if it's time to revisit
introducing a git repo+ref to the commitfest app.

I find the need to find the latest patch on the list, apply it, and fix
it up really frustrating. "git am --3way" helps a lot, but only if the
patch is created with "git format-patch".

Perhaps it's time to look at whether git can do more to help us with the
testing and review process.

-- Craig Ringer                   http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training & Services