With feature freeze behind us, I'd like to propose that now is a good
time for a pgindent run. It's possible we'd need another one before
9.5 is branched off from HEAD, but a run now ought to take care of 95%
of the cleanup needed. I see a couple of advantages to doing it now:
1. Patches that are now postponed to 9.6 will need to be rebased against
pgindented sources anyway. Might as well give their authors more time
for that rather than less.
2. Code that matches the project layout conventions is easier to read
and review, IMO (not that I'm especially in love with the pgindent
layout, but I'm used to it). Also any issues like commit d678bde65
would be taken care of, which is an objective reason why reviewing is
easier.
The only significant downside I can think of is that, if we determine
that any of the recent feature additions are so sketchy that they need
to be reverted, having to undo just a portion of the pgindent commit
before reverting the feature commit would be a pain. But I don't think
we should optimize on the assumption that that will happen.
regards, tom lane