Re: pgindent run?

On January 24, 2018 11:34:07 AM PST, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>Andres Freund <andres(at)anarazel(dot)de> writes:
>> There'd be one or two edge cases of bad formatting, but the
>> end result would be far less painful than what we have today, were
>> basically nobody can format their patches without a lot of manual
>> cherry-picking or large scale unrelated whitespace changes.
>
>IMO, the big problem here is that people commit large changes without
>having pgindent'd them first.

Well, I think it'd really have to be every patch that's indented properly. And that's hard given the way typedefs.list is maintained. Without most new typedefs included one continually reindents with a lot of damage.

It'd be less bad is we automated the maintenance of the lists so
a) patch authors can automatically add to the list and include that in commits
b) the committed list gets updated automatically every few days based on bf results
c) there's automated whole tree pgindent runs every few days.

Andres
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.