Re: Lessons from commit fest - Mailing list pgsql-hackers

From NikhilS
Subject Re: Lessons from commit fest
Date
Msg-id d3c4af540804152336j481dbaamb063192c943a2f6c@mail.gmail.com
Whole thread Raw
In response to Re: Lessons from commit fest  ("Brendan Jurd" <direvus@gmail.com>)
Responses Re: Lessons from commit fest
List pgsql-hackers
Hi,

>  The idea that we "fix" stylistic issues on the fly is not sustainable.
>  We should offer help and mentorship to new patch submitters in all
>  areas (including stylistic) but they should do the work. It is the only
>  way we will mold them to submit patches in the proper way.
>

I agree.  As a submitter I would much rather get an email saying e.g.
"Hey, your patch is nice but the code style sticks out like a sore
thumb.  Please adopt surrounding naming convention and fix your
indentation per the rules at [link]." than have it fixed silently on
its way to being committed.

With the former I learn something and get to improve my own work.
With the latter, my next patch is probably going to have the exact
same problem, which is in the long term just making extra work for the
reviewers.

I think, us patch-submitters should be asked to do a run of pg_indent on the files that we have modified. That should take care of atleast the indentation related issues. I looked at the README of src/tools/pgindent, and it should be easy to run enough (or is it not?). Only one thing that caught my eye was:

1) Build the source tree with _debug_ symbols and all possible configure options

Can the above point be elaborated further? What all typical and possible configure options should be used to get a clean and complete pg_indent run?

And I think adopting surrounding naming, commeting, coding conventions should come naturally as it can aide in copy-pasting too :)

Regards,
Nikhils
--
EnterpriseDB http://www.enterprisedb.com

pgsql-hackers by date:

Previous
From: Decibel!
Date:
Subject: Re: Lessons from commit fest
Next
From: tomas@tuxteam.de
Date:
Subject: Re: pulling libpqtypes from queue