Horton

Re: unified vs context diffs (was Re: Strange Windows problem, lock_timeout test request)

Robert Haas <robertmhaas@gmail.com>

View thread View on postgr.es
From: Robert Haas <robertmhaas@gmail.com>
To: Stephen Frost <sfrost@snowman.net>
Cc: Claudio Freire <klaussfreire@gmail.com>, Heikki Linnakangas <hlinnakangas@vmware.com>, Andrew Dunstan <andrew@dunslane.net>, Boszormenyi Zoltan <zb@cybertec.at>, Hari Babu <haribabu.kommi@huawei.com>, Craig Ringer <craig@2ndquadrant.com>, Hans-Jürgen Schönig <hs@cybertec.at>, Ants Aasma <ants@cybertec.at>, Tom Lane <tgl@sss.pgh.pa.us>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>, Amit kapila <amit.kapila@huawei.com>
Date: 2013-02-25T12:56:40Z
Lists: pgsql-hackers 
On Sun, Feb 24, 2013 at 4:31 PM, Stephen Frost <sfrost@snowman.net> wrote:
> * Claudio Freire (klaussfreire@gmail.com) wrote:
>> > As another point, it's also the very first thing that we document in
>> > http://wiki.postgresql.org/wiki/Reviewing_a_Patch to check for.
>>
>> TBH, that wiki link seems to suggest that *having context* is the
>> point of the requirement (to be able to merge with fuzz).
>
> The PG wiki link states "Is the patch in context diff format?" and
> provides a link to the wikipedia article about *that specific format*.
> There's absolutely zero confusion over what "context diff format" means.

True, but I'm with Heikki: it's a pedantic and unhelpful guideline.
Everyone here who reviews patches regularly knows how to, and probably
does, convert between those formats with regularity.  Making patch
submitters feel badly because they've used the "wrong" format does not
advance the goals of the project.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company