Re: the war on trailing whitespace

From: Junio C Hamano <junkio@cox.net>
Date: Feb 28, 2006, 00:10 UTC
Message-ID: <7vbqwswdf7.fsf@assigned-by-dhcp.cox.net>
In-Reply-To: <44038B73.4030004@bigpond.net.au>

Peter Williams <pwil3058@bigpond.net.au> writes:

Show 11 quoted lines

>> This whitespace policy should be at least per-project (people
>> working on both kernel and other things may have legitimate
>> reason to want trailing whitespace in the other project),
>
> I'd be interested to hear these reasons.  My experience is that people
> don't put trailing white space in deliberately (or even tolerate it if
> they notice it) and it's usually there as a side effect of the way
> their text editor works (and that's also the reason that they don't
> usually notice it).  But if my experience is misleading me and there
> are valid reasons for having trailing white space I'm genuinely
> interested in knowing what they are.

For example:

	http://compsoc.dur.ac.uk/whitespace/

Jokes aside, I can imagine people want to keep format=flowed text messages (i.e. not programming language source code) under git control. Maybe pulling and pushing would be the default mode of operation for those people, so what git-apply does would not be in the picture for those people, but who knows.

One way to find it out is to push out a strict one and see who screams ;-), but the point is I am reluctant to make a stricter policy the default, thinking, but not knowing as a fact, that it is good enough for everybody.