Re: [PATCH] scripts/checkpatch.pl: Add warning about leading contination tests

[William Allen Simpson <william.allen.simpson@gmail.com>]

J. Bruce Fields wrote:
> Where does this preference come from?
> 
David Miller -- in response to a patch of mine that used:
  - trailing && on existing lines that already had trailing &&, and
  - leading && on existing lines that already had leading &&, and
  - leading && on new code.

He decided he wants "consistency", existing code be damned.


> In
> 
> 	excessivelylongcondition
> 	&& anotherreallylongcondition
> 	&& yetanotherunbelievablylongcondition
> 	&& yetanotherwellyougettheidea
> 
> I want to be able to keep the &&'s all justified.
> 
Agree with you and Jean Delvare and thousands of other developers.


> Or look for well-typeset math or CS texts and try to find any that leave
> operators dangling on the right.
> 
Agreed.


> I don't really care much about this particular point, but: the
> checkpatch output is already getting too verbose to be useful, without
> adding advice that's actually the opposite of what I'd normally want to
> do....
> 
Yes, you are agreeing with a point Jean raised here, too.

Count me as opposed to this patch.

When I first looked at CodingStyle back in August, one thing that appealed
to me was the laid-back simpler style -- very few, very clear rules.

I'd prefer an addition to CodingStyle clarifying that we should not argue
about this minutiae.