Re: [PATCH] Don't ignore write failure from git-diff, git-log, etc.

From: Linus Torvalds <torvalds@linux-foundation.org>
Date: May 26, 2007, 16:18 UTC
Message-ID: <alpine.LFD.0.98.0705260910220.26602@woody.linux-foundation.org>
In-Reply-To: <87bqg724gp.fsf@rho.meyering.net>

On Sat, 26 May 2007, Jim Meyering wrote:

Show 5 quoted lines

>
> Each git command should report such a failure.
> Some already do, but with the patch below, they all do, and we
> won't have to rely on code in each command's implementation to
> perform the right incantation.

The patch is wrong.

Some write errors are expected and GOOD.

For example, EPIPE should not be reported. It's normal. The user got bored. It might be hidden by the SIGPIPE killing us, but regardless, reporting it for the normal log/diff thing is just not correct. EPIPE isn't an error, it's a "ok, nobody is listening any more".

Also, PLEASE don't do this:

> +		if (0 <= fcntl(fileno (stdout), F_GETFD)

That's totally unreadable to any normal human.

You don't say "if zero is smaller or equal to X". You say "if X is larger than or equal to zero". Stop messing with peoples minds, dammit!

Anybody who thinks that code like this causes fewer errors is just fooling himself. It causes *more* bugs, because people have a harder time reading it.

Maybe you and Junio have taught yourself bad manners, but you're a tiny tiny part of humanity or the development community. Junio can do it just because while he's just a single person, he's a big part of the git coding base, but anybody else who does it should just be shot.

			Linus