Re: [PATCH] copy.c: make copy_fd preserve meaningful errno

From: Jeff King <peff@peff.net>
Date: Nov 21, 2014, 17:54 UTC
Message-ID: <20141121175446.GA26650@peff.net>
In-Reply-To: <xmqqa93k1m3g.fsf@gitster.dls.corp.google.com>

On Fri, Nov 21, 2014 at 09:48:19AM -0800, Junio C Hamano wrote:

Show 13 quoted lines

> Michael Haggerty <mhagger@alum.mit.edu> writes:
> 
> > On 11/21/2014 10:14 AM, Michael Haggerty wrote:
> >> Couldn't we save ourselves a lot of this "save_errno" boilerplate by
> >> making error() and warning() preserve errno? [...]
> >
> > Never mind; I see that Peff already submitted a patch to this effect.
> 
> My understanding of the conclusion of those four patches was that
> only a single updated one is needed, and "moving save/restore inside
> error()" did not have to survive.
> 
>   http://article.gmane.org/gmane.comp.version-control.git/259911

Yeah, the callsite I had intended ended up needing to save it across more than just error(). And I think that is probably why we have never done any errno-handling inside error() before (it is certainly not the first time I thought of doing such a thing): real-world cases tend to be a little more complicated.

That being said, if this copy() case is one that could benefit, I do not have any problem with picking up (or just reusing the concept of) my 1/4 from that series. It probably does not _hurt_ anything, and if it can help in even a few cases, it may be worth it.

-Peff