Re: [PATCH] bisect: print abbrev sha1 for first bad commit

From: Jeff King <peff@peff.net>
Date: May 9, 2015, 04:07 UTC
Message-ID: <20150509040704.GA31428@peff.net>
In-Reply-To: <20150509014152.GA31119@tsaunders-iceball.corp.tor1.mozilla.com>

On Fri, May 08, 2015 at 10:03:41PM -0400, Trevor Saunders wrote:

Show 18 quoted lines

> On Fri, May 08, 2015 at 05:29:42PM -0700, Stefan Beller wrote:
> > On Fri, May 8, 2015 at 4:46 PM, Trevor Saunders <tbsaunde@tbsaunde.org> wrote:
> > > its rather silly especially considering the next line contains the
> > > full hash again.
> > 
> > Maybe we can omit it altogether then?
> 
> SO we'd print something like
> 
> the first bad commit is
> Commit abcdefabcdefabcdefabcdefabcdefabcdefabcd
> Author foo@ba.com
> 
> blah blah blah
> 
> ? That seems reasonable to me.  If we're going that far does it also
> make sense to drop printingthe lines about which trees have changed and
> just print the commit message / author / hash?

Yeah, I have always found bisect's output somewhat silly. It prints the "--raw" diff output, which is not incredibly useful. And then to top it off, it does not feed the "--recursive" switch to the diff, so you don't even get to see the real list of changed files.

I suspect the most minimal we could go is:

  git log --format='The first bad commit is %h %s' $bad

and then let the user inspect further from there using the hash. But I think it would also be reasonable to just do a straight "git log -1 $bad" with no with no diff.

(Actually, it looks like all this is generated in bisect.c:show_diff_tree, so it would have to be written in C; but it should be pretty easy to tweak the display options).

-Peff