Re: Output from "git blame A..B -- path" for the bottom commit is misleading

From: John Keeping <john@keeping.me.uk>
Date: May 8, 2014, 21:38 UTC
Message-ID: <20140508213806.GA19464@serenity.lan>
In-Reply-To: <xmqq8uqc2dt5.fsf@gitster.dls.corp.google.com>

On Thu, May 08, 2014 at 01:52:38PM -0700, Junio C Hamano wrote:

Show 25 quoted lines

> I am tempted to say "blame that is run without the --porcelain
> option is a end-user facing Porcelain, and people should not be
> reading its output in their scripts" and change the behaviour of the
> "-b" option to instead show something like this instead:
>     
>     ^cc29195 (Unknown        2014-04-18 11:21:43 -0700 103) 
>     7bbc458b (Kyle J. McKay  2014-04-22 04:16:22 -0700 104) test_expect_...
>     ^cc29195 (Unknown        2014-04-18 11:21:43 -0700 105)         test...
>     7bbc458b (Kyle J. McKay  2014-04-22 04:16:22 -0700 106)         git ...
>     ^cc29195 (Unknown        2014-04-18 11:21:43 -0700 107)         test...
> 
> which shows the commit object name, its bottom-ness and the
> timestamp, or even
> 
>              (                                         103) 
>     7bbc458b (Kyle J. McKay  2014-04-22 04:16:22 -0700 104) test_expect_...
>              (                                         105)         test...
>     7bbc458b (Kyle J. McKay  2014-04-22 04:16:22 -0700 106)         git ...
>              (                                         107)         test...
> 
> which does away with the misleading information altogether.
> 
> I myself is leaning towards the latter between the two, and not
> overriding "-b" but introducing another "cleanse the output of
> useless bottom information even more" option.

I'd be tempted to leave the SHA-1 to indicate "sometime before" but delete the author and date (I also think it looks a bit nicer to omit the opening parenthesis):

    ^cc29195                                           103) 
    7bbc458b (Kyle J. McKay  2014-04-22 04:16:22 -0700 104) test_expect_...
    ^cc29195                                           105)         test...
    7bbc458b (Kyle J. McKay  2014-04-22 04:16:22 -0700 106)         git ...
    ^cc29195                                           107)         test...

On a slight tangent, I tried this in a fairly young repository and got this (with master at v2.0.0-rc2-4-g1dc51c6):

$ git blame Makefile | head -5 7a3fc144 (John Keeping 2013-12-26 17:37:53 +0000 1) REL_VERSION = v0.2 5c9829f9 (John Keeping 2013-07-29 17:03:26 +0100 2) 5c9829f9 (John Keeping 2013-07-29 17:03:26 +0100 3) # The default target is... ^f7fae99 (John Keeping 2013-03-24 17:14:40 +0000 4) all:: ^f7fae99 (John Keeping 2013-03-24 17:14:40 +0000 5)

f7fae99 is the initial commit in the repository, so shouldn't the last two lines blame to that, not a non-existent ancestor?