From mboxrd@z Thu Jan  1 00:00:00 1970
From: john at keeping.me.uk (John Keeping)
Date: Wed, 12 Aug 2015 16:02:58 +0100
Subject: [PATCH 0/3] Blame UI
In-Reply-To: <CAHmME9p4KL7Ld+1KD00Le2Z-M7eSQ+3Guiiy9Bg2TxrbhoeBXg@mail.gmail.com>
References: <cover.1439384224.git.john@keeping.me.uk>
 <CAHmME9p4KL7Ld+1KD00Le2Z-M7eSQ+3Guiiy9Bg2TxrbhoeBXg@mail.gmail.com>
Message-ID: <20150812150258.GJ30507@serenity.lan>

On Wed, Aug 12, 2015 at 03:44:19PM +0200, Jason A. Donenfeld wrote:
> On Wed, Aug 12, 2015 at 3:03 PM, John Keeping <john at keeping.me.uk> wrote:
> > This is an attempt at adding "blame" support to CGit.  Because git.git
> > doesn't include the blame implementation in libgit.a, I decided to
> > simply invoke git-blame(1) and parse its porcelain output.
> 
> Could you possibly just do:
> 
> #include "git/builtin/blame.c"
> 
> And then call the static functions?

I was trying to stay to a well defined interface.  It looks like we'd
have to copy reasonable chunks of cmd_blame() and emit_other() in order
to drive the static functions.

I'll give it a go and see what it looks like, but I expect it to result
in more pain during future maintenance than using a separate process.