From: Jeff King <peff@peff.net>
Date: Sun, 08 Jun 2014 12:09:36 GMT
Subject: Re: [PATCH v3 1/4] replace: add --graft option
Message-ID: <20140608120936.GB21827@sigill.intra.peff.net>
In-Reply-To: <20140608120439.GA21827@sigill.intra.peff.net>

On Sun, Jun 08, 2014 at 08:04:39AM -0400, Jeff King wrote:

> diff --git a/builtin/blame.c b/builtin/blame.c
> index a52a279..1945ea4 100644
> --- a/builtin/blame.c
> +++ b/builtin/blame.c
> @@ -2313,7 +2313,7 @@ static struct commit *fake_working_tree_commit(struct diff_options *opt,
>  		    ident, ident, path,
>  		    (!contents_from ? path :
>  		     (!strcmp(contents_from, "-") ? "standard input" : contents_from)));
> -	commit->buffer = strbuf_detach(&msg, NULL);
> +	set_commit_buffer(commit, strbuf_detach(&msg, NULL), msg.len);

Side note: this is wrong, as the fake commit created by blame here does
not have its "index" field set. This is a bug waiting to happen the
first time somebody uses a slab in builtin/blame.c. It looks like
merge-recursive does the same thing in make_virtual_commit.

We probably want to provide a function to allocate a commit, including
the index, and use it consistently.

I'll try to work up a series doing that, and the fuller slab API I
mentioned in the previous message, but probably not until tomorrow.

-Peff