From: Junio C Hamano <gitster@pobox.com>
Date: Thu, 13 Sep 2012 17:23:13 GMT
Subject: Re: [PATCH v2] cherry-pick: don't forget -s on failure
Message-ID: <7v8vcdalby.fsf@alter.siamese.dyndns.org>
In-Reply-To: <20120913073324.GA14383@suse.cz>

Miklos Vajna <vmiklos@suse.cz> writes:

> +void append_signoff(struct strbuf *msgbuf, int ignore_footer)
> +{
> +	struct strbuf sob = STRBUF_INIT;
> +	int i;
> +
> +	strbuf_addstr(&sob, sign_off_header);
> +	strbuf_addstr(&sob, fmt_name(getenv("GIT_COMMITTER_NAME"),
> +				getenv("GIT_COMMITTER_EMAIL")));
> +	strbuf_addch(&sob, '\n');
> +	for (i = msgbuf->len - 1 - ignore_footer; i > 0 && msgbuf->buf[i - 1] != '\n'; i--)
> +		; /* do nothing */
> +	if (prefixcmp(msgbuf->buf + i, sob.buf)) {
> +		if (!i || !ends_rfc2822_footer(msgbuf))
> +			strbuf_addch(msgbuf, '\n');
> +		strbuf_addbuf(msgbuf, &sob);
> +	}
> +	strbuf_release(&sob);
> +}

Hrm, what is this thing trying to do?  It does start scanning from
the end (ignoring the "Conflicts:" thing) to see who the last person
that signed it off was, but once it decides that it needs to add a
new sign-off, it still adds it at the very end anyway.