Re: [PATCH v2 03/14] sequencer: lib'ify write_message()

From: Junio C Hamano <gitster@pobox.com>
Date: Aug 29, 2016, 20:27 UTC
Message-ID: <xmqqfupn9v76.fsf@gitster.mtv.corp.google.com>
In-Reply-To: <6cc2e834ae9e595625fbbc6922613994098fd7dc.1472219214.git.johannes.schindelin@gmx.de>

Johannes Schindelin <johannes.schindelin@gmx.de> writes:

Show 14 quoted lines

> Instead of dying there, let the caller high up in the callchain
> notice the error and handle it (by dying, still).
>
> The only caller of write_message(), do_pick_commit() already checks
> the return value and passes it on to its callers, so its caller must
> be already prepared to handle error returns, and with this step, we
> make it notice an error return from this function.
> ...
> @@ -223,7 +226,7 @@ static int fast_forward_to(const unsigned char *to, const unsigned char *from,
>  
>  	read_cache();
>  	if (checkout_fast_forward(from, to, 1))
> -		exit(128); /* the callee should have complained already */
> +		return -1; /* the callee should have complained already */

This hunk does not seem to have anything to do with write_message() conversion, other than that its only caller, do_pick_commit(), also calls write_message(). checkout_fast_forward() itself can die when it cannot write the index, though, so this hunk may deserve to be in its own patch that teaches checkout_fast_forward() to instead return an error if/when its caller desires.

With the updated message, the series has become far easier to review, and the reordering them to sequencer-pick-revisions at the beginning makes quite a lot of sense.