On 12/27/2016 06:16 PM, Junio C Hamano wrote:
> Michael Haggerty <mhagger@xxxxxxxxxxxx> writes:
>
>> Sorry I didn't notice this earlier, but the `LOCK_REPORT_ON_ERROR`
>> constant introduced by
>>
>> 3f061bf "lockfile: LOCK_REPORT_ON_ERROR", 2016-12-07
>>
>> sets that constant to the value 2,...
>
> Sorry I didn't notice this earlier, either. Thanks for spotting.
>
> -- >8 --
> From: Junio C Hamano <gitster@xxxxxxxxx>
> Date: Tue, 27 Dec 2016 09:12:09 -0800
> Subject: [PATCH] lockfile: move REPORT_ON_ERROR bit elsewhere
>
> There was LOCK_NO_DEREF defined as 2 = 1<<1 with the same value,
> which was missed due to a huge comment block. Deconflict by moving
> the new one to 4 = 1<<2 for now.
The fix is obviously correct. I'm not sure I like it that comments are
being blamed for mistakes :-P
Perhaps defining these constants within an `enum` would make it clearer
that they are a group and help prevent problems like this in the future
(though that could be done later rather than as part of this hot fix).
Michael
> Signed-off-by: Junio C Hamano <gitster@xxxxxxxxx>
> ---
> lockfile.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/lockfile.h b/lockfile.h
> index 16775a7d79..7b715f9e77 100644
> --- a/lockfile.h
> +++ b/lockfile.h
> @@ -137,7 +137,7 @@ struct lock_file {
> * ... this flag can be passed instead to return -1 and give the usual
> * error message upon an error.
> */
> -#define LOCK_REPORT_ON_ERROR 2
> +#define LOCK_REPORT_ON_ERROR 4
>
> /*
> * Usually symbolic links in the destination path are resolved. This
>