Re: [PATCH] git-rev-list: fix --header

[Junio C Hamano <junkio@cox.net>]

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

>> Wouldn't this be simpler and do the same thing, I wonder?  The
>> point being that "--pretty --header" and "--header --pretty"
>> traditionally did not make --header to override --pretty.
>
> I thought, why not fix that bug, too? After all, it is counterintuitive 
> what "--header --pretty" does, and it was easy to fix.

I checked with 0.99.9m and both "--pretty --header" and
"--header --pretty" gives preference to --pretty.  I think your
patch changes it to favor whichever comes later.

I thought it could be considered a bug to accept --header and
--pretty at the same time without complaining, but if you want
to forbid it, you could error out.  However, that might break
existing Porcelains, and that's why I suggested to keep the
traditional "--pretty wins over --header" behaviour.

If gitk were the only Porcelain we care about that uses
--header, it would make more sense to change the rule to
"--header trumps --pretty" and "gitk --pretty" would magically
start working.

Still undecided.  As you say it is an easy change, so I'd rather
leave the behaviour as before for now.