update-index --really-refresh unsets assume-unchanged bit

From: Clemens Buchacher <drizzd@aon.at>
Date: May 4, 2010, 08:57 UTC
Message-ID: <20100504085722.GA32217@localhost>
In-Reply-To: <20100501092720.GB15941@localhost>

On Sat, May 01, 2010 at 11:27:20AM +0200, Clemens Buchacher wrote:

Show 5 quoted lines

>  --really-refresh::
>  	Like '--refresh', but checks stat information unconditionally,
> -	without regard to the "assume unchanged" setting.
> +	without regard to the "assume unchanged" setting. The "assume
> +	unchanged" bit is unset for all paths.

Scratch that latter part, please. I just noticed the bit is unset only for modified files. If the file matches the index, or even if it has been deleted in the work tree, the bit is _not_ unset.

So the current behavior is quite strange. I see several possible interpretations of --really-refresh:

 Update index for specified paths, disregarding the assume-unchanged bit,

 a) ... and do not ignore work tree changes for tracked files any more.
    I.e., unset the assume-unchanged bit for all files.

 b) ... and do not ignore work tree changes for the specified paths any more.
    I.e., unset assume-unchanged bit only for specified paths.

 c) ... but continue ignoring work tree changes as before.
    I.e., do not unset the assume-unchanged bit for any paths.

(and the current behavior)

 d) ... and do not ignore work tree changes for modified files any more.
    I.e., unset assume-unchanged bit for modified paths.

I believe c) is the most useful, since it allows the user to deactivate the assume-unchanged bit temporarily. All other options potentially change the assume-unchanged bit, which the user may or may not want. In my opinion, it makes no sense to bundle the two operations "update index disregarding the assume-unchanged bit" and "(conditionally) unset assume-unchanged bit."

Clemens