Re: [PATCH] git-stash: fix flag parsing

From: Junio C Hamano <gitster@pobox.com>
Date: Sep 27, 2010, 04:36 UTC
Message-ID: <7vzkv3hkg1.fsf@alter.siamese.dyndns.org>
In-Reply-To: <1285366534-10490-1-git-send-email-brian@gernhardtsoftware.com>

Brian Gernhardt <brian@gernhardtsoftware.com> writes:

Show 9 quoted lines

> Currently git-stash uses `git rev-parse --no-revs -- "$@"` to set its
> FLAGS variable.  This is the same as `FLAGS="-- $@"`.  It should use
> `git rev-parse --no-revs --flags "$@"`, but that eats any "-q" or
> "--quiet" argument.  So move the check for quiet before rev-parse.
>
> Signed-off-by: Brian Gernhardt <brian@gernhardtsoftware.com>
> ---
>
>  Not the most elegant solution, but it works.

Thanks.

rev-parse as a flag parser was useful hack when most of the log family was "rev-list | diff-tree" with various options, but the recent push to make flags consistent across commands inevitably has made it fundamentally impossible for the function to function sensibly in all cases, as it itself has to share some options, like "-q", with others.

After the push of rewriting everything as C builtin, combined with the improvement of parse-options, I have to say that rev-parse flag parser hack has outlived its usefulness. The former means there is less reason to script end-user commands as a "rev-list | diff-tree" pipeline (I am not talking about special purpose statistics commands that use rev-list here), and the latter means that it has become easier for a Porcelain script to parse _its_ command line options and come up with a set of sensible options to underlying rev-list and the command at the downstream of the pipe, without following the original hacky pattern of "we let random options and args and without interpreting them we guess which ones to throw at rev-list and which ones to throw at diff-tree".