Re: [PATCH] repack: add `repack.honorpackkeep` config var

From: Junio C Hamano <gitster@pobox.com>
Date: Jan 28, 2014, 09:21 UTC
Message-ID: <xmqq8uu0mpg8.fsf@gitster.dls.corp.google.com>
In-Reply-To: <20140128060954.GA26401@sigill.intra.peff.net>

Jeff King <peff@peff.net> writes:

Show 13 quoted lines

> The git-repack command always passes `--honor-pack-keep`
> to pack-objects. This has traditionally been a good thing,
> as we do not want to duplicate those objects in a new pack,
> and we are not going to delete the old pack.
> ...
> Note that this option just disables the pack-objects
> behavior. We still leave packs with a .keep in place, as we
> do not necessarily know that we have duplicated all of their
> objects.
>
> Signed-off-by: Jeff King <peff@peff.net>
> ---
> Intended for the jk/pack-bitmap topic.

Two comments.

 - It seems that this adds a configuration variable that cannot be
   countermanded from the command line. It has to come with either a
   very good justification in the documentation describing why it is
   a bad idea to even allow overriding from the command line in a
   repository that sets it, or a command line option to let the
   users override it. I personally prefer the latter, because that
   will be one less thing for users to remember (i.e. "usually you
   can override the configured default from the command line, but
   this variable cannot be because of these very good reasons").

 - In the context of "pack-objects", the name "--honor-pack-keep"
   makes sense; it is understood that pack-objects will _not_ remove
   kept packfile, so "honoring" can only mean "do not attempt to
   pick objects out of kept packs to add to the pack being
   generated." and there is no room for --no-honor-pack-keep to be
   mistaken as "you canremove the ones marked to be kept after
   saving the still-used objects in it away."

   But does the same name make sense in the context of "repack"?

Thanks.