Re: [PATCH v2] Custom compression levels for objects and packs

From: Dana How <danahow@gmail.com>
Date: May 9, 2007, 06:46 UTC
Message-ID: <56b7f5510705082346m32d3c48dj987fd9b0a6118c10@mail.gmail.com>
In-Reply-To: <alpine.LFD.0.99.0705082031370.24220@xanadu.home>

On 5/8/07, Nicolas Pitre <nico@cam.org> wrote:

Show 10 quoted lines

> On Tue, 8 May 2007, Dana How wrote:
> > On 5/8/07, Nicolas Pitre <nico@cam.org> wrote:
> > > On Tue, 8 May 2007, Junio C Hamano wrote:
> > > If we want the fallback logic to work, at some point we must remember if
> > > the current value is the default or if it is the result of an explicit
> > > config option.
> > I can leave as-is, or use a magic value like -99 and
> > depend on it not colliding with values in zlib.h.
> And where would you set those variables to a sensible default in the
> absence of any config option?

This is why I used the _seen variables, but they are not necessary -- see 3rd option below.

Unfortunately we agreed a day or two ago to use a config rule like used_value = isset(var1) ? var1 : isset(var2) ? var2 : DEFAULT. This doesn't interact well with each variable being processed completely independently in git_config() and the callbacks it calls. The isset() value is "out-of-band"; either store it in the _seen variables, or some special value in used_value .

Which makes the most sense:
* Leave _seen as-is;
* Move pack.compression recognition into config.c which means
  the _seen variables would all be local to config.c;
* Use some special value, and if still present replace it with the default
  at the end of git_config() using extra code;
* Change the config rule to something simpler.

I like the 2nd and the 4th. You didn't like the 4th. Shall I change to the 2nd?

Thanks,

-- 
Dana L. How  danahow@gmail.com  +1 650 804 5991 cell