Re: [PATCH v2 5/6] receive-pack: Wrap status reports inside side-band-64k

From: Shawn O. Pearce <spearce@spearce.org>
Date: Feb 5, 2010, 21:53 UTC
Message-ID: <20100205215344.GC27498@spearce.org>
In-Reply-To: <7vd40j1j2a.fsf@alter.siamese.dyndns.org>

Junio C Hamano <gitster@pobox.com> wrote:

Show 17 quoted lines

> "Shawn O. Pearce" <spearce@spearce.org> writes:
> 
> > If the client requests the side-band-64k protocol capability we
> > now wrap the status report data inside of packets sent to band #1.
> > This permits us to later send additional progress or informational
> > messages down band #2.
> >
> > If side-band-64k was enabled, we always send a final flush packet
> > to let the client know we are done transmitting.
> 
> Two questions.
> 
>  - Why does use_sideband, the variable with the same name as a boolean
>    variable used by other parts of the system to decide whether we should
>    or should not use the sideband communiocation, get a value other than 0
>    or 1?  What is the benefit of using it to keep an actual value?  Does
>    the benefit outweigh the confusion factor?

I was following the existing convention of use_sideband is maximum-packet-size in server code, and boolean in client code.

Well, I do diverage a bit in the client, in the client side
use_sideband = 2 in builtin-fetch-pack.c for side-band-64k and
use_sideband = 1 in builtin-send-pack.c for the same thing.

>  - What happens if client wants only side-band, not 64k?  This is just
>    theoretical and "we don't bother" is a perfectly acceptable answer.  I
>    am just curious ;-).

Why bother? What client doesn't understand side-band-64k but would understand this?

-- 
Shawn.