From: Jeff King <peff@peff.net>
Date: Wed, 28 Mar 2007 22:15:17 GMT
Subject: Re: [PATCH] commit: fix pretty-printing of messages with "\nencoding "
Message-ID: <20070328221517.GA25430@coredump.intra.peff.net>
In-Reply-To: <7v7it1f29s.fsf@assigned-by-dhcp.cox.net>

On Wed, Mar 28, 2007 at 03:10:07PM -0700, Junio C Hamano wrote:

> The function is correct; the only reason it may recode to
> non-utf8 is the user (or Porcelain such as qgit or gitk)
> explicitly asked to do so -- from the final output they will get
> the message in user-native encoding and without the extra
> encoding header, thus we retain the backward compatible
> behaviour before the re-encoding feature was introduced.

I see. But then why do we not simply strip the encoding header entirely,
and return in native format? Why do we insert a new encoding header
_only_ when there was one previously?

-Peff