Re: [PATCH v7 0/7] convert: add support for different encodings

From: Jeff King <peff@peff.net>
Date: Feb 28, 2018, 13:21 UTC
Message-ID: <20180228132116.GA32272@sigill.intra.peff.net>
In-Reply-To: <20180228082005.GA16857@tor.lan>

On Wed, Feb 28, 2018 at 09:20:05AM +0100, Torsten Bögershausen wrote:

Show 9 quoted lines

> >   2. auto-detect utf-16 (your patch)
> >      - Just Works for existing repositories storing utf-16
> > 
> >      - carries some risk of kicking in when people would like it not to
> >        (e.g., when they really do want a binary patch that can be
> >        applied).
> 
> The binary patch is still supported, but that detail may need some more explanation
> in the commit message. Please see  t4066-diff-encoding.sh

Yeah, but if you don't have binary-patches enabled we'd generate a bogus patch. Which, granted, without that you wouldn't be able to apply the patch either. But somehow it feels funny to me to generate something that _looks_ like a patch but you can't actually apply.

I also think we'd want a plan for this to be used consistently in other diff-like tools. E.g., "git blame" uses textconv for the starting file content, and it would be nice for this to kick in then, too. Ditto for things like grep, pickaxe, etc.

I have some patches that reuse some of the textconv infrastructure for this, which should mostly make it "just work" everywhere. They need a little more polishing before I post them, but you can take a look at:

  https://github.com/peff/git.git jk/textconv-utf16

if you want.

-Peff