Re: [BUG] Unicode filenames handling in `git log --stat`

From: Torsten Bögershausen <tboegi@web.de>
Date: Aug 10, 2022, 08:40 UTC
Message-ID: <20220810084017.gnnodcbt5lyibbf6@tb-raspi4>
In-Reply-To: <xmqqsfm4prqk.fsf@gitster.g>

On Tue, Aug 09, 2022 at 10:55:31PM -0700, Junio C Hamano wrote:

Show 14 quoted lines

> Calvin Wan <calvinwan@google.com> writes:
>
> > Hi Alexander,
> >
> > Thank you for the report! I attempted to reproduce with the steps you
> > provided, but was unable to do so. What commands would I have to run
> > on a clean git repository to reproduce this?
>
> Sounds like a symptom observable when the width computed by
> utf8.c::git_gcwidth(), using the width table imported from
> unicode.org, and the width the terminal thinks each of the displayed
> character has, do not match (e.g. seen when ambiguous characters are
> involved, https://unicode.org/reports/tr11/#Ambiguous).
>

I am not fully sure about that - I can reproduce it with Latin based file names as well:

 git log --stat
[snip]
 Arger.txt  | 1 +
 Ärger.txt | 1 +
   2 files changed, 2 insertions(+)

From this very first experiment I would suspect that we use strlen() somewhere rather then utf8.c::git_gcwidth()

More digging needed (but I don't promise anything today)