Re: [PATCH v2] ci: disallow directional formatting

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Nov 4, 2021, 10:19 UTC
Message-ID: <nycvar.QRO.7.76.6.2111041111570.56@tvgsbejvaqbjf.bet>
In-Reply-To: <xmqqk0ho4xo5.fsf@gitster.g>

Hi Junio,

On Wed, 3 Nov 2021, Junio C Hamano wrote:

Show 14 quoted lines

> "Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com>
> writes:
>
> > +# U+202a..U+2a2e: LRE, RLE, PDF, LRO and RLO
> > +# U+2066..U+2069: LRI, RLI, FSI and PDI
> > +regex='(\u202a|\u202b|\u202c|\u202d|\u202e|\u2066|\u2067|\u2068|\u2069)'
> > +
> > +! git ls-files -z ':(attr:!binary)' |
> > +LC_CTYPE=C xargs -0r git grep -Ele "$(LC_CTYPE=C.UTF-8 printf "$regex")" --
>
> One thing for the future, and one thing for the present.
>
>  - Do some languages we would add to po/ hierarchy in the future
>    possibly want to use these sequences as legitimate contents?

I mulled over that. And I think you're right. If a right-to-left translation needs to refer to, say, a `git` invocation, the part that shows the commandline surely would need to be guarded within directional formatting code points. We currently only have translated messages that read left-to-right, for example we lack Arabic and Hebrew translations. Those would be likely to contain such code points on purpose.

I therefore added `:(exclude)*.po` to the command.

Show 7 quoted lines

>  - Do we need ls-files?
>
>
> For the latter, shouldn't the attribute-based pathspec work just
> fine with "git grep"?  i.e.
>
>     git grep -l -E -e $pattern -- ':(exclude,attr:binary)'

D'oh. You're right! I fixed it.

Ciao, Dscho