Re: [PATCH v9 4/8] utf8: add function to detect a missing UTF-16/32 BOM

From: Lars Schneider <larsxschneider@gmail.com>
Date: Mar 6, 2018, 23:00 UTC
Message-ID: <CF78BB76-B42C-4A83-B18C-3FBAB0339346@gmail.com>
In-Reply-To: <xmqq8tb43jfe.fsf@gitster-ct.c.googlers.com>

Show 16 quoted lines

> On 06 Mar 2018, at 23:53, Junio C Hamano <gitster@pobox.com> wrote:
> 
> Lars Schneider <larsxschneider@gmail.com> writes:
> 
>>> Also "UTF16" or other spelling
>>> the platform may support but this code fails to recognise will go
>>> unchecked.
>> 
>> That is true. However, I would assume all iconv implementations use the
>> same encoding names for UTF encodings, no? That means UTF16 would never be
>> valid. Would you agree?
> 
> After seeing "UTF16" and others in "iconv -l | grep -i utf" output,
> I am not sure what you mean by "Would you agree?"  People can say in
> their .gitattributes file that this path is in "UTF16" without dash
> and that is what will be fed to this coe, no?

On macOS I don't see UTF16... but I just checked on my Linux box and there it is. Consequently, we need to check both versions: with and without dash.

I'll reroll ASAP (I try to do it first thing in the morning).

Thanks, Lars