Re: [PATCH v11 10/10] convert: add round trip check based on 'core.checkRoundtripEncoding'

From: Junio C Hamano <gitster@pobox.com>
To: Eric Sunshine <sunshine@sunshineco.com>
Cc: "Lars Schneider" <lars.schneider@autodesk.com>,
	"Git List" <git@vger.kernel.org>,
	"Torsten Bögershausen" <tboegi@web.de>,
	"Johannes Sixt" <j6t@kdbg.org>, "Jeff King" <peff@peff.net>,
	"Ramsay Jones" <ramsay@ramsayjones.plus.com>,
	"Johannes Schindelin" <Johannes.Schindelin@gmx.de>,
	"Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>,
	"Lars Schneider" <larsxschneider@gmail.com>
Subject: Re: [PATCH v11 10/10] convert: add round trip check based on 'core.checkRoundtripEncoding'
Date: Fri, 09 Mar 2018 12:22:55 -0800	[thread overview]
Message-ID: <xmqq1sgt578g.fsf@gitster-ct.c.googlers.com> (raw)
In-Reply-To: <CAPig+cTH7wmrBwiyBxr=D1g6dTw65ZRfGPX_ok2PYaMoGJk0Dg@mail.gmail.com> (Eric Sunshine's message of "Fri, 9 Mar 2018 15:18:08 -0500")

Eric Sunshine <sunshine@sunshineco.com> writes:

> On Fri, Mar 9, 2018 at 12:35 PM,  <lars.schneider@autodesk.com> wrote:
>> [...]
>> Add 'core.checkRoundtripEncoding', which contains a comma separated
>> list of encodings, to define for what encodings Git should check the
>> conversion round trip if they are used in the 'working-tree-encoding'
>> attribute.
>> [...]
>> Signed-off-by: Lars Schneider <larsxschneider@gmail.com>
>> ---
>> diff --git a/convert.c b/convert.c
>> @@ -1150,7 +1227,7 @@ static const char *git_path_check_encoding(struct attr_check_item *check)
>>         /* Don't encode to the default encoding */
>> -       if (!strcasecmp(value, default_encoding))
>> +       if (is_encoding_utf8(value) && is_encoding_utf8(default_encoding))
>>                 return NULL;
>
> This change belongs in 6/10, not 10/10, methinks.

It is actually worse than that, no?  When default_encoding is
(somehow) configured not to be UTF-8, e.g. "Shift_JIS", we used to
avoid converting from Shift_JIS to Shift_JIS, but the optimization
no longer happens with this code.

In any case, I think same_encoding() is probably a good thing to use
here at step 6/10, so the point is moot, I guess.