Re: [PATCH 1/2] t0027: Tests for core.eol=native, eol=lf, eol=crlf

From: Junio C Hamano <gitster@pobox.com>
Date: Aug 28, 2014, 19:20 UTC
Message-ID: <xmqqha0wa09z.fsf@gitster.dls.corp.google.com>
In-Reply-To: <53EFBC3A.5020805@web.de>

Torsten Bögershausen <tboegi@web.de> writes:

Show 13 quoted lines

> Add test cases for core.eol "native" and "" (unset).
> (MINGW uses CRLF, all other systems LF as native line endings)
>
> Add test cases for the attributes "eol=lf" and "eol=crlf"
>
> Other minor changes:
> - Use the more portable 'tr' instead of 'od -c' to convert '\n' into 'Q'
>   and '\0' into 'N'
> - Style fixes for shell functions according to the coding guide lines
> - Replace "txtbin" with "attr"
>
> Signed-off-by: Torsten Bögershausen <tboegi@web.de>
> ---

It appears that I missed this patch? You seem to have rerolled the corresponding 2/2, to which I responded ($gmane/255507). Is this one still viable/necessary?

If you are doing the whole-sale style fixes for this script, can you also fix the case/esac statement in create_gitattributes? The case arm labels are indented one level too deep, making it harder to spot them than necessary.

Test files "setup master" step creates appear to end in an incomplete line. Is this intended or by mistake? Making sure things work even on files that end in an incomplete line is a good thing, but it looks somewhat strange not to test normal cases (in other words, it makes it appear as if normal cases work OK but incomplete lines cause corner case bugs and these tests are meant to check them, or something).

Thanks.