Re: git bug report

[Junio C Hamano <gitster@pobox.com>]

Paul Watson <pwatson2@wellmed.net> writes:

> 9:43:55.45 2023-08-04  C:\src\t\scripts>TYPE .\t1.txt
> this is t1.txt

> 9:43:57.92 2023-08-04  C:\src\t\scripts>TYPE .\t2.txt
> this is t2.txt
>
> 9:43:58.04 2023-08-04  C:\src\t\scripts>"C:\Program Files\Git\cmd\git.exe" diff --exit-code --no-index  --ignore-all-space  --shortstat  .\t1.txt .\t2.txt
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> 9:43:58.14 2023-08-04  C:\src\t\scripts>ECHO %ERRORLEVEL%
> 0

This is not specific to Windows port and it can be reproduced on a
random Linux box.

    $ echo one >1
    $ echo two >2
    $ git diff --no-index --shortstat 1 2; echo "<<$?>>"
     1 file changed, 1 insertion(+), 1 deletion(-)
    <<1>>
    $ git diff --no-index --shortstat -w 1 2; echo "<<$?>>"
     1 file changed, 1 insertion(+), 1 deletion(-)
    <<0>>

Note that I omitted "--exit-code" in the above reproduction, as it
is always used in the no-index mode.  There seems to be interaction
with "-w" and not using "-p", as this is not limited to "--shortstat".

    $ git diff --no-index -w --stat 1 2; echo "<<$?>>"
     1 => 2 | 2 +-
     1 file changed, 1 insertion(+), 1 deletion(-)
    <<0>>

    $ git diff --no-index -w --numstat 1 2; echo "<<$?>>"
    1	1	1 => 2
    <<0>>

All of the above that exits with 0 status will exit with 1 when -p
is added to the command line.

Also, this is not limited to the no-index mode.

    $ echo one >1
    $ git add 1
    $ echo two >1
    $ git diff --exit-code --numstat 1; echo "<<$?>>"
    1	1	1
    <<1>>
    $ git diff --exit-code --numstat -w 1; echo "<<$?>>"
    1	1	1
    <<0>>

So the minimum reproduction seems to be

  * the diff machinery is asked to do --exit-code (no-index
    implicitly does it)
  * -w is used
  * -p is *not* used
  * to compare two different files.

Thanks for a bug report.

Patches welcome ;-)