Re: [PATCH v0 2/3] t/t3400-rebase.sh: add more tests to help migrating git-rebase.sh to C

[Nguyen Thai Ngoc Duy <pclouds@gmail.com>]

2009/5/21 Jakub Narebski <jnareb@gmail.com>:
> Nguyễn Thái Ngọc Duy    <pclouds@gmail.com> writes:
>
>> These new tests make sure I don't miss any check being performed before
>> rebase is proceeded (which is well tested by other tests)
>>
>> Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
>> ---
>>  t/t3400-rebase.sh |   28 ++++++++++++++++++++++++++++
>>  1 files changed, 28 insertions(+), 0 deletions(-)
>>
>> diff --git a/t/t3400-rebase.sh b/t/t3400-rebase.sh
>> index 6e391a3..37f86ab 100755
>> --- a/t/t3400-rebase.sh
>> +++ b/t/t3400-rebase.sh
>> @@ -41,9 +41,37 @@ test_expect_success \
>>       git tag topic
>>  '
>>
>> +test_expect_success 'rebase on dirty worktree' '
>> +     echo dirty >> A &&
>> +     ! git rebase master'
>
> Shouldn't you use test_must_fail instead? From t/test-lib.sh
> (paraphrasing):
>
>  Writing this as "! git rebase master" is wrong, because
>  the failure could be due to a segv.  We want a controlled failure.

Right. Shouldn't we have another patch to fix this once and for all? I
did "grep -F '! git'" and find a few places applicable too.

>> +
>>  test_expect_success \
>>      'the rebase operation should not have destroyed author information' \
>>      '! (git log | grep "Author:" | grep "<>")'
>
> Errrr... what?  Why git-log and not git-cat-file?  Why grep twice?
> Additionally you do not check here that author is unchanged, only that
> is not destroyed.

Errr.. this is not from me. No idea why.
-- 
Duy