Re: [PATCH v7 kunit-next 3/4] kunit: subtests should be indented 4 spaces according to TAP

[shuah <shuah@kernel.org>]

On 3/25/20 4:03 PM, Alan Maguire wrote:
> On Wed, 25 Mar 2020, shuah wrote:
> 
>> On 3/13/20 8:44 AM, Alan Maguire wrote:
>>> Introduce KUNIT_INDENT macro which corresponds to 4-space indentation,
>>> and use it to modify indentation from tab to 4 spaces.
>>>
>>> Suggested-by: Frank Rowand <frowand.list@gmail.com>
>>> Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
>>> Reviewed-by: Frank Rowand <frank.rowand@sony.com>
>>> ---
>>>    include/kunit/test.h                |  7 +++-
>>>    lib/kunit/assert.c                  | 79
>>>    +++++++++++++++++++------------------
>>>    lib/kunit/test.c                    |  6 +--
>>>    tools/testing/kunit/kunit_parser.py | 10 ++---
>>>    4 files changed, 54 insertions(+), 48 deletions(-)
>>>
>>> diff --git a/include/kunit/test.h b/include/kunit/test.h
>>> index f7b2ed4c..d49cdb4 100644
>>> --- a/include/kunit/test.h
>>> +++ b/include/kunit/test.h
>>> @@ -84,6 +84,10 @@ struct kunit_resource {
>>>    /* Size of log associated with test. */
>>>    #define KUNIT_LOG_SIZE	512
>>>    
>>> +/* TAP specifies subtest indentation of 4 spaces. */
>>> +#define KUNIT_INDENT	"    "
>>> +#define KUNIT_INDENT2	KUNIT_INDENT KUNIT_INDENT
>>
>> Sorry for a late comment on this.
>>
>> What's the reason to do it this way? Why wouldn't you define
>> it as 8 spaces long string?
>>
> 
> I could have I suppose; I thought it makes it a bit easier
> to read as above (though it did generate a checkpatch
> warning; I thought readability was more important in this
> case, but I can alter if needed).
>   
Please do. Couple of things. KUNIT_INDENT2 doesn't really
tell me much. Same with KUNIT_INDENT

Please make the names more descriptive. Something along the
lines of

KUNIT_INDENT_4SPACE
KUNIT_INDENT_8SPACE

>> Also can you please make sure to run checkpatch --strict on the
>> patches you send?
>>
> 
> Sure! There were also some other line-too-long warnings
> generated as a result of this patch, but when I fixed those
> checkpatch complained about splitting strings across multiple
> lines.  The only way out was to reduce the amount of information
> in the log messages, which I didn't want to do.  In future I can
> note checkpatch warnings that I couldn't find a way to fix in the
> commit message if that would help?
> 

I understand. This is an error though. I am willing to ignore line-too
long warnings for the most part. I don't like to see errors in general.

thanks,
-- Shuah