[U-Boot] [PATCH] cmd_nvedit.c: clean up with checkpatch

[Macpaul Lin]

HI all,

2011/4/15 Mike Frysinger <vapier at gentoo.org>:
>> > up to Wolfgang how he feels about ifdef indentation
>>
>> In this specific case of #ifdef indentation I feel that the original
>> form (which causes checkpatch warnings) is actually easier to read, so
>> I tend to keep it.  But I am aware that this is inconsequent as we ask
>> for "indentation by TAB only" everywhere else.

For this specific #ifdef case, I will agree to keep the origin format,
it's exactly easier to read.
Because I'm not in the office, I can only send the fixed patch v2 until 4/18.

> well, this is a case where i would say a soft rule is OK -- i.e. allow both
> options and let the active maintainer of the code in question decide.  i too
> prefer the current style as i find it easier to read, but if someone
> maintaining code i never work on wants to be strict about tabs, then it's no
> sweat off my back.
>
> so, not to dupe another thread, but i'd say "use common sense".  but that
> probably too isn't consistent/clear enough for many people.
> -mike

Yes, use common sense is really too lose for many people.

Checkpatch is really help for people to maintain the consistence for
coding style and also avoid the coding method which might lead logic
failure.
However, the drawback is that we still have effort to think about if
the warning should be correct up according to every case that has been
 reported by checkpatch.

For example, 80 charecters is the most often case. Sometimes the code
is really easier to be read in the single one line. Sometimes the
complicated code that should be execute in one line is really hard to
be modified into short format to avoid exceeding the 80 charecters
lenght rule. Although most of time people can decide which format is
better for human reading than checkpatch's parsing result.
We still have effort to inform the patch commiters to modified the
code and discuss with it.

Also, the coding style clean up for the old code already exist in git
repository will help the new contributers and reduce the disscusion
effort passively.
That's why I think we can do clean up for old code slowly when someone
have time.

Maybe list a exception form and some soft rules on the web page is
sometimes help to the contributers just new to the u-boot project.

I have an idea to recruit some colledgue students locally as
volunteers to clean up the old code according to "checkpatch". They
just do code clean up. By doing this activity will also make them be
familiar with the open source projects. At the same time they can also
be familiar with git CVS. Of course a guideline for such code clean up
activity must be work out first.

Don't know what do you think of it?

Thanks.

-- 
Best regards,
Macpaul Lin