[MPlayer-dev-eng] [bikeshed] Coding style

[Ivan Kalvachev]

On 6/17/08, Luca Barbato <lu_zero at gentoo.org> wrote:
> Michael Niedermayer wrote:
>> On Tue, Jun 17, 2008 at 10:57:13AM +0200, Diego Biurrun wrote:
>>> On Tue, Jun 17, 2008 at 05:35:34AM +0200, Michael Niedermayer wrote:
>>>> On Tue, Jun 17, 2008 at 12:39:04AM +0200, Diego Biurrun wrote:
>>>>> On Mon, Jun 16, 2008 at 08:39:33PM +0200, Benjamin Zores wrote:
>>>>>> Uoti Urpala a écrit :
>>>>>>> On Mon, 2008-06-16 at 13:45 +0200, Luca Barbato wrote:
>>>>>>>> Given that some people seems to be fine with the cleanup weekend
>>>>>>>> here
>>>>>>>> something to start a nice flame:
>>>>>>>>
>>>>>>>> We could decide a coding style and try to be consistent as much as
>>>>>>>> possible.
>>>>>>>>
>>>>>>>> I like trying to leave spaces when possible but I like better having
>>>>>>>>
>>>>>>>> *pointer, conditional blocks should keep the bracket in the same
>>>>>>>> line,
>>>>>>>> functions the line immediately after.
>>>>>>> Standard K&R style as generated by the -kr option of the indent
>>>>>>> program,
>>>>>>> without any "personal" variations. Some deviations can be allowed but
>>>>>>> they should be viewed as that, deviations from the recommended style.
>>>>>>> Trying to create an MPlayer-specific style is a bad idea which will
>>>>>>> cause unnecessary arguments without any real benefit and with real
>>>>>>> harm
>>>>>>> such as making the style less obvious to people not familiar with
>>>>>>> MPlayer.
>>>>>> I tend to agree on that.
>>>>>> We might choose one well-known coding style because, obviously,
>>>>>> there is no chance that everyone agree on something to define an
>>>>>> MPlayer
>>>>>> one.
>>>>>>
>>>>>> Sure, some might not like the chosen one but it's a matter of tradeoff
>>>>>> and can't be worse that the current situation.
>>>>> Seconded (or rather thirded).
>>>> I object, we should agree on the style most developers like best. The
>>>> style
>>>> we would choose is what people here will have to live with for quite
>>>> some
>>>> time. If a overwhelming majority dislikes some detail of K&R it would be
>>>> simply wrong to ignore that.
>>>> Iam not saying iam against indent -kr style, iam just saying that it is
>>>> very wrong to not vote on each issue individually. But rather to bundle
>>>> them up in what the person in charge would prefer to see ...
>>> These two paragraphs contain so many (unnecessarily) loaded statements
>>> that it becomes quite hard not to flame in response.
>>
>> If one considers every disagreeing statement, loaded ...
>>
>>
>>> Anyway, this thread is marked [bikeshed] for a good reason.  Voting on
>>> each issue individually is a recipe for endless bikeshedding.
>>
>>> This is
>>> not a good idea.
>>
>> It is a good idea to vote on the exact details of the style that we will
>> have
>> to live with in the foreseeable future.
>>
>>
>>> This is necessarily about tradeoffs and compromises.
>>
>> Yes, compromisses and tradeoffs between the developers, not the compromise
>> of everyone accepting whatever style uoti picked.
>>
>
> Uoti has his idea, I have another, Michael has another different from
> the former 2.
>
> I guess the best common ground is entertain the bonfire and come up with
> a patch for indent =)

How about do the minimum.
Aka
remove all whitespaces from end of line,
replace all tabs with spaces
and do 4 space indent.

Leave the rest as it is.


Doing this may not even break patches,
as there is option "-l" to ignore whitespace changes.
It should work as long as we don't move stuff up and down.