Re: [Yade-dev] clang-format (Was: Re: Removing trailing white space, yay or nay?)

2019-10-22 Thread Anton Gladky
Hi Bruno,

OK, if we have an agreement on that, we will merge MR 298.
Thus the .clang-format will be in the root of the trunk and people
can start to use it.

We will have a test phase (2-3 months), so we can probably find
and fix some issues in .clang-format, update documentation.
Also CI-pipeline can be prepared, but only with Allow-To-Fail option.

If everything is successful - we can then reformat the files, which were not
formatted on-the go during the test phase. Cang-format-step can then
be switched on strictly in the CI-pipeline.

Best regards

Anton

Am Di., 22. Okt. 2019 um 11:38 Uhr schrieb Bruno Chareyre
:
>
> Thanks Anton, it makes sense to me.
>
> On Mon, 21 Oct 2019 at 21:02, Anton Gladky  wrote:
>>
>> If we do a one-shot-reformatting - it is also OK. But I would then prefer
>> to set the author of this commit, something to "clang-formatter",
>> Just to identify, that this particular change was done by this action.
>
>
> Now I'm unsure myself. Maybe for stable files which we rarely touch the 
> "cons" of reformatting are more than the "pros".
> A special commiter is a very good point. And even if it's done per-file it 
> would be better to have a commit by that clang-formatter on the top of actual 
> commit rather than altogether - so the changes don't get hiden.
>
>>
>>
>> For example, my IDE shows at each line, who was the last author
>> of the particular line  (git blame basically). Sometimes it is useful to 
>> contact
>> the author of the line/code personally.
>
>
> Yep, "git blame" will be meaningless. Fortunately that's not the only way to 
> browse history, but yeah, that's the main downside.
>
>>
>> But again, it is up to the majority to decide, whether to use this tool or 
>> not.
>> But the code is getting more readable, more uniform and professional
>
>
> It would be an improvement in my view.
> Cheers
>
> Bruno
> ___
> Mailing list: https://launchpad.net/~yade-dev
> Post to : yade-dev@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~yade-dev
> More help   : https://help.launchpad.net/ListHelp

___
Mailing list: https://launchpad.net/~yade-dev
Post to : yade-dev@lists.launchpad.net
Unsubscribe : https://launchpad.net/~yade-dev
More help   : https://help.launchpad.net/ListHelp


Re: [Yade-dev] clang-format (Was: Re: Removing trailing white space, yay or nay?)

2019-10-22 Thread Bruno Chareyre
Thanks Anton, it makes sense to me.

On Mon, 21 Oct 2019 at 21:02, Anton Gladky  wrote:

> If we do a one-shot-reformatting - it is also OK. But I would then prefer
> to set the author of this commit, something to "clang-formatter",
> Just to identify, that this particular change was done by this action.
>

Now I'm unsure myself. Maybe for stable files which we rarely touch the
"cons" of reformatting are more than the "pros".
A special commiter is a very good point. And even if it's done per-file it
would be better to have a commit by that clang-formatter on the top of
actual commit rather than altogether - so the changes don't get hiden.


>
> For example, my IDE shows at each line, who was the last author
> of the particular line  (git blame basically). Sometimes it is useful to
> contact
> the author of the line/code personally.
>

Yep, "git blame" will be meaningless. Fortunately that's not the only way
to browse history, but yeah, that's the main downside.


> But again, it is up to the majority to decide, whether to use this tool or
> not.
> But the code is getting more readable, more uniform and professional
>

It would be an improvement in my view.
Cheers

Bruno
___
Mailing list: https://launchpad.net/~yade-dev
Post to : yade-dev@lists.launchpad.net
Unsubscribe : https://launchpad.net/~yade-dev
More help   : https://help.launchpad.net/ListHelp