Re: [GNUnet-developers] reindenting the tree, choosing a formater

[Christian Grothoff]

There is already a minor mode for emacs to run uncrustify, you can find
it in Taler's exchange.git/contrib/uncrustify*, so there shouldn't be a
need to write a new extension for emacs.

We should probably copy those files into the gnunet.git as well.

That said, we still do need the Git hook to prevent "bad" commits.

On 9/8/19 1:49 PM, N wrote:
> Okay, the obvious question now is documentation, even if short, at the
> same time while I switch this over. I've spent the morning reading into
> this and for emacs it almost seems like I have to continue working on an
> extension for emacs. for vim there is formatprg (but can it walk a directory
> structure upwards until it finds a named configuration file?).
> 
> I would remove the clang-format entry in .dir-locals.el. But other than
> that, what's the most editor independent way to solve this? pre-commit
> hook?
> 
> Schanzenbach, Martin transcribed 2.7K bytes:
>> This is news to me, too. That being said, I would prefer if the commit would 
>> simply be rejected if not
>> conforming (i.e. if running the formatter results in a different output). I 
>> would hate it if the commited
>> code != my local copy after the push.
>>
>>> On 8. Sep 2019, at 12:25, N <address@hidden> wrote:
>>>
>>> ok. I've never seen the email by florian which christian mentioned
>>> to me. I'm okay with uncrustify.
>>>
>>> Should I wait for more of us to reply or get to it today?
>>>
>>> Schanzenbach, Martin transcribed 2.5K bytes:
>>>> Fine with me. I just tried it. Works great in vim just like clang-format.
>>>>
>>>>> On 8. Sep 2019, at 05:45, Christian Grothoff <address@hidden> wrote:
>>>>>
>>>>> Signed PGP part
>>>>> On 9/7/19 9:00 PM, N wrote:
>>>>>>> Sure, assuming you're talking about using Florian's uncrustify style.
>>>>>>
>>>>>> Oh? I thought we would be using clang-format.
>>>>>> Do you know how much it differs in style? I've used
>>>>>> clang-format in my hook. I can run uncrustify on the
>>>>>> tree.
>>>>>
>>>>> Given that uncrustify looks significantly better -- and also offers some
>>>>> more flexibility (i.e. "leave as is") I think we should just go for
>>>>> uncrustify _exclusively_.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>
>>>
>>>
>>>> _______________________________________________
>>>> GNUnet-developers mailing list
>>>> address@hidden
>>>> https://lists.gnu.org/mailman/listinfo/gnunet-developers
>>>
>>
> 
>

signature.asc
Description: OpenPGP digital signature