[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Taler] Coding style clang-format

From: Christian Grothoff
Subject: Re: [Taler] Coding style clang-format
Date: Wed, 24 Apr 2019 20:01:05 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.6.1

On 4/24/19 2:14 PM, address@hidden wrote:
> If you apply this wholesale for git.taler.net,
> make sure that you do NOT apply this for gnurl.
> I generally try to avoid having unnecessary work
> with gnurl. Code reformating would fall under
> "unnecessary work".
> While I do this for other forks I maintain myself,
> gnurl is not like those forks (which were initiated
> for other reasons and don't have upstreams anymore/
> not as active as curl).

Fully agreed, this will certainly not apply to Gnurl or other larger
projects where we have someone else upstream. As Florian mentioned, this
should be at least per-repository, and sometimes even for a particular
repository we may need per directory or even per file policies.

> On the whole topic of clang-format:
> I'm torn between liking it and not liking it.
> The argument Martin brought up that you don't really require llvm
> seems not to be true - you still need to compile llvm to get a
> usable clang-format, at least reading into the build system of it.
> For developers this overhead is okay, but at the same time it's
> annoying that you might have to build llvm until a new release is
> out. 

Sure, but I suspect we can live with a minor annoyance for a while.
At least I've so far not heard anyone provide me with a true blocking
argument on any of the lists. The biggest concern we should have was
raised by Darshit Shah (GNU Wget maintainer) who mentioned that
clang-format can result in slightly different output given different
clang-format versions. So if our hook were to reject pushes too often
because of such smaller differences, that _could_ be a reason to not
enforce this via a Git hook. Anyway, I think we'll soon try this and see
how it goes.

Attachment: signature.asc
Description: OpenPGP digital signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]