clang-fromat-ignore

clang-fromat-ignore

[Patrick Venture]

I haven't been able to find a good answer on this, but from reading
the man page, there's no built-in mechanism for having clang-format
skip some files.  I wanted to check on this before fixing the script
that calls clang-format in our CI.  I've added a submodule to a
project, and that project doesn't follow our style -- so ignoring the
folder lock, stock, is the right approach.

Patrick

Re: clang-fromat-ignore

[James Feist]

On 08/08/2018 11:25 AM, Patrick Venture wrote:
> I haven't been able to find a good answer on this, but from reading
> the man page, there's no built-in mechanism for having clang-format
> skip some files.  I wanted to check on this before fixing the script
> that calls clang-format in our CI.  I've added a submodule to a

I'm not sure how the CI handles this, but if the clang-format is ran 
before the submodule init, and each build is from a clean tree, 
shouldn't that work?

> project, and that project doesn't follow our style -- so ignoring the
> folder lock, stock, is the right approach.
> 
> Patrick
> 

Re: clang-fromat-ignore

[Patrick Venture]

On Wed, Aug 8, 2018 at 11:43 AM, James Feist
<james.feist@linux.intel.com> wrote:
> On 08/08/2018 11:25 AM, Patrick Venture wrote:
>>
>> I haven't been able to find a good answer on this, but from reading
>> the man page, there's no built-in mechanism for having clang-format
>> skip some files.  I wanted to check on this before fixing the script
>> that calls clang-format in our CI.  I've added a submodule to a
>
>
> I'm not sure how the CI handles this, but if the clang-format is ran before
> the submodule init, and each build is from a clean tree, shouldn't that
> work?

That must be my local CI running issue - that I've initialized it.
This is going to make running locally more difficult, but thanks --
that indicates it'll be fine.
https://gerrit.openbmc-project.xyz/#/c/openbmc/phosphor-ipmi-flash/+/11859/
for an example of it fine.

>
>
>> project, and that project doesn't follow our style -- so ignoring the
>> folder lock, stock, is the right approach.
>>
>> Patrick
>>
>