+1. Having an enforceable rule is sometimes more important than the rule itself (e.g., 4 vs. 2 spaces indent).
On Thu, Jun 15, 2017 at 9:59 AM, Alexander Rojas <alexan...@mesosphere.io> wrote: > +1 It is always frustrating to rely in clang format to realize it generate > the wrong style, even for old Mesos contributors > > Alexander Rojas > alexan...@mesosphere.io > > > > > > On 15. Jun 2017, at 04:32, Michael Park <mp...@apache.org> wrote: > > > > I'm increasingly hearing that many contributors who want to contribute to > > Mesos find that > > it's often difficult to do so. One of the big issues is due to our > > formatting rules which is not > > automated. As a result, we've had many reviews that are overwhelming in > > style comments > > with only a couple of comments on functionality. > > > > This is very frustrating for contributors, and is also a large burden on > > the committers to > > remember, review and explain the formatting sections of the style guide. > > > > I introduced *ClangFormat* a long time ago as our formatting tool, but it > > was only > > a supplementary tool since it didn't yet conform fully to the style > guide. > > We've done a lot of > > work to narrow this gap and the gap is actually quite small at this > point. > > However, the existence > > of such a gap is enough to stir discussions and render the tool useless > for > > some people. > > > > I think we should close this gap by adopting ClangFormat as our > formatting > > guideline. > > > > I don't have a fully fleshed out plan just yet. I'd like to push for this > > effort again, > > as I find it to be very important. > > > > I'm just seeking for +1s if you'd like to see a fleshed out plan for > this. > > > > Thanks, > > > > MPark > >
