I'm glad to hear that there is a desire to improve this situation. However, I want to point out what seems like very flawed reasoning to me:
On Wed, Aug 20, 2014 at 10:43 AM, Kate Stone <katherine_st...@apple.com> wrote: > We shouldn’t be arbitrarily different, and where we do differ we should be > able to describe the specific rationale (as I’m sure Greg will gladly do > when it comes to line length and naming conventions.) Unfortunately, if you're going to differ whenever you have a specific rationale, it undermines the benefit of matching the LLVM style at all. If you think the LLVM style is bad (in many places, I think it is bad) then advocate for changing it. But advocate for changing it *from the position of conforming to it, even if not everything is precisely to your liking*. Much of it is not to my liking, but I write my code that way to be consistent. If you always reserve the right to diverge because of a reason (rather than change the standards when they are flawed), then it is isn't a coding standard at all, it's a set of coding suggestions to be ignored at will. Having LLVM's style, and LLVM's-slightly-modified-style-with-28-LLDB-specific-tweaks-I-have-to-lookup-or-remember is not a substantial improvement IMO over having two totally divergent styles. -Chandler
_______________________________________________ lldb-dev mailing list lldb-dev@cs.uiuc.edu http://lists.cs.uiuc.edu/mailman/listinfo/lldb-dev
