[PATCH] D33589: clang-format: consider not splitting tokens in optimization

[Manuel Klimek via Phabricator via cfe-commits]

klimek added a comment.

In https://reviews.llvm.org/D33589#920160, @Typz wrote:

> ping ?


I'm working on understanding this better :) I've refactored the code a bit so I could fully understand the problem, which I now do (sorry for this taking a while, but it took me multiple hours to work through this). I'm now convinced that you're right that we need the reflow/breaking logic to make the decision about the trade-offs, as we need the penalty for the potentially reflown code, as opposed to the penalty for the code if we do nothing.
Given that, my current gut feeling is that we'll want to go a bit further in that direction and make the change more local in breakProtrudingToken. I'll give that a spin and report back with what I find.


https://reviews.llvm.org/D33589