deadalnix

@lemire if times allow, I'll port the D code I wrote to do that. In the meantime, I'll put it there fore reference: https://github.com/snazzy-d/sdc/blob/877b095128a50a3f60376e55774668cabbece76a/src/source/packedfloat.d#L215-L311

@JelleZijlstra I'm not sure that the proposed solution is the right one, but there is clearly a need here. The root problem is IMO that once you reformatted in place,...

I admittedly did not look at the implementation, so i might be wrong, but I'd expect the `--diff` flag to be implemented as follow: 1. Run the formatting algorithm on...

@cooperlees I generally do something similar myself. Nevertheless, I'm not talking about tools we made ourselves, but off the shelf pipeline that companies use. Specifically for `rustfmt`, this can be...

This would definitively need a test case.

There is something I don't get. How is the test causing the track trace generator to throw while generating the stack trace?

Merge conflict resolved.

I rebased on top of the right branch, sorry for the confusion. Is it good to merge? if not, what's missing?

The perf problem can be solved by having the function be perfectly regular function under the hood, and have the runtime lazily duplicate the stack on demand when something actually...

I think you might want to let the HHVM guys know, then. Clearly, you have a lot to teach them.