Simon Krajewski

Can't we just always use the `__APPLE__` version?

I've disabled the BigInt tests on eval today because tests runtime went from 0.3 seconds to 50+ seconds on my machine. I haven't identified the actual cause yet.

This is pretty funky because `B++` itself isn't legal, and that's what we get after the inlining. What do you actually expect this to generate?

Let's forget about the inlining for now and look at the direct `B++` case: ```haxe abstract X(Int) { public function new(v:Int) { this = v; } } class Main {...

Well, `@:op` isn't involved here. The problem is that we currently don't re-type field access after inlining. I'm not even sure if our architecture allows this because access is generally...

I'm fine with this, though I'm slightly worried about the overall approach here because we're already at 4 permutations with persistent and new. I can't immediately tell how else to...

Oh no...

I'm seeing an error on `readClassPaths` already: ``` Processing Arguments [--cwd,c:/Users/simn/Desktop/Newfolder,-D,display-details,--no-output,--times,-D,macro-times,build.hxml,--display,{"jsonrpc":"2.0","id":2,"method":"server/readClassPaths"}] Defines dce=std,display_details=1,eval=1,haxe=4.3.0-rc.1,haxe3=1,haxe4=1,haxe_ver=4.300,interp=1,macro_times=1,source_header=Generated by Haxe 4.3.0-rc.1+da375d759,sys=1,target.name=eval,target.sys=1,target.threaded=1,target.unicode=1,true=1 Using signature 52ee16b3475adf9153eb4803835ac0a3 Display position: ?: -1--1 1,evl: changed directories: [] 1,evl: Cached 0...

It only happens if you `trace("Hello World")` because then the output gets messed up... x) The other problem is probably related to `--cwd` in some way.

I made it stop after the first compilation if there's a `--display`. That's not very elegant, but if this is what used to happen anyway then it should be fine.