Jonathan Coates

Results 393 comments of Jonathan Coates

![image](https://user-images.githubusercontent.com/4346137/81494318-e0f79a00-929f-11ea-8ad5-a2bc9d2564ba.png) So I ran a the benchmarks over all commits up until master. I love that even commits which shouldn't have changed anything still show up on the benchmarks, which...

I've absolutely no clue - I ran the benchmarks using the same setup as the above graph, and had a massive performance decay which is not in line with what's...

Haven't had a proper look at your changes since Sunday. Sorry, will try to do so this evening when I have more time (and mental bandwidth). --- But just some...

On `master`: ``` # Run complete. Total time: 00:23:40 PerformanceBenchmark.binarytrees thrpt 15 3.977 ? 0.137 ops/s PerformanceBenchmark.fannkuch thrpt 15 5.974 ? 0.225 ops/s PerformanceBenchmark.nbody thrpt 15 0.599 ? 0.024 ops/s...

Obvious disclaimer that I'm not super well-versed in OC internals, so may be misreading things... Looking at RelayPeripheral, it appears that the result of `Component.invoke` is returned immediately without any...

@payonel I'm afraid yes. I suspect you'll need to handle all of the cases in `ScalaClosure.toLuaValue`, as you'll also have arrays inside maps. Sorry! https://github.com/MightyPirates/OpenComputers/blob/e452b1fb5ef8bd13e845b0fe18d3752b9c0a59be/src/main/scala/li/cil/oc/util/ScalaClosure.scala#L48-L54

I'm not aware of some "standard" bundled cable API, but it would be something I'd be interested in seeing - ComputerCraft has the same joys of trying to get every...

I _think_ (afraid I'm more familiar with LuaJ 2's codebase) the problem occurs here: https://github.com/luaj/luaj/blob/daf3da94e3cdba0ac6a289148d7e38bd53d3fe64/src/core/org/luaj/vm2/LuaClosure.java#L570 `errorHook` is the function responsible for calling the xpcall callback. It's being called with `getMessage`...

I've found the [LuaJIT wiki's algorithm section](http://wiki.luajit.org/New-Garbage-Collector#gc-algorithms) to be rather useful too. It's rather sparse on implementation details, but offers a decent description of some algorithms.

I wonder if the more appropriate thing to do would be to do something like #125, and just allow turtles to right click arbitrary blocks. It obviously requires more work...