Chen Liang
Chen Liang
These emerge as a result of stitch's handling of `block.bone_block.break` into `BLOCK_BONE_BLOCK_BREAK`. Imo this propagation logic is fine. I propose to keep it as-is, without adding or removing.
My bad! Then imo we should consider this together with #402. In there, I voted to remove duplicate `Block` suffix. Here, I feel I am even more compelled to do...
@aikar The main issue with forge is that it needs gradle to build. Put it in a separate repo?
Hmm, maybe a heap dump is needed to examine what is using the memory
> Why do you want to decompile the client using Enigma? You can use the genSources task from Fabric Loom, it's less memory intensive because it does only decompilation instead...
hmm, the resolution is to add all inner class info to enigma mappings despite that these classes have no mappings.
Agreed, I was just thinking of the same thing and I found this issue on the top of the list :laughing:
Enigma's concern is real as if we have class `A` and `B extends A`, `static void work()` in `A`, then `B.work()` compiles to `invokestatic` on owner `B` name `work` type...
Enigma considers tokens with `$` in its name as broken somehow. Dunno who added that or why that was added.
These imports will be confusing as well, as they include more imports than what's assumed by javadoc