Closed DarthSharkie closed 5 months ago
Bah, I have to fix how the debug stuff is ignored. Will do that tomorrow. Removing the DebugParticleAnim1
breaks ModItems
; removing ModItems
breaks Apel
. Sigh.
For now, I've removed the .gitignore bits. Need more thought on how to ignore those file/files.
Seems good to me
i think i should merge it
Ignore the
DebugParticleAnim1.java
file. We use that for testing, no need to pollute history with that file.Fix the
setRotation
NPE by extracting the normalization logic for reuse.Fix another NPE in
beforeChildRender
metadata. This one might need a quick discussion, since I have it requiring an explicittrue
value to render. If that key is unset in theEnumMap
, it will not draw by default. I suspect previous testing was setting that in the interceptor being defined. Let me know if we should default to drawing if unset; it's a simple fix.This actually raises a larger question on how we treat that metadata. By mapping, there are messy casts required, null checks before auto-unboxing, and a generally unappealing interface. I propose we switch to nested classes: they can use the same names, but have clearly typed/named fields. The updates can be done in-place on the metadata, and we can limit setters for fields (like the
ServerWorld
) that shouldn't change.Vector3f
s already have a bunch of in-place methods, so it should be reasonably expected these*Data
classes would do the same. If you'd like to see what this looks like, I'll put together another PR in the next day or two.