- bring fixes to 1.16
- Fix for all lighting where it only set sky lighting if not present, even if it was trying to set block lighting (unlikely to have caused issues as block lighting seems always to be present, but wrong nonetheless)
First noticed incident of operations ruining ChunkSections. Do not build and use this unless you're testing.
Rushed some of the changes, gotta sleep. Would be nice to get a review of this one from @mattbdev and @dordsor21
This commit will allow the branch to build properly but the plugin will not function properly due to the lack of a proper adapter implementation. Proceeding will require the implementation of the SideEffects system from upstream (865c3a24d2 (diff-8fd33296e427c87d0296ad7f3ccc050a)).
* Start work on lighting engine (does not build)
* Implement getLighting
* Setting, flushing light etc works. Getting light should be working..?
* Better queue/chunk handling
* Use correct location for lighting update
* Correct set location, remove debug
* cleanup a little
* Fix fixlight
* Apply to all versions for the numpties
* Remove lighting extent if not using
* Actually bitmask blocks when setting in chunks
* Initialise Maps and Dequeues with inital size
* format
* Documentation maybe
- Should ensure the server doesn't tick the chunksection while we're editing it
- Basically using the old issue with FAWE breaking block ticking by setting this to zero :)
- Revert back to default FAWE behaviour where placed blocks do not tick by default (until a restart, or another edit is done in the same chunksection without changing the block)
* Increase performance slightly when trimming.
If the chunk section is all one blocks (common in plotworlds) it'll be a nice little boost.
* Cache whether blocks are ticking or not. Greatly reduces the time required to create a palette
* collapse 5 lines to 2.
* Also apply to 14 and 15 for the numpties
* Cleanup
Actually ignore the exception - remove my debug print.
Remove double semi-colon
* Apparently 1.14/15 matter too still.
* Fix FAWE killing chunk ticks. Also allow ticking for set blocks.
* Reduce for-loop to being one loop in createPalette
* set the value in the set array and break so we don't continue onwards in the set's ordinal switch statement
- Re-read chunksections the first time they're "loaded" for an operation
- Reset the chunksection if there are block changes when setting blocks via reflection
- These are maybe gonna affect performance a bit? Idk. Seems to be alright for me
- These are maybe gonna make more issues? Yeah maybe, but I couldn't find any
I run a fork of paper which replaces the visibleChunks and updatingChunks
field for gc performance reasons - visibleChunks is updated via
cloning updatingChunks, and at high chunk counts this causes gc issues
due to the humongous allocation. Unfortunately the only solution is to
not clone the map - which is why the field is removed.
Instead of BukkitAdapter#getPlayerChunk using the visibleChunks field,
it now uses a MethodHandle for PlayerChunkMap#getVisibleChunk. This method is
present on spigot & paper (only protected on spigot - which is why reflection is required),
and I preserve the same thread-safety it provides in my fork - so this solution
will not break compatibility with craftbukkit, spigot, and paper.
Changed Lock to subclass ReentrantLock since that is the only subclass DelegateLock should be using. The lock should also never be null so I added an annotation in the constructor.
I also removed some code and replaced it with PaperLib code to just clean up things a little bit.
Adapter updated to 1.15.2, mostly a clone of 1.15 / 1.15.1's adapter. I encountered no issues from this, but the field names for our reflections may need to be double-checked for accuracy in this Minecraft version.
refactor some region dependencies
swithc masking extent to a filter (so it can be parallelized)
Some unfinished refactoring of the filters
remove piston messages from strings.json
Add regen
Add //history [find|restore|rollback|summary|clear]
- history commands are interactable
- inspect brush info is interactable
Commands are now logged to a searchable database
Fix some cases of id/ordinal mismatch
Lazy tags means tiles/ents are not translated into the nms NBTBase until it is needed. Should be faster in cases where getFullBlock is called, but nbt is not always needed.
Commands like Copy and Paste, where the input/output are both nms worlds, can entirely bypass WorldEdit translating to and from the WorldEdit JNBT classes.
