* Register platforms and commands in a more proper way
Platforms are now register "on load"/initialize/etc., and preferreds are
chosen just prior to command registration events. This allows us to
properly hook in to reload. Additionally, we now have the capability to
reload data with /minecraft:reload, though this is not yet implemented.
* Correct javadocs on queryCapability
* Flatten BukkitImplAdapter Lifecycled
* The scramble is complete
* Fix version decl for Bukkit Log4J
* Swap out to a compatibility layer for LogManager
Slightly slower, but compatible across all Java versions
* Piston released with SL4JF removal
* Clean up declarations a bit
* Show a message with install instructions when running the jar file
* Typo
* Fixed checkstyle issue
* Slightly modify message to adapt to the type of install
* Clean up the display code a lot
* Allow line breaks before semi
Co-authored-by: Octavia Togami <octavia.togami@gmail.com>
* Update so many dependencies, merge Forge/Fabric for final
* Clean up contrib docs for Gradle change
* Fix setting compat flags while using toolchain
* Fix deprecation in doc printer
* Restore proper forge JAR name
* Add dist classifier for mod jar
* Properly relocate new bStats
* Fix jar used from fabric
* Fix fabric bom
* Dup the shaded classes for consistency
* Sync Forge/Fabric log4j versions, de-dup
* Downgrade both log4j. This will work
* Update some plugins as well
* Drop the fabric force stuff
* Use duplicate strategy to directly merge jar
* Alter the CUI lifecycle to be more consistent
This handling reduces the platform-specific behaviour of CUI handshakes,
and ensures that the appropriate state is maintained when a client
connection ends but a LocalSession remains cached.
* Apply review suggestions
* fabric: Update to new networking API
* session: Correct field grouping for cui attempts
* Move to packaged translations
* License that file
* Fix some checkstyle too
* Support reloading translation files
* Add missing dash for unpack directory
* Mark unpacked zips as owner-only
* Added a method to check whether a block can be placed at a location
* Make the snow system only place at locations that it can by game logic
* Make canPlaceAt a default method
* Fixed PR comments
* Use a World field
* Just throw directly
* IAE makes more sense
* Remove unused import
* Check for editsession without a world
* Fixed an issue where upwards resolved schematics would break schem list
* Missed diff
* PR changes
* Further fixes
* Improve sorting logic
* Pass in comparator rather than using sort type
Co-authored-by: Octavia Togami <octavia.togami@gmail.com>
* Localise the ExceptionConverter
* Further work on localising exceptions
* Everything but input parsing is now localised
* Finish making errors localisable
* Fixed a few notes from review
* Fixed another spot
* More changes
* Specialise parser errors
* Add RegenOptions to the API
* Integerate regen options into //regen
* Rename isRegenBiomes to shouldRegenBiomes
* Go through the EditSession for setting biomes
* Respect 3D biome support in regen
* Add RegenOptions implementation for 1.16 Bukkit adapter.
Co-authored-by: wizjany <wizjany@gmail.com>
* Initial work on 3D biomes. Not implemented for Bukkit, and history is unfinished
* Further implement 3D Biomes. It's now usable on Forge/Fabric
* Fix a few cases of apply. Gotta fix the deprecated methods at some point
* Add Bukkit support
* Add clipboard support. Schematics are still not great, but the Sponge schem format doesn't support this yet.
* Fix the last of the issues
* Fixed up review comments
* Fixed review notes
* Simplify
* max is inclusive
* Standardize non-abstract compatibility handling
Co-authored-by: Octavia Togami <octavia.togami@gmail.com>
* Deprecate BiomeRegistry, etc.
* Update some libraries, e.g. text
* Move to new translation renderer
* Revert "Deprecate BiomeRegistry, etc."
This reverts commit 59a5d6c92aec52739a8dc68ac3d23898af7593dd.
This was not a good idea for potential mod shenanigans.
* Move BiomeData#getName to BiomeRegistry, use i18n
* Use getRichName instead of getName
* Implement getRichName for NullBiomeRegistry
* Add getRichName for blocks
* Relocate net.kyori.minecraft
* Update adapters for getRichBlockName
* Add getRichName for items
* Update adapters for getRichItemName
* Update adapters JAR for merge
* Speed up initial change buffering by removing streams and optional wrapping
* Bump the adapters for an experimental performance boost (Purely Bukkit 1.15.2)
* Bump adapters for fixes
* Bump adapters for further performance boosts
* Migrate fast mode to a BlockUpdates section, allowing fine-grained control
* Rename to SideEffect and clean up usage of the set
* Implement a 'usable' state
* Add a test command to demonstrate SideEffects. This is non-final
* Move //sideeffect into //fast
* Improve the box for //fast
* Switch to a system that includes "delayed" as a mode
* Remove the concept of configurable side effects, all are configurable. And each platform specifies what it supports.
* Implement side effects on other platforms (and spigot versions)
* Only use the side effects from the world editing platform
* Update from review
* Rename WorldApplyingExtent to SideEffectExtent
* More review fixes
* Update the World#applySideEffects return type to return the applied side effects
* Fixed requested changes
* Fixed checkstyle
* Improve setBlock javadocs
* Further improve
* Switch parameter name
* Fixed typo in translation key
* Temporarily remove DELAYED
* static final
* Side Effects = 7.2.0 snapshot
* Add new experimental snapshot API
This API intends to replace the existing snapshot API in WorldEdit 8.
It is currently experimental, and the old API is soft-deprecated. Once
it is determined to be stable, the old API will be removed and replaced
with this new one.
It uses TrueVFS instead of TrueZIP, due to the latter's status as
matinence-only, and the face that TrueVFS is NIO-FileSystem compatible,
allowing us to integrate it with NIO's native zipfs. The TrueVFS system
does encompass multiple modules, but users and download + install
truevfs-profile-default as a fatjar. We depend on that profile to show
this as the intended solution.
* Port commands to experimental snapshots. Legacy ones are in sibling classes.
* Add many tests, fix bugs
* Clean up imports
* Verify sorting behaviors, fix bugs
* Add license to test
* Add support for DIM region folders
* Fix tests for Windows paths
* Rewrite tests for even more coverage, fix bugs
* Clean up naming on tests
* Add test for legacy format
* Stop trying to abuse `file://`, use `snapfs:` instead
* Simplify iterPaths
* Use File.separator in FS snap tests
* Fix some bugs from T9N PR (#538)
* Use printInfo/printError to keep colors right
* Escape quotes before they go in the map
* I18N-ify new snapshot code
* Rebase translation work for easier rebasing
* Fixed gradle setup
* Further translations
* Add a few more strings
* Remove support for pluralisation to match Minecraft and alleviate issues with over-complexity
* Further translations
* Added a tonne more translations
* Added the remaining translation nodes
* When custom files are found, use the base files for default values.
* Only download translations if a key is provided
* Fixed typo in brush radius too large
* Allow translation files to be missing entries, and use the english version in place.
* Fixed issues brought up in review
* Reverse translation order (#533)
* Fix a lot of PR comments
* Add JD to the Operation warned field due to no private fields in J8
* Only warn when there are messages
* Re-do commits to avoid awful rebase
* You can load and save a schematic file now. Still gotta setup ability to use commands as a console actor.
* Add a world override concept to LocalSession, and allow a lot more commands to be performed by actors.
* Fixed commands, and set the loaded schematic as the world override in CLI
* Properly load tags
* Added 1.14.4 data values
* Allow a majority of commands to be performed by the console.
* Fixed a lot of PR requested changes
* Added a Locatable interface and use that for getting the location of the player in commands.
* Added script support. Currently requires a newline at the end of the script.
* Shade everything to allow this to run locally - should probably minimize this to an extent later.
* Actually hook up the version
* Added a //world command to set the override
* Fixed a missed checkstyle issue
* Added CommandBlock support to Bukkit
* Make command block support configurable
* Minor cleanup and implementing a few of the final functions
* Fixed most issues from PR
* Improve UX, saving is now automatic and unknown command messages show
* Better save docs and support any clipboard format
* Include the entire formats list
* Arrays.copyOf
* Clear the world override if the selector is called on another world.
* Update logging extent to allow basic logging with non-player actors
* Remove LocatedBlock overhead in LBL map
* Add new space-efficient block map, with thourough testing
* Drop ordering property, add full insertion test
* Add licenses
* Fix mocked platform conflicts
* Disable full block map testing for faster builds
* Re-implement BlockMap with fastutil maps
* Re-write chunk batching to be memory efficient
* Make MultiStageReorder use BlockMap
* Increase LBL load factor, fix long-pack limit detection
* Fix infinite loop in chunk batching
* Save memory in history by cleaning up MSR
* Re-implement LocatedBlockList in BlockMap
* Fix data race with BlockType lazy fields
* Make IDs ALWAYS present, only runtime-consistent. Use for memory efficiency in BlockMap
* Remap inner structure of BlockMap for smaller maps
* Remove containedBlocks fields, not very efficient
* Fix minor de-optimizing bug in stage reorder
* Make long packed y signed
* Add extended Y limit configuration option
* Add licenses
* Store 3 ints for unoptimized BV list
* Add final to BitMath
* Correct int-cast for long-packing
The new command now writes a json file to WorldEdit's working directory with instructions on which chunks to delete, which is read by the plugin/mod at startup and calls the ChunkDeleter.
The chunk deleter parses the json and iterates the instructions, backing up .mca files as it goes and overwriting the offset headers with 0 wherever a chunk needs to be deleted.
This allows Minecraft to reclaim the space used for that chunk, as well as forcing it to be generated from scratch next time the area is loaded.
