Merged functionality of "RiftHandler" class into "RiftRegistry" class
Moved the "Location" class
Why the RiftRegistry couples the Rift ID to the Rift **location** and
the PocketRegistry couples the Pocket ID to the actual Pocket
**instance**:
-Each Rift is a Tile-Entity, which means that it has its infromation can
be stored using the Tile-Entitiy's information saving system.
-Each Pocket is barely more than an information storage container, which
depends on having an external information saving system, which the
PocketRegistry provides for it.
If there is a problem reading the blacklist file from disk, previously,
null would be returned from the DDSaveHandler.readBlacklist method. This
would result in a crash later on down the line when we tried to iterate
the blacklist. Now we return an empty array. I'd like to be able to
diagnose the issues causing the blacklist file to be corrupted, but
unfortunately when this issue crops up, people only post the crash :(
If the server disconnected you due to a timeout or whatever, the pocket
manager would appropriately unload all dimension & pocket data, but it
would do so moments before the world unloaded, meaning that entities would
sometimes, on update or as part of their renderer, attempt to access the
dimensional data. This places a new guard condition that attempts to stop
that from crashing the client.
On startup, existing sub-dimensions of personal dimensions were being
given the wrong worldprovider by the launch code. The reason for this is
that subdimensions are not added to the player -> dimension mapping for
personal dimensions, because there is only one canonical "personal"
dimension. The launch code was using presence or absence in the mapping
to determine whether a dimension is personal, rather than the dimension
type. This is now fixed.
Some users were reporting very sporatic crashes in
generateLightBrightnessTable() when the server was starting up. It's
possible that very occasionally, the server will register the pocket
dimension before any pocket dimension data is initialized, forcing the
light brightness table to get generated server side with bad values.
In some cases (such as another mod crashing on startup or something), the
server shutdown even was getting called before anything was actually
initialize. In these cases, some uninitialized objects were having
cleanup methods called on them, when they were null. There are now guard
conditions to prevent the resulting NPE's.
Major change is addition of fractal rift rendering, currently first
pass.
Curves are registered and pregenerated in mod_pocketDim.
Rifts look up these curves, choose one, rotate it, and render it.
The render is a TESR that does stuff. Hard to explain, look at
RenderRift in the code and look at the actual rifts in game to get an
idea of what it does.
I had to add a triangulation library to accomplish this. Will hopefully
do something else that drag around all this.
(I tried(and used comments))
Fixed a crash from manipulating rift data on the client side. I let this
happen because it seemed like TileEntityRift already did that before.
This crash also exposed another issue: that server-side functions are
being used on the client side. I'm not sure how pervasive this is but
some client dimensions are being constructed with the server-side
constructor.
1. Fixed a design flaw in PocketManager. We originally assumed that all
requests to PocketManager.getDimensionData() had to be legitimate
requests for dimensions that existed. That was true in most cases, but
for things like processing user commands, it was dangerously optimistic.
It was possible that a flaw in DD's usage of that function could be
exploited by a player to trick the mod into pre-registering dimension
data for a non-existent dimension. That would declare the dimension as a
root. DD would crash later if Forge ever allocated that ID for a pocket
dimension. The new implementation is almost the same as the old one, but
allows us to differentiate between cases when we can eagerly create
dimension data, and cases in which the absence of a dimension should
cause a crash to alert us of a design flaw.
2. Remove the pocket regeneration code from PocketBuilder. We simply
don't support pocket regeneration and it's unlikely it'll ever be
implemented because it's a difficult issue. Wiping out pockets
completely is easier. We can always recover the code from this commit if
it's needed later.
3. Minor changes: removed some debug prints from PocketManager and
changed some static accesses in PocketBuilder.
Fixed an issue. DD would crash when MC created a completely new world
because onChunkLoad() would be called before onWorldLoad(). That's not
the usual order. PocketManager would be unloaded at that point and would
return a null dimension.
Autocorrected indentation in PocketManager. I'll be working on changing
PocketManager to prevent the risk of creating data for a non-existent
dimension through a bad call to getDimensionData().
Made some improvements to TileEntityRift. The main reason for these
changes was to remove the field nearestRift - we should not hold on to
references to links. Now we simply track the location of the nearest
rift. I also confirmed that closeRift() and updateNearestRift() must be
allowed to run on both the client and the server. If the client doesn't
run those functions, then adjacent rifts don't connect as expected and
the rift removal animation doesn't work.
Removed a pointless check in yCoordHelper and corrected some comments in
RiftRegenerator. It turns out that ChunkProviderServer.chunkExists()
returns whether a chunk is loaded, not whether it has already been
created.
Removed unnecessary code for the Transdimensional Trapdoor. Most of it
was code dedicated to updating TileEntityTransTrapdoor.hasRift. That
flag was never used for anything.
1. Added code so that Transdimensional Trapdoors detect that they have
been broken and schedule rift regeneration at their location. This had
previously been neglected. Trapdoors deserve a little more attention.
2. Tweaked the breakBlock() code for BlockRift and BaseDimDoor so that
rift regeneration is only scheduled if the underlying block was removed.
We don't want that to happen if the only change was for metadata.
1. Removed code from BaseDimDoor that was already implemented almost
identically in BlockDoor. Clarified some uses of setBlock() by changing
them to setBlockToAir() instead.
2. Removed TileEntityDimDoor.invalidate() and moved the regeneration
scheduling code to BaseDimDoor.breakBlock(). I would prefer to move away
from overriding the invalidate() method. This also simplifies the code
since we don't need to perform some of the checks we had in
breakBlock().
1. Made it so that rifts regenerate when rift blocks are replaced by
other blocks.
2. Changed the rift regeneration scheduling functions to streamline
their use in other classes. Common code that was needed to validate
links before calling those functions has been moved into them so that
the checks are always performed internally.
1. Implemented scheduled rift regeneration in RiftRegenerator. The
previous randomized selection algorithm has been removed completely. All
regeneration is scheduled now. We perform numerous checks to make sure
that regenerating a rift is safe.
2. Removed FastRiftRegenerator as RiftRegenerator performs roughly the
same task but with more flexibility. Updated TileEntityDimDoor to use
RiftRegenerator instead for creating rifts when doors are broken.
3. Modified EventHookContainer to receive the chunk loaded event. We
iterate over the list of links in a loaded chunk and schedule them for
regeneration.
4. Reorganized the code in BlockRift. Divided the list of immune blocks
into two lists - one for DD blocks and one for regular MC blocks.
RiftRegenerator has to be able to distinguish between the two types.
5. Factored out some duplicate code from ItemRiftSignature and
ItemStabilizedRiftSignature. Most of the block immunity checks were used
to check if it would be safe to spawn a rift when using one of those
items. BlockRift.tryPlacingRift() covers all that logic in a single
function and makes the item code a little simpler.
Minor changes to DimLink to simplify DimLink.getDestinationOrientation()
and to clarify the output of DimLink.toString() when no destination is
available.
Rewrote the NewDimData.deleteLink() version that would accept x, y, and
z as parameters. There was some redundant code for getting the
parameters from a Point4D instance just to create another one to find
the target link. Now we pass the source point in directly.
Made some minor changes to NewDimData. Fixed some comments for the rift
search functions that incorrectly indicated the search would not detect
rifts adjacent to the center of the search range. That behavior changed
some time ago. Also added an unused field called "chunkMapping" for
associating chunks with lists of links. It'll be used for implementing
queuing of rift regeneration when chunks load.
1. Reorganized our code to initialize tick receivers each time the
server starts rather than once when the mod is initialized. This is
needed because reusing a single instance of each class across different
single-player sessions could cause scheduled events for one world to
leak into another world. This approach ensures that we discard all
pending events.
2. Separated the implementation of Limbo decay from a tick receiver that
periodically triggers fast decay. All of the decay code has been kept in
LimboDecay, while the ticking is handled by LimboDecayScheduler. This
change separates some functionality that should be independent, but
also, it's needed so that BlockLimbo can have access to LimboDecay's
methods without holding on to a tick receiver instance.
3. Minor change: renamed ChunkLoaderHelper.loadChunkForcedWorlds() to
loadForcedChunkWorlds().
1. Renamed CommonTickHandler to ServerTickHandler. Given that it only
handles server ticks, this seems like a reasonable name. Also changed
its profiler label to the new name.
2. Deleted ClientTickHandler and removed any references to it in
mod_pocketDim. It was never used for anything.
Made various changes to clarify code in DDTeleporter. For instance, we
had a whole switch block that was used to give the same outcome on every
case except the default. I rewrote the code there to remove the block.
Also changed DDTeleporter.checkDestination() since it was redoing the
destination orientation checks unnecessarily, changing the entity's yaw
when it shouldn't have side effects, and some other little things.
1. Changed build.gradle so that it edits the version numbers in
mod_pocketDim.java and mcmod.info.
2. Changed mod_pocketDim to use a placeholder for its version number and
fixed an annotation that was wrong. It would cause our mod_pocketDim
instance to not initialize properly. I removed what seemed to be
workarounds that were hiding the problem.
3. Fixed space in TileEntityDimDoor.invalidate() and corrected a
non-static access to a static field in mod_pocketDim.
4. Changed mcmod.info so that it users placeholders for the mod version
and MC version values.
We previously returned the Vanilla counterparts to our doors as the
items to be used by the pick block button - used in Creative mode for
replicating nearby blocks. This was incorrect because we would want to
return the actual door item needed to place the particular door in
question.
More importantly, this might solve our issues with WAILA reporting the
wrong information when players look at our doors. I read the code for
the most recent version of WAILA and it uses various functions to choose
how to identify a block. The result of Block.idPicked() is probably used
as the main source for the identities of our blocks.
Fixed the way in which we handle redirecting links to blacklisted
dimensions. The previous method always converted links into safe exits.
This lead to strange situations that could be seen as bugs. For
instance, using a dungeon entrance in a root dimension would generate an
exit door and a supporting platform directly above the entrance door.
That also meant that any visited dungeons would be unusable if they were
reset.
We now do different things depending on the location of the link and its
type. If the link is a dungeon link, then its destination is reset to
allow a new dungeon to form. For other link types, if the link is in a
pocket dimension, then it becomes a safe exit link, because it could be
the only way out. If it's in a root dimension, then there are no
reasonable destinations, so the teleport request is cancelled.
Made some minor changes to DDTeleporter to get rid of a few warnings.
Also tweaked and commented DDTeleporter.initializeDestination(). There
were inappropriate references to a link's internal variables instead of
using its getters and there was a subtle link overwrite for blacklist
destinations. Without comments, it would probably be unclear that
overwriting that link is safe. This changes are in preparation for
fixing issues with blacklisted destination redirection.
Rewrote CommandTeleportPlayer for the same reason as usual. There were a
few bugs before, such as that lookups for players were limited to within
the world from which the command sender was using the command. Players
in other dimensions could not be teleported. The command would also
place people in the ground because it did not adjust its coordinates for
the way that DDTeleporter interprets them - as the location of the top
block of a door and the player's head.
I removed CommandDeleteAllLinks because of the significant risk that it
would harm a server. It's possible that someone could run this by
accident instead of CommandDeleteRifts. It takes the same arguments or
even no arguments and it would immediately wipe all links in a
dimension. We can restore it later if it's really needed.
Added a minor check to PocketManager.loadDimension() so that an attempt
to load an unregistered dimension won't be passed on to Forge where it
would cause an exception - which is then caught by Forge before it can
cause problems. This isn't strictly necessary, but it's a nice
consideration.
Increased the maximum Monolith aggro level and the cap range values by a
factor of 1.25. The max aggro increase is to slow down how long it takes
Monoliths to max out because they're just a little too fast right now.
The cap adjustments will preserve the range of texture states they can
have while idling.
1. Rewrote CommandResetDungeons to improve clarity and remove bugs. This
version of the command preserves valid links. Those were previously
removed, which would break some dungeons unnecessarily.
2. Fixed NewDimData.setParentToRoot(). The function did not account for
the possibility that the target's parent might still exist and would
need to be updated, leading to conflicting data. It also did not reset
pack depth. We now correctly update a dimension and all its descendants.
1. Rewrote or removed a few bits that were causing minor warnings.
2. Rewrote deleteDimensionFiles() and deleteDimensionData() to remove
unnecessary casts and checks. We can confirm that those checks are
unnecessary because those functions are only used inside PocketManager.
If they were ever exposed externally, then we would need to add checks
again.
Rewrote CommandDeleteRifts for clarity and to remove several flaws.
Previously, the command would have removed non-immune blocks that were
in the same location as a rift. It also set blocks in the command
sender's dimension rather than the target dimension, so blocks would
have potentially disappeared from the wrong world.
1. Fixed some warnings in the affected commands.
2. Removed checks for whether a command is running on the server or
client side. We have performed those checks inconsistently throughout
our commands without problems. I assume that they must be running on the
server side only. If I'm wrong, we can add a check to DDCommandBase.
3. Minor punctuation change in DDCommandResult
1. We weren't giving old tickets to the doors that owned them. That
would result in doors requesting new tickets and the old ones wouldn't
be released. Each time a server rebooted, a new ticket would be created.
Then Opis would report that many chunks were forcefully loaded in a
pocket because it doesn't consider overlapping tickets. We now give
doors their old tickets when they're reloaded and we release extra
tickets referring to the same door. That will also deal with the excess
tickets that already exist on servers.
2. Rewrote the logic for checking if a Golden Dimensional Door is
allowed to force-load a pocket. We now check if the door is within the
horizontal bounds of the pocket. This prevents the confusing scenario
where someone places a door far away from the pocket but the only chunks
affected are in the pocket.
3. Fixed the calculation for determining which chunks must be
force-loaded to cover a pocket. This has the benefit that fewer chunks
should need to be loaded. It should be enough to load 16 chunks. We
previously loaded 25 chunks just to err on the side of caution.
4. Golden Dimensional Doors only try to initialize as chunk loaders
once. We previously allowed them to keep trying every tick until they
could get a ticket.
Added the Universal Limbo config option to the world config settings.
When enabled, it causes players to get teleported to Limbo if they die
in any dimension except Limbo. It's disabled by default.
This feature was requested by Mr_Turing.
Added code to EventHookContainer.revivePlayerInLimbo() so that the
player's food level is restored upon respawning in Limbo. Apparently it
was possible for players to have death loops from starving to death in
Hardcore Limbo.
Changed the function call for playing the rift-closing sound to match
other similar calls in Minecraft - the sound should target the center of
the block. I also switched the last flag argument. It seems to determine
whether distance between the player and the source should be checked
when playing back the sound, although documentation is lacking so I
can't be sure. Other blocks that use that function use "false", while
things involving sounds that should be heard uniformly, such as
rainfall, use "true".
---
I noticed that this code runs on the client and server. Some parts
modify link data and rifts on the client side. We should really stop
this from happening as it could lead to inconsistencies.
Added support for using Stabilized Rift Signatures from dispensers under
specific conditions. Everything works except for playing the sound that
occurs when an SRS is used. Not sure why that's not working.
1. Removed several fields and functions referring to the newer rendering
code. The functions were just cluttering up the code and the fields were
consuming additional memory that was never being used for anything.
2. Updated NBT reading and writing functions to give some tags proper
names and to remove references to unused fields.
Removed the tag for riftCloseTimer because it was unnecessary alongside
the shouldClose flag. If a server reboots while a rift is closing, the
rift can start over upon reloading.
3. Renamed some fields and functions to have better names.
4. Changed the various checks for closing rifts. There were a few
redundant parts. We don't have to put calls to "this.invalidate()"
everywhere on top of explicitly removing the tile entity and destroying
its block.
5. Rewrote update timing checks. The rift spread and Enderman spawning
calls have been separated to distribute the impact of updating a rift.
Also, a flaw in the timing logic meant that the calculations for
particle offsets would only run when a rift was first created. That's
been fixed.
1. Fixed the bug where the setting that controls whether Endermen can
spawn from rifts was being ignored. It was never checked at all.
2. Cleaned up some formatting and annotations.
3. Removed call to World.removeBlockTileEntity() following a call to
World.setBlockToAir(). The latter function already handles removing the
tile entity.
1. Changed hasEffect() override since we were overriding a deprecated
version.
2. Fixed a bug where we checked if a block could be edited before
deciding whether to change the Y coordinate of the rift to be placed.
Sometimes we would place the rift in a different block. This is the
result of sticking in support for special blocks like grass and snow
without considering the impact on surrounding code. It also contradicted
comments that specifically said special blocks were ignored...
3. Cleaned up the code for checking for special blocks.
4. Fixed a bug in loading NBT data. There were no null checks on
orientation data. If a Rift Signature or Stabilized Rift Signature had
been created in a version of DD before orientations were set up, then it
cause an exception when so much as looked at in later versions of DD.
5. Partially implemented free redirects for Stabilized Rift Signatures.
The check to determine if a redirect is being done is missing.
1. Changed the crafting recipes for most DD items to use Ender Pearls
instead of Stable Fabric. The items that still use Stable Fabric are
Dimensional Doors, Golden Dimensional Doors, Rift Blades, and Stabilized
Rift Signatures. Steven had already made this change in another branch
but I'd like to push this out with several bug fixes. The SRS recipe is
different from his version - it's now just 4 Iron Ingots and a Stable
Fabric.
2. Change Stabilized Rift Signatures back to consuming Ender Pearls
instead of Stable Fabric.
Finished implementing Personal Pockets
-any pocket created from within a personal pocket retains personal
status
-exit doors cannot be used in any personal pockets
-personal status is saved with dimData
fixed a bug that let trapdoors get around locks
fixed FoR not rendering properly
-inventory and world
Changed the max stack size of ItemGoldDoor to 16 as it is for Vanilla
doors on some modpacks. Later versions of Vanilla have door stacking to
64 so this will need to change eventually.
1. Changed EventHookContainer to remove a check against
BaseItemDoor.getDoorToPlace(). The checks performed there can be done
in BaseItemDoor.tryToPlaceDoor(), which removes the need for callers to
know more internal details about how doors are handled. I moved the
checks inside.
2. Renamed vanillaDoorMapping to doorItemMapping. It now maps dim door
items to themselves to remove the need for various checks we were
performing. Updated BaseItemDoor's constructor to reflect this change.
3. Removed BaseItemDoor.getDoorToPlace() and integrated its
functionality into BaseItemDoor.tryToPlaceDoor().
4. Changed BaseItemDoor.tryToPlaceDoor() so that it simply returns false
if a given item stack cannot be used to place any doors. We don't need
to check if the item is an ItemDoor or anything like that now.
1. Cleaned up some spacing and unused imports in EventHookContainer.
Also changed an indirect reference to BaseItemDoor.trytoPlaceDoor() to a
direct reference seeing as the function is static and should be accessed
that way.
2. Renamed getDoortoItemMapping() to getDoorBlock(). The original named
had a minor capitalization mistake and implied that it would return a
mapping table or would associate doors to items. The function actually
associates items to door blocks.
1. Stopped Monoliths from teleporting players while in Limbo. This was a
serious issue on some modpacks if players got unlucky.
2. Decreased the required Monolith aggro level to start spawning
particles around a target player. The required level was so high,
combined with the current Monolith speed, that players would hardly see
the particles.
3. Disabled Monolith sounds in Limbo. Some of the sounds were really
annoying in Limbo. Usually they only get to play for a moment before a
player is teleported, but since no teleports occur in Limbo and the area
is full of Monoliths, the constant noise is aggravating. It would also
drown out the background music.
* Fixed a mistake in the crafting recipe for Golden Dim Doors
* Removed static import of fields from mod_pocketDim. There were already
qualified references to those fields in some areas. We should be
consistent.
* Fixed issue with Monoliths detecting players through walls
* Changed aggro values from bytes to shorts
* Fixed aggro updates so that aggro levels can decrease
* Fixed upper bound on aggro clamping
* Added client/server-side checks to a few functions to save on
performing pointless checks, such as making calculations for spawning
particles on the server
Removed lots of obsolete and unused code from Monoliths. The code is
subdivided more clearly now. The aggro level is sent over a data watcher
instead of sending the texture state.
Keybounce noticed that he was having a persistent item ID conflict with
DD's Golden Doors - the normal doors not the dimensional variant. I
discovered that we have been assigning them the same ID as the Golden
Door block ever since they were first introduced 7 months ago. It didn't
break immediately since Forge adds +256 to item IDs.
* Changed saving code to create backups by moving existing files rather
than creating copies and deleting the originals.
* Removed final call to PocketManager.save() in PocketManager.unload().
Since we no longer check if the caller is the client or server and
unload() must be called from both, this prevents clients from trying to
save pocket data locally. A final save() call wasn't needed anyway.
Added a flag to NewDimData so that we can avoid writing dimensions to
disk if they haven't been modified. They're still rewritten when the
server shuts down.
Made saving a bit more robust, now we only delete ones that where not
modified. Also overhauled door placement, its all handled by the
eventHandler now.
* Changed DDTeleporter to stop us from generating exits to Witchery's
Spirit World - this would cause people to lose their items upon leaving
the dimension.
* Changed GatewayGenerator to stop us from generating gateways or rift
clusters in the Spirit World
Changed FillContainersOperation so that empty dispensers are not filled
with a stack of arrows on import. Just in case someone needs to use
empty dispensers in their design. As far as I can remember, this won't
affect any of our dungeons - arrow traps are rare. The only dungeon that
I can remember is one by Balgor and it was exported with all of its
dispensers loaded.
* Added a small treasure room that opens when the puzzle is solved and
made the exits in the corners slightly more apparent by placing redstone
lamps.
* Tweaked the redstone a little to correct for a brief signal cutoff.
* Tried to add fireworks that would shoot out when the puzzle was
solved, but it was too hard to secure them against theft while also
getting consistent explosions.
Updated the dungeon pistonFallRuins so that it causes less extreme lag
by slowing down and spacing out the timing for the piston traps. Also
reinforced the dungeon against tampering and added lava inside each
piston drop.
XombyCraft has been kind enough to research which provider and biome IDs
are used by major mods. He found some ranges of free IDs and suggested
changing our defaults to sidestep conflicts with Biomes o' Plenty.
Fixed a nasty crash to desktop. It happens when ChickenChunks is
installed and a new world is generated with HardcoreLimboEnabled = true.
It appears that ChickenChunks forces a chunk to generate in Limbo if
LimboProvider.canRespawnHere() = true, which is the case if hardcore
Limbo is enabled. Our Monolith and gateway generation code runs as a
tick handler instead of through standard world gen calls. I believe
Limbo is unloaded immediately after the chunks are generated because no
players are around. That would cause DimensionManager to return null for
Limbo because it's not loaded, which would crash our code. We probably
dealt with this for Monoliths by adding a check. Now it happened again
because we didn't take precautions while calling the gateway generation
method. I've added code to forcefully load Limbo if it's not loaded.
Changed the condition on LimboProvider.canRespawnHere() so that players
can respawn in Limbo even if LimboEnabled is false. LimboEnabled only
controls whether players are sent to Limbo when they die in a pocket. It
does not prevent players from ending up in Limbo because of Monoliths.
If Hardcore Limbo is enabled, it stands to reason that people should be
respawning in Limbo anyway.
