Since merging by distance only creates false positives (one would have
to increase the bounding box for an entry, but the angles would never be
100% right), we never merge, instead we don't add a new entry if the
position is the same, the time value is not updated in this case. For
validity of an entry you always have to consider the time span until the
previous (younger) entry or until now for the latest entry.
Rough changes:
* Use an interface for accessing trace entries.
* Use a linked structure for the actual trace.
* Use maximum age and size to limit the number of stored entries.
* Use a pool to somewhat limit object creation (size may need
configuration or scaling with number of players).
* Since the trace starts empty, have the field be final.
* Keep trace elements if settings are changed, cut size if necessary.
* Remove obsolete tests.
Potentially missing:
* Usage of LocationTrace has not been checked if we need to account for
the time of the latest entry not necessarily being updated (!).
* New tests, e.g. accounting for the expiration of entries.
Follow ups:
* (Extend fight/loop checks to a latency window mechanism.)
Upcoming changes will roughly be:
* Change implementation to a double linked structure.
* Implement/use something like ListIterator.
* Never merge entries, instead use some pool and time/extre-n as limits.
* A basic latency window implementation just for the LocationTrace for
preliminary experiments. [Track hit/miss all time + recent so and so
seconds, some extra cancelling/invalidation mechanics, allow to test
complement 0->window start and possibly window-end-> max latency for
some cases, cancelling mechanics may contain a buffer or a mixture of a
buffer relating to average miss rate]
Prepare using VehicleUpdate and PlayerMove instead of VehicleMove for
vehicle moving. This change isn't intended to change
anything/much on the surface.
* Implement native IEntityAccessPositionAndLook for 1.9_R1 and 1.9_R2.
* Alter method visibility and parameters.
* Common pre-conditions.
* Route contents of both VehicleUpdateEvent and PlayerMoveEvent through
a common related method (also named onVehicleUpdate).
* Remove RichLivingEntityLocation, to be able to simplify more.
* Refine interfaces for locations (IGet... ISet... vs, I... for both).
* Implement location related interfaces in some places, related changes.
* Override hashCode for some of the location related classes. Use that
for storing location hashes instead of Location.hashCode. Auxiliary
methods for hashCode in LocUtil.
* Add onIce to LocationData.
* Renaming player vs. vehicles (likely incomplete).
* Possibly other related/random changes.
Line count is high for this change, despite not so complex. Next step is
to change VehicleChecks to use past move tracking to estimate from where
a vehicle is moving (left not compiling). Due to the lack of teleport
events, and due to entity last location being mostly useless, we have no
choice but to hard-set-back on anything that looks strange.
Directly following:
* Boat fly check based on VehicleUpdateEvent and fetching last pos.
* Implement a native access based provider for
EntityAccessLastPositionAndLook, after testing the reflection based one.
Likely following:
* Implement the same fly checks based on PlayerMoveEvent for horses and
pigs too, for the case server-side fly checking is disabled.
* Configurability for individual types of enbtities, at least a flag for
activation.
* Not sure if a fall-back to VehicleMoveEvent should be kept, setup
shouldn't be all too complicated.
Major: Sketch vehicle envelope check.
* Renaming fields, methods, packages. Moving classes to other packages.
* Additions and refactoring for set-back handling and location tracking.
* Increase amount of debug logging.
* Adjustments to current vehicle set back handling.
* AuxMoving: call clear() on setMCAccess.
Minor: Adjust block change tracking implementation.
* Use a class instead of an id, in order to keep track of used entries.
* Allow reuse of an id, if the block still is intersecting.
* Improves situation for simple setup, issues remaining:
* Random UNKNOWN teleport by server potentially interfering.
* Distances > 1.0, possibly resulting from split move handling.
* On-ground estimation and passable.
* Blocks with gravity are worse (likely on-ground).
* More in-depth checking of constraints of implementation.
* Note that the block change tracker currently is disabled by default.
Naming:
* Possibly not final.
* survivalfly: Does not fit at all.
* Separate official checks + sections for speed and fly: nope.
* runfly: not running.
* speedfly: mix up with too fast flying.
* moving: too much redundancy with moving.vehicle.moving
* envelope: so and so, possibly better with tags to be set.
Content:
* Prepare a vehicle moving envelope check (just basic pre-coding
bookkeeping and refactoring).
* Split vehicle checks to another class. Needs refactoring.
* Add ability for simple generic instance registration by an interface.
* Add new sub check type MOVING_VEHICLE, with configuration section,
move existing stuff there (moving.vehicles.., morepacketsvehicle).
* Breaks at least the use of check type MOVING_MOREPACKETSVEHICLE.
FunFact:
* Try CheckType setup with MOVING_MOREPACKETS_VEHICLE(MOVING, ...) to
see why there is a test for this kind of thing.
Bugs:
* Old configuration paths don't seem to get removed with @Moved.
Missing:
* More refined vehicle checks.
Issues on quick testing:
* End portal seems bugged, resulting in desync, entering in creative
mode. Switching to survival via console allows end portal travel after
kick+relog. Can't tell if this is a 1.9 bug or due to building NCP vs.
an older version of ProtocolLib.
This won't cover all effects of internal changes, some blocks might be
interpreted wrongly still, some shapes may have changed in an
incompatible way (e.g. skulls).
(@Samistine).
Slight deviations:
* Random formatting changes.
* Use a linked list for changedCommands at first.
* Use list.clear after iteration, instead of calling remove repeatedly.
* descendingIterator instead of ListIterator + (has+) previous.
Represents the first "simplistic" approach to block change tracking,
only attempting to make vertical push/pull work.
It seems that we need to add on-ground checking accounting for piston
moves as well, otherwise anything with pistons retracting will lead to
survivalfly violations. Pistons extending and retracting may also
randomly move around players, including dragging them into the piston
block with the bounding box (not center of player).
In order to make on-ground work, we might need to check in another
place, possibly check where resetFrom an resetTo are set. Performance
questions might remain, there might also be a slight redesign necessary,
in order to run some sub-routines more side-effect free, to check
several branches, including after-failure checking.
This breaks testing for UNKNOWN_VERSION, if that is used externally.
Access methods are added for testing for unknown versions.
PR mentioning access methods:
94c4da3267
* Group selected classes into sub-packages of moving.
* Rename classes.
* Must use LinkedList for velocity entries.
* Prepare SimpleAxisVelocity + entry for use-once accounting.
(Might not be the final naming.)
Using profiles:
* Profile "minimal" will build by default, excluding all dedicated cb
dependencies, making it easier to quickly test stuff.
* Set the property 'cbdedicated' to true, in order to build all, using
the profile "all".
* The properties BUILD_NUMBER and BUILD_SERIES don't seem to set to
defaults anymore, so they have to be set manually (BUILD_NUMBER does get
set on jenkins).
* Not actually a fix for anything we encountered.
* Nailed down blockinteract.visible raytracing issues to bad end-points
for raytracing.
* Also test/prepare logging test-cases for raytracing in general. Not
enabled, because we should have some flag/permission/command to check
before logging ~ 5KB per interact event.
This makes logging all violations potentially useful to use alongside
with the "ncp debug player" command in production environments. The flag
debugonly must be set with at least one backend being activated.
Similar to TestNCP but reduced/different features:
* Config: trace for the log file and notify to send to notify channel.
* It's not possible to confine whose messages you receive (yet).
Meant for better local/quick testing in the first place.
