The game minetest.conf flags directly overwrote the global minetest.conf default values, resulting in unwanted erased mapgen flags.
* Fix set_mapgen_setting
Unified flags handling in C++ and Lua Settings API
-> Reading only, for now. Writing can be implemented later, if needed.
API function to read the currently active flags
-> was impossible from Lua
Co-authored-by: Wuzzy <wuzzy2@mail.ru>
This is necessary under Windows systems, and direct IPv6 connections.
Windows universally disallows ':' from occuring in filenames.
Other disallowed characters on Windows:
\ / * ? " < > |
are not relevant to hostnames, IPv4 or IPv6 addresses.
Anyone who has got an existing server map saved on Linux with ':' in
the world save will want to keep that save.
GenericCAO::getPosition() did not take the camera offset into account
LocalPlayer attachment cleanup: Use sane getParent() function
Make that getPosition() (GenericCAO and LocalPlayer) always return the absolute position
* Formspec: Fix priorities for version < 3
1) Introduce 'priority' to 'FieldSpec'
2) Sort elements based on 'priority'
3) Assign 'name' to the Item Image Button's image to show tooltips again
Low-disruption first step towards removing the hardcoded cave liquid
code. Since MT 5.0.0 cave liquids can be defined and located by
biome definitions instead.
In games that do not yet use biome definitions to define and locate
cave liquids (MTGame does), lava will now appear below
y = water_level - 256 instead of below 'lava depth' (usually y = -256).
Therefore no change in most mapgens if using the default 'lava depth'.
Previously, the only way to disable the 3D noise tunnels was to set
'cave width' > 1.0, however doing so did not disable the very intensive
noise calculations or the generation loop.
All the other types of cave generation (randomwalk caves, caverns)
can already be independently and completely disabled.
This feature is now needed more because the small randomwalk caves are
now available for use as an alternative to the 3D noise tunnels.
Add mapgen parameters to set the range of the random number of
randomwalk caves per mapchunk, and to set the proportion that are
flooded with liquids.
Default values are, for now, unchanged from the previous hardcoded
values.
Add parameters to allow small randomwalk caves
Disabled by default for now as they have never been present in the
non-mgv6 mapgens.
Revert gamma upper limit to 3.0 because that was raised based on
a misunderstanding and had no benefit. A sane upper limit is
needed as players on a competitive server tend to use the maximum.
Set gamma lower limit to 0.33 for consistency with 3.0.
Set sane limits on alpha, beta, boost and enforce these in code
to limit values entered in minetest.conf and to avoid easy cheating
by editing settingtypes.txt.
Improve documentation and 'readable' setting names.
Clarify that gamma does not significantly affect natural night light.
light.cpp: Various codestyle and comment improvements.
This value reduces how far randomwalk caves can extend into the
mapchunk padding, to avoid 'out of voxelmanip' cave nodes which cause
flattened cave walls.
Testing shows that a value of 2 (instead of 10) is enough to make
'out of area' nodes extremely rare.
Reducing this value results in a higher chance of overlap and
connection with caves of neighbour mapchunks.
- Add style properties for overriding the the hovered/pressed state
- By default, hovered buttons are a lighter version of the base color
- By default, pressed buttons are a darker version of the base color
- Add hovered bg image support for image buttons (style property)
This reverts commit 1db3d252cff9e8d61fecf1052d7497813851da51.
Temporary solution until a compatible solution is found to define both - formspec and fullscreen backgrounds.
Does not clear the parent's attachment information when the child is deleted locally.
Either it was removed permanently, or just temporary - we don't know, but it's up to the server to send a *detach from child" packet for the parent.
1. Pass current camera offset to shader, so shader have access to the global coordinates
2. Pass animation timer to fragment shader. C++ code is already there, just wasn't declared in the shader
3. Delay animation timer wrap-around (from 100s to about 16 minutes)
The setting now accepts values between 1 and 8 in the
Advanced Settings menu.
Values 0 and 1 now behave the same way (setting it to 1 won't
disable MSAA anymore), so there's no need to expose 0 as a value.
This closes#8939.
While preserving the general character of dungeon structure.
Slightly increase the range of standard room horizontal size, while
preserving the average horizontal size.
Return to classic maximum large room size of 16x16x16.
Make 1 in 4 dungeons have a 1 in 8 chance for each room being 'large',
making multiple large rooms possible for the first time.
Make 1 in 8 dungeons allow diagonal corridors, to make these a little
more common.
Make corridor width vary from 1 to 2, but forced to 2 if diagonal
corridors are allowed, to make them passable.
Add some comments.
Safety-guards for CSM callbacks to abort on a bad implementation
Only run callbacks when the mods are loaded (and with it: builtin)
Duplication checks inside constructors
Before opening the file for writing, its file size is tested. If it exceeds 50 MB, it is moved to debut.txt.1, otherwise the log is appended to the old messages. An old debut.txt.1 is removed if it already exists.
Calculate 'i' inside 'y' loop to avoid 'i' and 'y' becoming
out of step due to the uses of 'continue'.
Simplify calculation of 'p2d' when coordinates are inverted.
Remove some unnecessary and unreliable 'index out of
voxelmanip' checks.
Move calculation of '&em' out of loops.
For 'Loop further down until not air' code, use y coordinate
to detect being out of voxelmanip, instead of checking index
which did not detect that.
Add and improve comments.
Indent the 'for (;; y--) {' loop.
Improve format and fix codestyle issues.
This change reduces the amount of sent data towards clients. Inventory lists that are already known to the player are skipped, saving quite some data over time.
Raises protocol version to 38 to ensure correct backwards-compatible code.
In MapgenV6::flowMud(), the previous implementation of coordinate
inversion caused the 2 inverted mudflow iterations (out of the 3
iterations) to not loop over the area, so only 1 non-inverted
iteration occurred.
Fix this bug but only iterate mudflow twice, as mapgen v6 has only
had 1 iteration for many years. There is now a good balance of 1
non-inverted iteration and 1 inverted iteration.
* Fix unnecessary exception use in Server::SendBlocks
The code in this method calls getBlockNoCreate and then
messes around with try...catch to skip blocks which are not
in the memory. Additionally, it repeatedly calls
m_env.getMap() inside this loop. Speed the code up by
extracting the m_env.getMap() out of the loop and getting
rid of the try...catch.
* Fix unnecessary exception use in Server::SendBlock
Another unnecessary try...catch is slowing down
Server::SendBlock. Remove that to speed it up and get a nice
side effect of simplifying the code in question.
* Fix unnecessary exception use in MMVManip::initialEmerge
Remove another unneeded exception usage from
MMVManip::initialEmerge to speed that code up and simplify
it but be careful to not remove the braces as there is a
TimeTaker in use there.
The Map::getSectorNoGenerate throws an exception but no other
code is really dependent on that. Fix the odd instance of
misuse in ClientMap::emergeSector and remove the exception
throwing version of the method along with the "NoEx" suffixes
in the names of these methods.
Update the profiler names to make more sense of what they actually represent
Move the profiler code from header to its source file
Use monospace font to align lines
Format the statistics line to align better with surrounding values
Refresh the profiler each 3 seconds (roughly)
The isNodeUnderground calls getBlockNoCreate which calls
getBlockNoCreateNoEx and throws InvalidPositionException
if the returned value is nullptr, which isNodeUnderground
then catches to return "false". Remove the try..catch in
isNodeUnderground by calling getBlockNoCreateNoEx instead
of getBlockNoCreate and checking the returned value for
nullptr.
Pointers shall be set to nullptr, not 0, according to the
coding standards. By implication they shall be compared with
nullptr, not 0, too. Fix this code to match that.
This fixes overridden items keeping their old groups in the group to
items mapping even after their groups have been changed in lua.
It also prevents a more widespread issue where overriding an item
will add its content ID *twice* to the mapping, resulting in odd
behaviour in features such as ABMs.
This fix consists of two parts:
- Clear the list of stored entities. This has no side-effects.
- Catch the case where active entities exist and print a message.
Clearing the active entitiy list has side-effects that should
be handled. (those entities are known to the environment and
to clients).
As avoiding those side-effects is more complex, and as this
problem is not expected to occur (with PR #4847 merged), there
is no real incentive to implement this ATM.
This issue was a contributing factor to bug #4217. With the other
contributing factor removed (PR #4847), this commit makes sure this
factor does not go unnoticed if it ever happens again.
The commit 526a9e4b66abaf83eb6b1aaa3e93375acd87b830 breaks
the non-GLES2 setups because the code that is intended to
handle that is behind "elseif()" which is interpreted as
"elseif(false)" and thus the code never gets executed. Fix
that by changing the offending line to else().
Additionally, to avoid breaking the server only build
(which shall not have a dependency on GL/GLU/GLES at all),
enclose the entire block code in if(BUILD_CLIENT).
* Fix some issues with minetest.clear_craft
- Fix memory leak
- Fix crafts with an output count not being cleared when clearing by
input.
- Fix recipe list being reversed when clearing by input.
* Add CraftInput::empty()
Avoid an unsuitable spawn position (which if outside mapgen limits can
cause a crash) if the main 0-3999 loop reaches its end. Fallback to a
spawn at 0,0,0.
Check the mapgen-returned 'spawn_level' value for being outside limits.
When 'air_count' reaches 2, move back down 1 to spawn in the lower
empty node.
If the spawn position is disallowed by 'objectpos_over_limit()', 'break'
from loop instead of 'continue' because positions above are probably
also over limit.
Reset 'air_count' to 0 if an obstruction is found, to make 'air_count'
consecutive empty nodes.
Allow spawn in 'airlike' drawtype nodes such as mod-added vacuum,
alien atmospheres, fog etc.
Add clarifying comments and improve codestyle.
This fixes an issue where when the engine looked up groups (for example,
in ABM node names), NodeDefManager's m_group_to_items would contain nodes
with a group value of zero, resulting in nodes with flammable = 0 being
burned by a fire mod with a group:flammable checking ABM.
It brings consistency to the behaviour described in the api
documentation, where zero and nil groups should be the same.
Previously, this wrongly returned ground level (a position containing
a solid node) as spawn level.
Return ground level + 2 (+ 2 to spawn above biome 'dust' nodes).
Improve codestyle and make more consistent with generateTerrain().
Due to commit ec3142a , UnitSAO's getArmorGroups() did not match
ServerActiveObject's, notably resulting in the lua get_armor_groups() call
returning nothing.
Document new meaning of immortal=1 for players
Disable breathing if player is immortal
Hide builtin statbars if player immortal (delayed)
Co-authored-by: ClobberXD <ClobberXD@gmail.com>
This prevents set_properties() calls that have nothing to do with hp_max or breath_max overriding the saved hp before another mod has the chance to set a player's intended hp_max (such as in on_joinplayer).
What happened:
1) Object data is received. Client begins to read the data
2) Client initializes all its children (gob_cmd_update_infant)
3) Children try to attach to parent (yet not added)
4) Parent initializes, is added to the environment
And somewhere in between, Irrlicht wrecks up the attachments due to the missing matrix node.
The solution here is to:
1) Use the same structure as ServerActiveObject
2) Attach all children after the parent is really initialized
v6 always last to discourage selection.
Special mapgens flat, fractal, singlenode, next to last. Of these, singlenode
last to discourage selection.
Of the remaining, v5 last due to age, v7 first due to being the default.
Re-add the random size range for large rooms.
Remove 'first_room_large' bool.
Add 'large_room_chance' parameter that can disable large rooms,
specify 1 large room, or specify a chance for large rooms.
If 1 or a chance is specified, the first generated room is large,
to take advantage of the intersection checks that are done for the
1st room only.
Move 'num_dungeons' to 'DungeonParams'.
Add new parameter 'num_rooms' to replace 'rooms_min' and 'rooms_max',
so that the mapgen has complete control over the number of rooms.
Add new bool 'first_room_large' so that the mapgen chooses this
instead of a hardcoded 1 in 4 chance.
Add new parameter 'room_size_large' to replace 'room_size_large_min'
and 'room_size_large_max', so that the mapgen has complete control
over this.
Biome-defined dungeon nodes was added as a feature to MT 5.0.0.
So now remove most of the hardcoded dungeon node code that assumes a
game has stone, sandstone, desert stone, and no other stone types.
If biome-defined dungeon nodes are not found, dungeon nodes fall back
to the 'cobble' mapgen alias if present, if not present they fall back
to biome-defined 'stone'.
Remove now-unnecessary mapgen aliases from MapgenBasic. Non-mgv6 games
now only need to define 3 to 5 mapgen aliases.
Document dungeon parameters.
Make c_lava_source fallback to c_water_source as both are used as cave
liquids.
They verify the provided value and error if a wrong value got provided
command line description for color was differnt on win32 but code did not handle any differenc
extended the command line description for world and worldname that it is clear that they only start a local game if used with --go
Fixes#7875
Rivers are disabled by default and will not be added to existing worlds.
Rewrite getSpawnLevelAtPoint() to be simpler and more consistent with
generateTerrain().
Add user-settable noise parameters for dungeon density to each mapgen,
except V6 which hardcodes this noise parameter.
Move the calculation of number of dungeons generated in a mapchunk out
of dungeongen.cpp and into mapgen code, to allow mapgens to generate
any desired number of dungeons in a mapchunk, instead of being forced
to have number of dungeons determined by a density noise.
This is more flexible and allows mapgens to use dungeon generation to
create custom structures, such as occasional mega-dungeons.
The old texture modifier is restored by passing `m_previous_texture_modifier`.
Either copy it manually or let the function parameter do that.
Victims so far:
8e0b80a Apr 2018
eb2bda7 May 2019
When multiple recipes are applicable, the recipes are prioritised in this order:
toolrepair < shapeless with groups < shapeless < shaped with groups < shaped
For cooking and fuel, items are prioritised over item groups
This allows games to specify biome cave liquids and avoid the old
hardcoded behaviour, but preserves the ability to have multiple
cave liquids in one biome, such as lava and water.
When multiple cave liquids are defined by the biome definition,
make each entire cave use a randomly chosen liquid, instead of
every small cave segment using a randomly chosen liquid.
Plus an optimisation:
Don't place nodes if cave liquid is defined as 'air'
* Optimize statbar drawing
The texture name of the statbar is a string passed by value.
That slows down the client and creates litter in the heap
as the content of the string is allocated there. Convert the
offending parameter to a const reference to avoid the
performance hit.
* Optimize texture cache
There is an unnecessary temporary created when the texture
path is being generated. This slows down the cache each time
a new texture is encountered and it needs to be loaded into
the cache. Additionally, the heap litter created by this
unnecessary temporary is particularly troublesome here as
the following code then piles another string (the resulting
full path of the texture) on top of it, followed by the
texture itself, which both are quite long term objects as
they are subsequently inserted into the cache where they can
remain for quite a while (especially if the texture turns
out to be a common one like dirt, grass or stone).
Use std::string.append to get rid of the temporary which
solves both issues (speed and heap fragmentation).
* Optimize animations in client
Each time an animated node is updated, an unnecessary copy of
the texture name is created, littering the heap with lots of
fragments. This can be specifically troublesome when looking
at oceans or large lava lakes as both of these nodes are
usually animated (the lava animation is pretty visible).
Convert the parameter of GenericCAO::updateTextures to a
const reference to get rid of the unnecessary copy.
There is a comment stating "std::string copy is mandatory as
mod can be a class member and there is a swap on those class
members ... do NOT pass by reference", reinforcing the
belief that the unnecessary copy is in fact necessary.
However one of the first things the code of the method does
is to assign the parameter to its class member, creating
another copy. By rearranging the code a little bit this
"another copy" can then be used by the subsequent code,
getting rid of the need to pass the parameter by value and
thus saving that copying effort.
* Optimize chat console history handling
The GUIChatConsole::replaceAndAddToHistory was getting the
line to work on by value which turns out to be unnecessary.
Get rid of that unnecessary copy by converting the parameter
to a const reference.
* Optimize gui texture setting
The code used to set the texture for GUI components was
getting the name of the texture by value, creating
unnecessary performance bottleneck for mods/games with
heavily textured GUIs. Get rid of the bottleneck by passing
the texture name as a const reference.
* Optimize sound playing code in GUIEngine
The GUIEngine's code receives the specification of the sound
to be played by value, which turns out to be most likely a
mistake as the underlying sound manager interface receives
the same thing by reference. Convert the offending parameter
to a const reference to get rid of the rather bulky copying
effort and the associated performance hit.
* Silence CLANG TIDY warnings for unit tests
Change "std::string" to "const std::string &" to avoid an
unnecessary local value copy, silencing the CLANG TIDY
process.
* Optimize formspec handling
The "formspec prepend" parameter was passed to the formspec
handling code by value, creating unnecessary copy of
std::string and slowing down the game if mods add things like
textured backgrounds for the player inventory and/or other
forms. Get rid of that performance bottleneck by converting
the parameter to a const reference.
* Optimize hotbar image handling
The code that sets the background images for the hotbar is
getting the name of the image by value, creating an
unnecessary std::string copying effort. Fix that by
converting the relevant parameters to const references.
* Optimize inventory deserialization
The inventory manager deserialization code gets the
serialized version of the inventory by value, slowing the
server and the client down when there are inventory updates.
This can get particularly troublesome with pipeworks which
adds nodes that can mess around with inventories
automatically or with mods that have mobs with inventories
that actively use them.
* Optimize texture scaling cache
There is an io::path parameter passed by value in the
procedure used to add images converted from textures,
leading to slowdown when the image is not yet created and
the conversion is thus needed. The performance hit is
quite significant as io::path is similar to std::string
so convert the parameter to a const reference to get rid of
it.
* Optimize translation file loader
Use "std::string::append" when calculating the final index
for the translation table to avoid unnecessary temporary
strings. This speeds the translation file loader up
significantly as std::string uses heap allocation which
tends to be rather slow. Additionally, the heap is no
longer being littered by these unnecessary string
temporaries, increasing performance of code that gets
executed after the translation file loader finishes.
* Optimize server map saving
When the directory structure for the world data is created
during server map saving, an unnecessary value passing of
the directory name slows things down. Remove that overhead
by converting the offending parameter to a const reference.
* Force send a mapblock to a player.
Send a single mapblock to a specific remote player.
This is badly needed for mods and games where players are teleported
into terrain which may be not generated, loaded, or modified
significantly since the last player visit.
In all these cases, the player currently ends up in void, air, or
inside blocks which not only looks bad, but has the effect that the
player might end up falling and then the server needs to correct for
the player position again later, which is a hack.
The best solution is to send at least the single mapblock that the
player will be teleported to. I've tested this with ITB which does this
all the time, and I can see it functioning as expected (it even shows
a half loaded entry hallway, as the further blocks aren't loaded yet).
The parameter is a blockpos (table of x, y, z), not a regular pos.
The function may return false if the call failed. This is most likely
due to the target position not being generated or emerged yet, or
another internal failure, such as the player not being initialized.
* Always send mapblock on teleport or respawn.
This avoids the need for mods to send a mapblock on teleport or
respawn, since any call to `player:set_pos()` will pass this code.
* Improve readability of debug menu by using '|'
* Restore whitespace to separate yaw and cardinal direction
Co-Authored-By: ClobberXD <ClobberXD@gmail.com>
* Optimize packet construction functions
Some of the functions that construct packets in
connection.cpp are using a const reference to get the raw
packet data to package and others use a value passed
parameter to do that. The ones that use the value passed
parameter suffer from performance hit as the rather bulky
packet data gets a temporary copy when the parameter is
passed before it lands at its final destination inside the
newly constructed packet. The unnecessary temporary copy
hurts quite badly as the underlying class (SharedBuffer)
actually allocates the space for the data in the heap.
Fix the performance hit by converting all of these value
passed parameters to const references. I believe that this
is what the author of the relevant code actually intended
to do as there is a couple of packet construction helper
functions that already use a const reference to get the
raw data.
* Optimize packet sender thread class
Most of the data sending methods of the packet sender thread
class use a value passed parameter for the packet data to be
sent. This causes the rather bulky data to be allocated on
the heap and copied, slowing the packet sending down. Convert
these parameters to const references to avoid the performance
hit.
* Optimize packet receiver thread class
The packet receiver and processor thread class has many
methods (mostly packet handlers) that receive the packed data
by value. This causes a performance hit that is actually
worse than the one caused by the packet sender methods
because the packet is first handed to the processPacket
method which looks at the packet type stored in the header
and then delegates the actual handling to one of the
handlers. Both, processPacket and all the handlers get the
packet data by value, leading to at least two unnecessary
copies of the data (with malloc and all the slow bells and
whistles of bulky classes).
As there already is a few methods that use a const reference
parameter for the packet data, convert all this value passed
packets to const references.
6125 is the time of first full light according to 'get_node_light()',
and the time of first full light visually when basic shaders are on.
This is the optimum default new world start time, taking all possible
games into account.
The previous time assumed a game similar to Minetest Game. Games
should set this setting themselves according to their needs.
It turns out there is no need to return the new value and
preserve the old one in random_turn, the procedure can be
made to modify the value in-place. This saves quite a bunch
of parameter and return value copying.
Previously, when basic shaders were enabled, the function
time_to_daynight_ratio() returned values jumping between 149 and 150
between times 4375 and 4625, and values jumping between 999 and 1000
between times 6125 and 6375, (and the corresponding times at sunset)
due to tiny float errors in the interpolation code.
This caused the light level returned by blend_light() to jump between
14 and 15, which became noticeable recently as those light levels were
given different visual brightnesses.
Add early returns to avoid the problematic interpolation, and to
avoid unnecessary running of the loop.
Makes the liquid waving shader per-nodedef like waving leaves/plants,
instead of being applied to all liquids.
Like the waving leaves/plants shaders, the liquid waving shader can
also be applied to meshes and nodeboxes.
Derived from a PR by t0ny2.
Like randomwalk caves, preserve nodes that have 'is_ground_content = false',
to avoid dungeons that generate out beyond the edge of a mapchunk destroying
nodes added by mods in 'register_on_generated()'.
Issue discovered by, and original PR by, argyle77.
This change permits to use up-to-date compilers, clang-tidy and
clang-format
It also refactor the tidy/format step to drop the binary selection from
scripts and perform it directly in travis
Shorter, simpler, clearer and more consistent with other mapgens,
while preserving functionality.
Base terrain shape is unchanged.
With the 'vary river depth' option disabled, river surface level
is unchanged.
Behaviour of the 4 heat/humidity/river depth options is very
slightly changed due to bugfixes and code cleanup (the mapgen is
'unstable').
Apply heat and humidity gradients above water_level instead of
above y = 0.
I removed the MapNode constructor which takes a nodename and gives the node's id or CONTENT_IGNORE
The code which used this constructor (two places) now handles the situation of not registered nodes correctly:
* minetest.set_node and similar functions make minetest crash when a not registered node is passed
* reverting a node with rollback aborts if the node is not registered
Use "append" method to construct the various game paths
instead of wasteful string concatenation. Additionally, use a
temporary to extract and reuse a result of a few common
subexpressions to further reduce the overhead.
The "what" parameter is being passed by value, most likely by
accident as the type is "const std::string". Convert it to a
reference by adding the missing "&".
* Fix color command line parameter ignorance
* coloured log: Support detecting the tty on windows
* Print an error message when setting something invalid as color mode instead of silently using mode never
* Revert "coloured log: Support detecting the tty on windows"
This reverts commit 4c9fc6366487ac0e6799e181796ca594797bb6f8.
It didn't work for travis and belongs to a separate PR
* Allow adjusting the log color with an environment variable
If --color is not passed to minetest, is used to decide on the log colorization.
Minetest settings can not be used instead of an environment variable because logs may appear before loading them.
* fix empty if body
This loop makes multiple passes over m_stack (type std::list) in order to remove all elements with a specified value. Replacing the loop with a call to std::list::remove does the same job, but in only one pass.
This sends the following header to a remote media server:
Referer: minetest://<server_name>:port
This was verified with CTF and the Minetest Public Remove Media
server. If the servername was a plain IPv6 address it will
contain `:` characters and will be encapsulated in `[]` to
be a valid URI.
The getS16NoEx() handler will return true unless there is a
`[num_emerge_threads]` line in the `minetest.conf` at which
point the excption handler part is reached. Due to the fact that
`defaultsettings.cpp` has a default value set for this setting,
that never will happen.
Because of this, the code will never check the number of threads on
the system, and keep `nthreads = 0`. If that happens, the value is
changed to `1` and only 1 emerge thread will be used.
The default should be set to `1` instead, due to the potential unsafe
consequences for the standard sqlite map files, but that should be a
separate commit that also adds documentation for that setting. This
commit focuses on removing this `hiding` bug instead.
* Drop the ID mapper, use a big u64 instead. This will permit to resync server ids properly with the manager code
* Modernize some code parts (std::unordered_map, auto)
* generate id on client part on U32_MAX + 1 ids, lower are for server ids
The reverted commit 968ce9af598024ec71e9ffb2d15c3997a13ad754
is suspected (through the use of bisection) of causing network slowdowns.
Revert for now as we are close to release.
Store the rotation in the node as a 4x4 transformation matrix internally (through IDummyTransformationSceneNode), which allows more manipulations without losing precision or having gimbal lock issues.
Network rotation is still transmitted as Eulers, though, not as matrix. But it will stay this way in 5.0.
