The first 'if' always evaluated to true, as it evaluated values which are the default
ones for Android and iOS respectively, so even if one of them was overridden, the other
one would be true.
Fixes#27658.
Network peers get_var/put_var
File get_var/store_var
GDScript/Mono/VisualScript bytes2var/var2bytes
Add MultiplayerAPI.allow_object_decoding member which deprecates PacketPeer.allow_object_decoding.
Break ABI compatibaility (API compatibility for GDNative).
Enum reference resolving will now search in the @GlobalScope if no class is specified and the enum cannot be resolved in the current class.
Added support for constant references in EditorHelp, e.g.: [constant KEY_ENTER] or [constant Control.FOCUS_CLICK]. It supports enum constants (the enum name must not be included).
Expands to Object.call, Object.set and Object.get for accessing members. This means it can also access members from scripts written in other languages, like GDScript.
With this change finally one can use compound collisions (like those created
by Gridmaps) without serious performance issues. The previous KinematicBody
code for Bullet was practically doing a whole bunch of unnecessary
calculations. Gridmaps with fairly large octant sizes (in my case 32) can get
up to 10000x speedup with this change (literally!). I expect the FPS demo to
get a fair speedup as well.
List of fixes and improvements:
- Fixed a general bug in move_and_slide that affects both GodotPhysics and
Bullet, where ray shapes would be ignored unless the stop_on_slope parameter
is disabled. Not sure where that came from, but looking at the 2D physics
code it was obvious there's a difference.
- Enabled the dynamic AABB tree that Bullet uses to allow broadphase collision
tests against individual shapes of compound shapes. This is crucial to get
good performance with Gridmaps and in general improves the performance
whenever a KinematicBody collides with compound collision shapes.
- Added code to the broadphase collision detection code used by the Bullet
module for KinematicBodies to also do broadphase on the sub-shapes of
compound collision shapes. This is possible thanks to the dynamic AABB
tree that was previously disabled and it's the change that provides the
biggest performance boost.
- Now broadphase test is only done once per KinematicBody in Bullet instead of
once per each of its shapes which was completely unnecessary.
- Fixed the way how the ray separation results are populated in Bullet which
was completely broken previously, overwriting previous results and similar
non-sense.
- Fixed ray shapes for good now. Previously the margin set in the editor was
not respected at all, and the KinematicBody code for ray separation was
complete bogus, thus all previous attempts to fix it were mislead.
- Fixed an obvious bug also in GodotPhysics where an out-of-bounds index was
used in the ray result array.
There are a whole set of other problems with the KinematicBody code of Bullet
which cost performance and may cause unexpected behavior, but those are not
addressed in this change (need to keep it "simple").
Not sure whether this fixes any outstanding Github issues but I wouldn't be
surprised.
User defined gizmos will haave higher preference than editor gizmos by
default. Also fixed some inconsistencies in the gizmos menu when using
custom gizmos.
Apparently we don't need to call mono_debug_close_image ourselves and we can call mono_image_close right away as it's not our duty to keep that reference.
Also fixed a wrong ifdef that was causing Mono to never be initialized if mscorlib was not found (which was the case with the utf8 assemblies path bug this commit fixes).
This condition was meant for exported projects only, not for the editor only.
When a singleton library was exposing NativeScript functionality,
the NativeScriptLanguage would attempt to terminate the library at
shutdown.
Since the GDNative module itself handles singleton libraries,
it closes all singleton libraries at shutdown as well. This double free
could cause a crash, since the library referenced would no longer be alive.
- This mode avoids loading any other resource.
- Search for class_name now uses this mode, to avoid loading in the scan
thread.
- Implement get_dependencies() for GDScript loader, now exporting
dependencies only should include the preloaded resources.
Previous version resulted in confusing (but actually right) errors about converting "from Object to Object", since CallError
does not include information about the actual types involved.
- Move "Mono" popup menu from the top right corner to `Projects -> Tools` as a submenu.
- Add "Build solution" button to the top right corner. Makes it more visible and quicker to access.
- Fix build list in the bottom panel unselect an item when clicking on empty space. Previously it would hide the issues panel but the item would remain selected, making it impossible to display the issues panel again if there was only one item.
This should fix the various issues with colours in Ogg Theora and WEBM playback.
(A reference project is attached to PR #26051, which this commit should be part of.)
This version of the commit, rather than moving x->RGBA handling into libsimplewebm, uses a colourspace field added to libsimplewebm by a PR there.
Thus, the commit that precedes this should be the synchronization & cleanup commit for that.
Also, this version is now clang-format friendly. I hope.
Recently, Dictionary::get() was introduced, which acts like a index
operator but allows the caller to specify a default value to return
instead of issuing an error.
This commit adds a new GDNative function that includes the default value.
Before the parser only checked if the catch-all branch has a return in
order to determine if the entire match block has a return.
This code block was assumed to always return.
match value:
"test":
print("test")
_:
return
Now as soon as one of the branches has no return, the entire match block
is marked to not have a return.
Scaling the depth was the wrong solution for the KinematicBody jitter
because it causes jitter with RigidBody. Instead scale the margin that
is ignored to allow KinematicBody to still pick up the ray shape in the
kinematic test when the shape is just at margin distance from another
one. This solution does not cause jitter with either KinematicBody or
RigidBody.
They had been reserved for future implementation, but we now have the
`match` CF keyword which does the same and more.
According to @reduz `do` was even added by mistake when copying from
the shader language parser, it was never intended to add support for
`do`... `while` loops, as the syntax would be awkward in GDScript,
and the added sugar is not worth it.
Fixes#25787.
mbedtls_ssl_read cannot be called with a NULL buffer even if len is 0,
as those are passed to memcpy and compilers doesn't like that.
Always pass a single byte (still len 0 so nothing is actually copied)
This property returns an instance of the singleton.
The purpose of this is to allow using methods from the base class like 'Connect'.
Since all Godot singletons inherit Object, the type of the returned instance is Godot.Object.
- Added code handling non-compound collision to recover_from_penetration_ray()
which is now needed due to the optimization avoiding the use of compound
collisions when only a single collision shape is used.
- Removed arbitrary margin applied in the collision algorithm of RayShapes
which causes jittered movement. For lack of a better replacement and for
lack of any explanation on why it has been introduced, it's now using the
shape's margin property instead which is small enough to not show visible
jitter.
- Tried to get rid of inconsistent uses of the collision margin.
- Removed hack from GodotDeepPenetrationContactResultCallback::addContactPoint
for RayShape collision as it's no longer needed as the collision algorithm
of RayShapes correctly calculates the contact normal for a while now.
Fixes#25227.
This is needed to avoid aborting due to missing mscorlib for projects that do not use C#.
If 'res://.mono/' exists, then we assume the project uses C#, in which case a missing mscorlib should still abort.
Only possible if the object class is a "native type". If the object class is a user class (that derives a "native type") then a script is needed.
Since CSharpLanguage does cleanup of script instance bindings when finished, cases like #25621 will no longer cause problems.
Fixed ~Object() trying to free script instance bindings after the language has already been removed, which would result in a NULL dereference.
Previously this was only done when building the script for running the game. This was a problem because the user could want to build the project manually with the "Build project" button, to then run the game from the command line or similar.
When a recursive declaration ends up in a GDScript file the
_guess_expression_type function would start looping and eventually run
out of stack space.
We now cap recusion for this function to 100 frames.
This fixes#25598
Avoid CSharpInstance from accessing its state after self destructing (by deleting the Reference owner).
It's now safe to replace the script instance without leaking or crashing.
Also fixed godot_icall_Object_weakref return reference being freed before returning.
The default value of the type is now used to initialise it.
export(int) A
Will now have A be 0 istead of Null even though it still showed as 0 before in the inspector, fixes#25357
This allows most demos to run without any ubsan or asan errors. There
are still some things in thirdpart/ and some things in AudioServer that
needs a look but this fixes a lot of issues. This should help debug less
obvious issues, hopefully.
This fixes#25217 and fixes#25218
To help users writing good cross-platform code, Godot's
`FileAccessWindows:open()` will issue a warning on case mismatch, which
happens here with capitalized extensions given by `PATHEXT` compared to
actual file extensions which are lowercase 99% of the time.
Fixes#25368.
Some construct (like match) actually depends on the second pass. This
adds some extra checks to not perform specific type-checks on release
since not all type information is available.
BaseIntermediateOutputPath seems to be empty by default. The workaround is to explicitly set it.
Also fixed passing char instead of char[] to String.Split. Why was this even working with Mono?
When exporting variables from a gdscript, default values of uninitialized variables would never be set. This caused the default value to be Variant::NIL, and when a user tried to reset the variable through the editor, an error would be thrown because too few arguments would be counted(end of argument list for calls are detected by NIL values).
Fixed by simply setting default value to an empty variant of the proper type in gdscript parser.
Some classes are represented internally with an underscore prefix, so we
need to make sure we match this representation when type-checking,
otherwise the check might fail on a valid scenario.
There's always a constructor, even if implicit, especially for native
types.
Also don't check for signature match on function call, since this
information is not available in release builds.
A lot of information is missing on release, and the checks might take a
performance hit. Also, having GDScript more lenient on release is
usually desirable.
Avoid sending encoded packet flags (reliable/unreliable/ordered) as
that's already been done by ENet itself and we can read them from the
incoming packet.
Use temporary cache directory instead of editor settings directory
in order to resolve encrypted file access needed for encrypting scripts
on all platforms.
Some used 'is_valid()' checks, others not. Validity is already checked in 'unref()',
and 'remove_resource_format_*()' has an ERR_FAIL condition on 'is_null()' already
(which shouldn't happen since we're only unregistering things that we previously
registered.
Also add missing GDCLASS statement in ResourceFormatLoaderVideoStreamGDNative,
missed in #20552 which was last amended before #19501 was merged.
During reloading in `GDScriptLanguage::reload_all_scripts` a placeholder instance that must remain so is replaced with a new placeholder instance. The state is then restored by calling `ScriptInstance::set` for each property. This does not work if the script is missing the properties due to build/parse failing.
The fix for such cases is to call `placeholder_set_fallback` instead of `set` on the script instance.
I took this chance to move the `build_failed` flag from `PlaceHolderScriptInstance` to `Script`. That improves the code a lot. I also renamed it to `placeholder_fallback_enabled` which is a much better name (`build_failed` could lead to misunderstandings).
Godot supports many different compilers and for production releases we
have to support 3 currently: GCC8, Clang6, and MSVC2017. These compilers
all do slightly different things with -ffast-math and it is causing
issues now. See #24841, #24540, #10758, #10070. And probably other
complaints about physics differences between release and release_debug
builds.
I've done some performance comparisons on Linux x86_64. All tests are
ran 20 times.
Bunnymark: (higher is better)
(bunnies) min max stdev average
fast-math 7332 7597 71 7432
this pr 7379 7779 108 7621 (102%)
FPBench (gdscript port http://fpbench.org/) (lower is better)
(ms)
fast-math 15441 16127 192 15764
this pr 15671 16855 326 16001 (99%)
Float_add (adding floats in a tight loop) (lower is better)
(sec)
fast-math 5.49 5.78 0.07 5.65
this pr 5.65 5.90 0.06 5.76 (98%)
Float_div (dividing floats in a tight loop) (lower is better)
(sec)
fast-math 11.70 12.36 0.18 11.99
this pr 11.92 12.32 0.12 12.12 (99%)
Float_mul (multiplying floats in a tight loop) (lower is better)
(sec)
fast-math 11.72 12.17 0.12 11.93
this pr 12.01 12.62 0.17 12.26 (97%)
I have also looked at FPS numbers for tps-demo, 3d platformer, 2d
platformer, and sponza and could not find any measurable difference.
I believe that given the issues and oft-reported (physics) glitches on
release builds I believe that the couple of percent of tight-loop
floating point performance regression is well worth it.
This fixes#24540 and fixes#24841
This fixes the previously wrong PR
Because we don't actually ship 'debug' templates to users make sure
the mono exporter picks the correct 'data' directory for export
templates.
This fixes#24752
Because we don't actually ship 'debug' templates to users make sure
the mono exporter picks the correct 'data' directory for export
templates.
This fixes#24752
Retrieved working implementation from 2.1 branch and adapted to
existing export preset system.
Added Script tab in export preset to export script as raw text,
compiled, or encrypted (same as in 2.1). The script encryption key is
visually validated. The script export mode and the key is saved per
per preset in `export_presets.cfg`, so it makes sense to ignore this
file in version control system.
Each custom exporting procedure can retrieve an export preset set
during project exporting. Refactored project export dialog a bit to
allow easier code comprehension.
- The `cpu-features.{c,h}` code was only used by chance by the webm
(libvpx) code, so I moved it there. It was actually introduced before
that and wasn't in use, and libvpx just happened to be able to
compile thanks to it being bundled.
It could potentially be compiled on the fly from the Android NDK, but
since we plan to replace the webm module by a GDNative plugin in the
near future, I went the bundling route.
- `ifaddrs_android.h` is already provided in the Android NDK as
`ifaddrs.h`, same as on other Unixes. Yet we cannot use it until we
up the min API level to 24, where `getifaddrs` is first defined.
I moved the files to `thirdparty/misc` and synced them with upstream
WebRTC (only indentation changes and removal of `static` qualifiers).
Also removes dropped thirdparty files from COPYRIGHT.txt after changes
in #24105 and #24145.
By introducing an intermediate proxy class for the array subscript
operator for String and CharString we can control better when CowData
will actually CoW.
This should improve performance of String usage for most cases.
Interface and callback api added for Videodecoder support.
Should be able to construct any format videodecoder using
only the given interface.
GSoC 2018 project.
str2var used to raise a blocking error when invalid input was passed. Now it logs an error message and
returns the input string. This solution was proposed in #13021.
Closes#11457 and #13021.
Fixes some out-of-bounds problems when there are more collisions than
have been requested for. This extends what has been done in 444b8471a3.
Fixesgodotengine/godot#23990
ForLoveOfCats
qarmin
Rémi Verschelde
Ignacio Etcheverry
marxin
Fabio Alessandrelli
volzhs
Daniel Rakos
Leonard
Thomas Herzog
Ignacio Roldán Etcheverry
Vivatchai Kaveeta
thomas.herzog
Sebastian Hartte
PouleyKetchoupp
karroffel
Timo Schwarzer
Bojidar Marinov
JFonS
Anish
Hugo Locurcio
Chaosus
2017A7PS0002G
Marcin Zawiejski
clayjohn
Juan Linietsky
George Marques
Juan Linietsky
Carter Anderson
QbieShay
Eric Rybicki
20kdc
Michael Alexsander Silva Dias
Andrea Catania
Daniel Spaniol
RomanAkberov
Aaron Franke
Andrii Doroshenko (Xrayez)
Hein-Pieter van Braam
Daniel Abbott
Will Nations
DualMatrix
Marc Gilleron
Martin Capitanio
Allan Daemon
jlahman
supatier
Shinryuuji
Kelly Thomas
Mateus Felipe C. C. Pinto
meditator
corrigentia
Ben Rog-Wilhelm
Bastiaan Olij
Aron Castro
Erik
Leon Krause
Florian Jung