Godot Wiki
Godot Wiki
Godot Wiki
Release latest
2 Engine 59
3 2D tutorials 87
4 3D tutorials 159
5 Networking 225
7 Miscellaneous 233
10 Languages 791
12 Compiling 831
13 Advanced 853
14 Contributing 885
i
ii
Godot Engine Documentation, Release latest
The main documentation for the site is organized into a few sections:
• Tutorials
• Reference
• Community
Tutorials 1
Godot Engine Documentation, Release latest
2 Tutorials
CHAPTER 1
1.1.1 Introduction
Imagine for a second that you are not a game developer anymore. Instead, You are a chef! Change your hipster outfit
for a toque and a double breasted jacket. Now, instead of making games, you create new and delicious recipes for your
guests.
So, how does a chef create a recipe? Recipes are divided in two sections, the first is the ingredients and the second is
the instructions to prepare it. This way, anyone can follow the recipe and savor your magnificent creation.
Making games in Godot feels pretty much the same way. Using the engine feels like being in a kitchen. In this kitchen,
nodes are like a refrigerator full of fresh ingredients to cook with.
There are many types of nodes, some show images, others play sound, other nodes display 3D models, etc. There’s
dozens of them.
1.1.2 Nodes
But let’s go to the basics. A node is a basic element for creating a game, it has the following characteristics:
• Has a name.
3
Godot Engine Documentation, Release latest
p=. |image1|
The last one is very important. Nodes can have other nodes as children. When arranged in this way, the nodes
become a tree.
In Godot, the ability to arrange nodes in this way creates a powerful tool for organizing the projects. Since different
nodes have different functions, combining them allows to create more complex functions.
This is probably not clear yet and it makes little sense, but everything will click a few sections ahead. The most
important fact to remember for now is that nodes exist and can be arranged this way.
1.1.3 Scenes
Now that the existence of nodes has been defined, the next logical step is to explain what a Scene is.
A scene is composed of a group of nodes organized hierarchically (in tree fashion). It has the following properties:
Theory is boring, so let’s change subject and go practical. Following a long tradition in tutorials, the first project will
be a hello world. For this, the editor will be used.
When godot executable is run outside a project, the Project Manager appears. This helps developers manage their
projects.
To create a new project, the “New Project” option must be used. Choose and create a path for the project and specify
the project name:
1.1.5 Editor
Once the “New Project” is created, the next step is opening it. This will open the Godot editor. Here is how the editor
looks when freshly opened:
As mentioned before, making games in Godot feels like being in a kitchen, so let’s open the refrigerator and add some
fresh nodes to the project. We’ll begin with a Hello World! To do this, the “New Node” button must be pressed:
This will open the Create Node dialog, showing the long list of nodes that can be created:
From there, select the “Label” node first. Searching for it is probably the quickest way:
And finally, create the Label! A lot happens when Create is pressed:
First of all, the scene is changed to the 2D editor (because Label is a 2D Node type), and the Label appears, selected,
at the top left corner of the viewport.
The node appears in the scene tree editor (box in the top right corner), and the label properties apear in the Inspector
(box in the bottom right corner).
The next step, will be to change the “Text” Property of the label, let change it to “Hello, World!”:
Ok, everything’s ready to run the scene! Press the PLAY SCENE Button on the top bar (or hit F6):
Aaaand.. Oops.
Scenes need to be saved to be run, so save the scene to something like hello.scn in Scene -> Save:
And here’s when something funny happens. The file dialog is a special file dialog, and only allows to save inside the
project. The project root is “res://” which means “resource path. This means that files can only be saved inside the
project. For the future, when doing file operations in Godot, remember that “res://” is the resource path, and no matter
the platform or install location, it is the way to locate where resource files are from inside the game.
After saving the scene and pressing run scene again, the “Hello, World!” demo should finally execute:
Success|image15|
Ok, It’s time to do some configuration to the project. Right now, the only way to run something is to execute the
current scene. Projects, however, have several scenes so one of them must be set as the main scene. This scene is the
one that will be loaded at the time the project is run.
These settings are all stored in the engine.cfg file, which is a plaintext file in win.ini format, for easy editing. There
are dozens of settings that can be set in that file to alter how a project executes, so to make matters simpler, a project
setting dialog exists, which is sort of a frontend to editing engine.cfg
To access that dialog, simply go to Scene -> Project Settings.
Once the window opens, the task will be to select a main scene. This can be done easily by changing the applica-
tion/main_scene property and selecting ‘hello.scn’.
With this change, pressing the regular Play button (or F5) will run the project, no matter which scene is being edited.
Going back to the project settings dialog. This dialog provides a lot of options that can be added to engine.cfg and
show their default values. If the default value is ok, then there isn’t any need to change it.
When a value is changed, a tick is marked to the left of the name. This means that the property will be saved to the
engine.cfg file and remembered.
As a side note, for future reference and a little out of context (this is the first tutorial after all!), it is also possible to
add custom configuration options and read them in run-time using the Globals singleton.
1.1.7 To Be Continued...
This tutorial talks about “Scenes and Nodes”, but so far there has been only one scene and one node! Don’t worry, the
next tutorial will deal with that...
1.2 Instancing
1.2.1 Rationale
Having a scene and throwing nodes to it might work for small projects, but as a project grows, more and more nodes
are used and it can quickly become unmanageable. To solve this, Godot allows a project to be separated in several
scenes. This, however, does not work the same way as in other game engines. In fact, it’s quite different, So please do
not skip this tutorial!
To recap: A scene is a collection of nodes organized as a tree, where they can have only one single node as the tree
root.
In Godot, a scene can be created and saved it to disk. As many scenes can be created and saved as desired.
Afterwards, while editing an existing or a new scene, other scenes can be instanced as part of it:
In the above picture, Scene B was added to Scene A as an instance. It may seem weird at first, but at the end of this
tutorial it will make complete sense!
Simply browse to inside the project location and open the “engine.cfg” file. The new project will appear on the list of
projects. Edit the project by using the ‘Edit’ option.
This project contains two scenes “ball.scn” and “container.scn”. The ball scene is just a ball with physics, while
container scene has a nicely shaped collision, so balls can be thrown in there.
p=. |image4|
|image5|
1.2. Instancing 13
Godot Engine Documentation, Release latest
Afterwards, push the ‘+’ shaped button, this is the instancing button!
Select the ball scene (ball.scn), the ball should appear in the origin (0,0), move it to around the center
of the scene, like this:
1.2. Instancing 15
Godot Engine Documentation, Release latest
There can be as many instances as desired in a scene, just try instancing more balls, or duplicating them (ctrl-D or
duplicate button):
1.2. Instancing 17
Godot Engine Documentation, Release latest
Select one of the many copies of the balls and go to the property editor. Let’s make it bounce a lot more, so look for
the bounce parameter and set it to 1.0:
The next it will happen is that a green “revert” button appears. When this button is present, it means we modified a
property from the instanced scene to override for a specific value in this instance. Even if that property is modified in
the original scene, the custom value will always overwrite it. Pressing the revert button will restore the property to the
original value that came from the scene.
1.2.5 Conclusion
Instancing seems handy, but there is more to it than it meets the eye! The next part of the instancing tutorial should
cover the rest..
1.3.1 Recap
Instancing has many handy uses. At a glance, with instancing you have:
• The ability to subdivide scenes and make them easier to manage.
• A more flexible alternative to prefabs (and much more powerful given instances work at many levels).
• A way to design more complex game flows or even UIs (UI Elements are nodes in Godot too).
But the real strong point of instancing scenes is that it works as an excellent design language. This is pretty much what
makes Godot special and different to any other engine out there. All the engine was designed from the ground around
this concept.
When making games with Godot, the recommended approach is to leave aside other design patterns such as MVC or
Entity-Relationship diagrams and start thinking games in a more natural way. Start by imagining the visible elements
in a game, the ones that can be named not by just a programmer but by anyone.
For example, here’s how a simple shooter game can be imagined:
It’s pretty easy to come up with a diagram like this for almost any kind of game. Just write down the elements that
come to mind, and then the arrows that represent ownership.
Once this diagram exists, making a game is about creating a scene for each of those nodes, and use instancing (either
by code [STRIKEOUT:more of that later] or from the editor) to represent ownership.
Most of the time programming games (or software in general) is spent designing an architecture and fitting game
components to that architecture. Designing based on scenes replaces that and makes development much faster and
more straightforward, allowing to concentrate on the game itself. Scene/Instancing based design is extremely efficient
at saving a large part of that work, since most of the components designed map directly to a scene. This way, none or
little architectural code is needed.
The following is a more complex example, an open-world type of game with lots of assets and parts that interact:
Make some rooms with furniture, then connect them. Make a house later, and use those rooms are the interior.
The house can be part of a citadel, which has many houses. Finally the citadel can be put on the world map terrain.
Add also guards and other NPCs to the citadel by previously creating their scenes.
With Godot, games can grow as quickly as desired, as only more scenes have to be made and instanced. The editor
UI is also designed to be operated by non programmers too, so an usual team development process involves 3D or 2D
artists, level designers, game designers, animators, etc all working with the editor interface.
Do not worry to much, the important part of this tutorial is to create awareness on how scenes and instancing are used
in real life. The best way to understand all this is to make some games.
Everything will become very obvious when put to practice, so, please do not scratch your head and go on to the next
tutorial!
1.4 Scripting
1.4.1 Introduction
Much has been said about tools that allow users to create video games without programming. It’s been a dream for
many independent developers to create games without learning how to code. This need has been around for a long
time, even inside companies, where game designers wish to have more control of the game flow.
Many products have been shipped promising a no-programming environment, but the result is often incomplete, too
complex or inefficient compared to traditional code. As a result, programming is here to stay for a long time. In fact,
the general direction in game engines has been to add tools that try to reduce the amount of code that needs to be
written for specific tasks, to speed up development.
In that sense, Godot has taken some useful design decisions towards that goal. The first and most important is the
scene system. The aim of it is not obvious at first, but works well later on. That is, to relieve programmers from the
responsibility of architecting code.
When designing games using the scene system, the whole project is fragmented in complementary scenes (not indi-
vidual ones). Scenes complement each other, instead of being separate. There will be plenty of examples about this
later on, but it’s very important to remember it.
For those with a good amount of programming expertise, this means a different design pattern to MVC. Godot promises
efficiency at the expense of dropping the MVC habits, which are replaced by the scenes as a complement pattern.
Godot also uses the extend pattern for scripting, meaning that scripts extends from all the available engine classes.
1.4.2 GDScript
[[GDScript]] (click link for reference) is a dynamically typed scripting language to fit inside Godot. It was designed
with the following goals:
• First and most importantly, making it simple, familiar and as easy to learn as possible.
• Making the code readable and error safe. The syntax is mostly borrowed from Python.
Programmers generally take a few days to learn it, and within two weeks feel comfortable with it.
As with most dynamically typed languages though, the higher productivity (code is easier to learn, faster to write,
no compilation, etc) is balanced with a performance penalty, but most critical code is written in C++ already in the
engine (vector ops, physics, math, indexing, etc), making the resulting performance more than enough for most types
of games.
In any case, if more performance is required, critical sections can be rewritten in C++ and exposed transparently to the
script. This allows for replacing a GDScript class with a C++ class without altering the rest of the game.
Before continuing, please make sure to read the [[GDScript]] reference. It’s a simple language and the reference is
short, should not take more than a few minutes to glance.
Scene Setup
This tutorial will begin by scripting a simple GUI scene. Use the add node dialog to create the following hierarchy,
with the following nodes:
• Panel * Label * Button
It should look like this in the scene tree:
And try to make it look like this in the 2D editor, so it makes sense:
1.4. Scripting 21
Godot Engine Documentation, Release latest
Adding a Script
Select the Panel node, then press the “Add Script” Icon as follows:
The script creation dialog will popup. This dialog allows to select the language, class name, etc. GDScript does not
use class names in script files, so that field is not editable. The script should inherit from “Panel” (as it is meant to
extend the node, which is of Panel type, this is automatically filled anyway).
Select the filename for the script (if you saved the scene previously, one will be automatically generated as sayhello.gd)
and push “Create”:
Once this is done, the script will be created and added to the node. You can see this both as an extra icon in the node,
as well as in the script property:
1.4. Scripting 23
Godot Engine Documentation, Release latest
To edit the script, pushing the icon above should do it (although, the UI will take you directly to the Script editor
screen). So, here’s the template script:
There is not much in there. The “_ready()” function is called when the node (and all it’s children) entered the active
scene. (Remember, It’s not a constructor, the constructor is “_init()” ).
A script basically adds a behavior to a node. It is used to control the node functions as well as other nodes (children,
parent, siblings, etc). The local scope of the script is the node (just like in regular inheritance) and the virtual functions
of the node are captured by the script.
1.4. Scripting 25
Godot Engine Documentation, Release latest
Handling a Signal
Signals are used mostly in GUI nodes, (although other nodes have them too). Signals are “emitted” when some specific
kind of action happens, and can be connected to any function of any script instance. In this step, the “pressed” signal
from the button will be connected to a custom function.
There is a GUI for connecting signals, just select the node and press the “Signals” button:
But this example will not use it. We don’t want to make things too easy. So please close that screen!
In any case, at this point it is clear that that we are interested in the “pressed” signal, so instead of doing it with the
visual interface, the connection will be done using code.
For this, there is a function that is probably the one that Godot programmers will use the most, this is get_node(). This
function uses paths to fetch nodes in the current tree or anywhere in the scene, relative to the node holding the script.
To fetch the button, the following must be used:
get_node("Button")
So, next, a callback will be added for when a button is pressed, that will change the label’s text:
func _on_button_pressed():
get_node("Label").set_text("HELLO!")
Finally, the button “pressed” signal will be connected to that callback in _ready(), by using connect.
func _ready():
get_node("Button").connect("pressed",self,"_on_button_pressed")
1.4. Scripting 27
Godot Engine Documentation, Release latest
# var a=2
# var b="textvar"
func _on_button_pressed():
get_node("Label").set_text("HELLO!")
func _ready():
get_node("Button").connect("pressed",self,"_on_button_pressed")
Running the scene should have the expected result when pressing the button:
Note: As it is a common mistake in this tutorial, let’s clarify again that get_node(path) works by returning the imme-
diate children to the node controlled by the script (in this case, Panel), so Button must be a child of Panel for the above
code to work. To give this clarification more context, if Button was a child of Label, the code to obtain it would be:
# not for this case
# but just in case
get_node("Label/Button")
And, also, try to remember that nodes are referenced by name, not by type.
1.5.1 Processing
Several actions in Godot are triggered by callbacks or virtual functions, so there is no need to check for writing code
that runs all time time. Additionally, a lot can be done with animation players.
However,it is still a very common case to have a script process on every frame. There are two types of processing, idle
processing and fixed processing.
Idle processing is activated with the Node.set_process() function. Once active, the Node._process() callback will be
called every frame. Example:
func _ready():
set_process(true)
func _process(delta):
#do something..
The delta parameter describes the time elapsed (in seconds, as floating point) since the previous call to _process().
Fixed processing is similar, but only needed for synchronization with the physics engine.
A simple way to test this is to create a scene with a single Label node, with the following script:
extends Label
var accum=0
func _ready():
set_process(true)
func _process(delta):
accum+=delta
set_text(str(accum))
1.5.2 Groups
Nodes can be added to groups (as many as desired per node). This is a simple yet useful feature for organizing large
scenes. There are two ways to do this, the first is from the UI, from tne Groups button:
And the second from code. One useful example would be, for example, to tag scenes which are enemies.
func _ready():
add_to_group("enemies")
This way, if the player, sneaking into the secret base, is discovered, all enemies can be notified about the alarm
sounding, by using SceneTree.call_group():
func _on_discovered():
get_tree().call_group(0,"guards","player_was_discovered")
The above code calls the function “player_was_discovered” on every member of the group “guards”.
Optionally, it is possible to get the full list of “guards” nodes by calling SceneTree.get_nodes_in_group():
var guards = get_tree().get_nodes_in_group("guards")
1.5.3 Notifications
Godot has a system of notifications. This is usually not needed to be used from scripting, as it’s too low level and
virtual functions are provided for most of them. It’s just good to know they exists. Simply add a Object._notification()
function in your script:
func _notification(what):
if (what==NOTIFICATION_READY):
print("This is the same as overriding _ready()...")
elif (what==NOTIFICATION_PROCESS):
var delta = get_process_time()
print("This is the same as overriding _process()...")
The documentation of each class in the class list shows the notifications it can receive. However, again, for most cases
script provides simpler overrideable functions.
As mentioned before, it’s better to use these functions. Nodes provide many useful overrideable functions, which are
described as follows:
func _enter_tree():
pass # When the node enters the _Scene Tree_, it become acive and this function is called. Childr
func _ready():
pass # This function is called after _enter_tree, but it ensures that all children nodes have also
func _exit_tree():
pass # When the node exists the _Scene Tree_, this function is called. Children nodes have all exi
func _process(delta):
pass # When set_process() is enabled, this is called every frame
func _fixed_process(delta):
pass # When set_fixed_process() is enabled, this is called every physics frame
func _paused():
pass # Called when game is paused, after this call, the node will not receive any more process cal
func _unpaused():
pass # Called when game is unpaused
To create a node from code, just call the .new() method, (like for any other class based datatype). Example:
var s
func _ready():
s = Sprite.new() # create a new sprite!
add_child(s) #add it as a child of this node
When a node is freed, it also frees all it’s children nodes. Because of this, manually deleting nodes is much simpler
than it appears. Just free the base node and everything else in the sub-tree goes away with it.
However, it might happen very often that we might want to delete a node that is currently “blocked” this means, the
node is emitting a signal or calling a function. This will result in crashing the game. Running Godot in the debugger
often will catch this case and warn you about it.
The safest way to delete a node is by using queue_free() instead. This erases the node during idle, safely.
func _someaction():
s.queue_free() # remove the node and delete it while nothing is happening
Instancing a scene from code is pretty easy and done in two steps. The first one is to load the scene from disk.
var scene = load("res://myscene.scn") # will load when the script is instanced
But ‘scene’ is still not a node containing subnodes. It’s packed in a special resource called PackedScene. To create the
actual node, the function PackedScene.instance() must be called. This will return the tree of nodes that can be added
to the active scene:
var node = scene.instance()
add_child(node)
The advantage of this two-step process is that a packed scene may be kept loaded and ready to use, so it can be used
to create as many instances as desired. This is specially useful, for example, to instance several enemies, bullets, etc.
quickly in the active scene.
1.6.1 Pong
In this simple tutorial, a basic game of Pong will be created. There are plenty of more complex examples in the demos
included with the engine, but this should get introduced to basic functionality for 2D Games.
1.6.2 Assets
For the sake of the old times, the game will be in 640x400 pixels resolution. This can be configured in the Project
Settings (see previous tutorials). The default background color should be set to black:
Create a [[class_node2d]] node for the project root. Node2D is the base type for the 2D engine. After this, add some
sprites ([[class_sprite]] node) and set each to the corresponding texture. The final scene layour should look similar to
this (note: the ball is in the middle!):
Save the scene as “pong.scn” and set it as the main scene in the project properties.
There are so many input methods for video games... Keyboard, Joypad, Mouse, Touchscreen (Multitouch). Yet this is
pong. The only input that matters is for the pads going up and down.
Handling all possible input methods can be very frustrating and take a lot of code. The fact that most games allow
controller customization makes this worse. For this, Godot created the “Input Actions”. An action is defined, then
input methods that trigger it are added.
Open the project properties dialog again, but this time move to the “Input Map” tab.
On it, add 4 actions: “left_move_up”,”left_move_down”,”right_move_up”,”right_move_down”. Assign the keys that
you desire. A/Z for left and Up/Down as keys should work in most cases.
1.6.5 Script
Create a script for the root node of the scene and open it (should have been explained in the previous tutorial!). The
script will inherit Node2D:
extends Node2D
func _ready():
pass
In the constructor, two things will be done. The first is to enable processing, and the second to store some useful
values. Such values are the dimensions of the screen and the pad:
extends Node2D
var screen_size
var pad_size
func _ready():
screen_size = get_viewport_rect().size
pad_size = get_node("left").get_texture().get_size()
set_process(true)
var ball_speed = 80
#direction of the ball (normal vector)
Get some useful values for computation. The first is the ball position (from the node), the second is the rectangles
(Rect2) of the pads. Sprites by defaut center the textures, so a small adjustment of size/2 must be added.
var ball_pos = get_node("ball").get_pos()
var left_rect = Rect2( get_node("left").get_pos() - pad_size/2, pad_size )
var right_rect = Rect2( get_node("right").get_pos() - pad_size/2, pad_size )
Then, now that the ball has a new position, it should be tested against everything. First, the floor and the roof:
if ( (ball_pos.y<0 and direction.y <0) or (ball_pos.y>screen_size.y and direction.y>0)):
direction.y = -direction.y
If one of the pads was touched, change direction and increase speed a little.
if ( (left_rect.has_point(ball_pos) and direction.x < 0) or (right_rect.has_point(ball_pos) and direc
direction.x=-direction.x
ball_speed*=1.1
direction.y=randf()*2.0-1
direction = direction.normalized()
If the ball went out of the screen, it’s game over. Game restarts:
if (ball_pos.x<0 or ball_pos.x>screen_size.x):
ball_pos=screen_size*0.5 #ball goes to screen center
ball_speed=80
direction=Vector2(-1,0)
Once everything was done with the ball, the node is updated with the new position:
get_node("ball").set_pos(ball_pos)
Only updating the pads according to player input. the Input class is really useful here:
#move left pad
var left_pos = get_node("left").get_pos()
get_node("left").set_pos(left_pos)
get_node("right").set_pos(right_pos)
And that’s it! a simple Pong was written with a few lines of code.
1.7.1 Introduction
If there is something that most programmers hate with passion, that is programming graphical user interfaces (GUIs).
It’s boring, tedious and unchallenging. Several aspects make matters worse such as:
• Pixel alignment of UI elements is difficult (so it looks just like the designer intends).
• UIs are changed constantly due to design and usability issues that appear during testing.
• Handling proper screen re-sizing for different display resolutions.
• Animating several screen components, to make it look less static.
GUI programming is one of the leading causes of programmer burnout. During the development of Godot (and
previous engine iterations), several techniques and philosophies for UI development were put in practice, such as
immediate mode, containers, anchors, scripting, etc. This was always done with the main goal of reducing the stress
programmers had to face while putting together user interfaces.
In the end, the resulting UI subsystem in Godot is an efficient solution to this problem, and works by mixing together a
few different approaches. While the learning curve is a little steeper than in other toolkits, developers can put together
complex user interfaces in very little time, by sharing the same set of tools with designers and animators.
1.7.2 Control
The basic node for UI elements is Control (sometimes called “Widget” or “Box” in other toolkits). Every node that
provides user interface functionality descends from it.
When controls are put in a scene tree as a child of another control, it’s coordinates (position, size) are always relative
to the parent. This sets the basis for editing complex user interface quickly and visually.
Controls receive input events by means of the _input_event() callback. Only one control, the one in focus, will receive
keyboard/joypad events (see set_focus_mode() and grab_focus().
Mouse Motion events are received by the control directly below the mouse pointer. When a control receives a mouse
button pressed event, all subsequent motion events are received by the pressed control until that button is released,
even if the pointer moves outside the control boundary.
Like any class that inherits from CanvasItem (Control does), a _draw() callback will be received at the begining and
every time the control needs to be redrawn (programmer needs to call update() to enqueue the CanvasItem for redraw).
If the control is not visible (yet aother CanvasItem property), the control does not receive any input.
In general though, the programmer does not need to deal with drawing and input events directly when building UIs,
(that is more useful when creating custom controls). Instead, controls emit different kinds of signals with contextural
information for when action occurs. For example, a Button emits a “pressed” signal when pressed, a Slider will emit
a “value_changed” when dragged, etc.
Before going into more depth, creating a custom control will be a good way to get the picture on how controls works,
as they are not as complex as it might seem.
Additionally, even though Godot comes with dozens of controls for different purposes, it happens often that it’s just
easier to attain a specific functionality by creating a new one.
To begin, create a single-node scene. The node is of type “Control” and has a certain area of the screen in the 2D
editor, like this:
var tapped=false
func _draw():
func _input_event(ev):
Then run the scene. When the rectangle is clicked/taped, it will go from blue to red. That synnergy between the events
and drawing is pretty much how most controls work internally.
1.7.5 UI Complexity
As mentioned before, Godot includes dozens of controls ready for using in a user interface. Such controls are divided
in two categories. The first is a small set of controls that work well for creating most game user interfaces. The second
(and most controls are of this type) are meant for complex user interfaces and uniform skinning trough styles. A
description is presented as follows to help understand which one should be used in which case.
This set of controls is enough for most games, where complex interactions or ways to present information are not
necessary. The can be skinned easily with regular textures.
• Label : Node used for showing text.
• TextureFrame : Displays a single texture, which can be scaled or kept fixed.
• TextureButton : Displays a simple texture buttons, states such as pressed, hover, disabled, etc can be set.
• TextureProgress : Displays a single textured progress bar.
Additionally, re-positioning of controls is most efficiently done with anchors in this case (see the [[GUI Repositioning]]
tutorial for more info).
In any case, it will happen often that even for simple games, more complex UI behaviors will be required. An example
of this is a scrolling list of elements (for a high score table, for example), which needs a ScrollContainer and a
VBoxContainer. These kind of more advanced controls can be mixed with the regular ones seamlessly (they are all
controls anyway).
The rest of the controls (and there are dozens of them!) are meant for another set of scenarios, most commonly:
• Games that require complex UIs, such as PC RPGs, MMOs, strategy, sims, etc.
• Creating custom development tools to speed up content creation.
• Creating Godot Editor Plugins, to extend the engine functionality.
Re-positioning controls for these kind of interfaces is more commonly done with containers (see the [[GUI Reposi-
tioning]] tutorial for more info).
1.8.1 Tutorial
This will be a simple tutorial to cement the basic idea of how the GUI subsystem works. The goal will be to create a
really simple, static, splash screen.
Following is a file with the assets that will be used:
attachment:robisplash_assets.zip
1.8.2 Setting Up
Create a scene with screen resolution 800x450, and set it up like this:
The nodes ‘background” and “logo” are of TextureFrame type. These have a special property for setting the texture to
be displayed, just load the corresponding file.
The node “start” is a TextureButton, it takes several images for different states, but only the normal and pressed will
be supplied in this example:
Finally, the node “copyright” is a Label. Labels can be set a custom font by editing the following property:
As a side note, the font was imported from a TTF, there is a [[Importing Fonts]] for importing fonts.
1.9 Animations
1.9.1 Introduction
This tutorial will explain how everything is animated in Godot. Godot animation system is extremely powerful and
flexible.
To begin, let’s just use the scene from the previous tutorial (splash screen). The goal will be to add a simple animation
to it. Here’s a copy just in case: attachment:robisplash.zip
First of all, add an AnimationPlayer node to the scene, make it a child of bg (the root node):
When a node of this type is selected, the animation editor panel will appear:
So, it’s time to create a new animation! Press the new animation button and name the animation “intro”.
After the animation has been created, then it’s time to edit it, by pressing the “edit” button:
1.9. Animations 41
Godot Engine Documentation, Release latest
Now this is when the magic happens! Several things happen when the “edit” button is pressed, the first one is that the
animation editor appears above the animation panel.
But the second, and most important, is that the property editor enters into “animation editing” mode. In this mode, a
key icon appears next to every property of the property editor. This means that, in Godot, any property of any object
can be animated:
Next, the logo will appear from the top of the screen. After selecting the animation player, the editor panel will stay
visible until manually hidden (or the animation node is erased). Taking advantage of this, select the “logo” node and
go to the “pos” property, move it up, to position: 114,-400.
Once in this position, press the key button next to the property:
1.9. Animations 43
Godot Engine Documentation, Release latest
As the track is new, a dialog will appear asking to create it. Confirm it!
img/addtrack.png
Change the logo position to 114,0 and a keyframe again. With two keyframes, the animation happens.
Pressing Play on the animation panel will make the logo descend. To test it by running the scene, the autoplay button
can tag the animation to start automatically when the scene starts:
And finally, when running the scene, the animation should look like this:
1.10 Resources
So far, Node have been the most important datatype in Godot, as most of the behaviors and features of the engine are
implemented through them. There is, though, another datatype that is equally as important. That is Resource.
Where Nodes focus on behaviors, such as drawing a sprite, drawing a 3D model, physics, GUI controls, etc,
1.10. Resources 45
Godot Engine Documentation, Release latest
Resources are mere data containers. This means that they don’t do any action nor process any information.
Resources just contain data.
Examples of resources are Texture, Script, Mesh, Animation, Sample, AudioStream, Font, Translation, etc.
When Godot saves or loads (from disk) a scene (.scn or .xml), an image (png, jpg), a scrit (.gd) or pretty much anything,
that file is considered a resource.
When a resource is loaded from disk, it is always loaded once. That means, if there is a copy of that resource already
loaded in memory, trying to load the resource again will just return the same copy again and again. This corresponds
with the fact that resources are just data containers, so there is no need to have them duplicated.
Typically, every object in Godot (Node, Resource, or anything else) can export properties, properties can be of many
types (like a string, integer, Vector2, etc) and one of those types can be a resource. This means that both nodes and
resources can contain resources as properties. To make it a litle more visual:
The resource properties can reference resources in two ways, external (on disk) or built-in.
To be more specific, here’s a Texture in a Sprite node:
Pressing the the “>” button the right side of the preview, allows to view and edit the resources properties. One of the
properties (path) shows where did it come from. In this case, it came from a png image.
When the resource comes from a file, it is considered an external resource. If the path property is erased (or never had
a path o begin with), it is then considered a built-in resource.
For example, if the path ‘”res://robi.png“‘ is erased from the “path” property in the above example, and then the scene
is saved, the resource will be saved inside the .scn scene file, no longer referencing the external “robi.png”. However,
even if saved as built-in, and even though the scene can be instanced multiple times, the resource will still always be
1.10. Resources 47
Godot Engine Documentation, Release latest
loaded once. That means, different Robi robot scenes instanced at the same time will still share the same image.
Loading resources from code is easy, there are two ways to do it. The first is to use load(), like this:
func _ready():
var res = load("res://robi.png") h1. resource is loaded when line is executed
get_node("sprite").set_texture(res)
The second way is more optimal, but only works with a string constant parameter, because it loads the resource at
compile-time.
func _ready():
var res = preload("res://robi.png") h1. resource is loaded at compile time
get_node("sprite").set_texture(res)
Scenes are also resources, but there is a catch. Scenes saved to disk are resources of type PackedScene, this means
that the scene is packed inside a resource.
To obtain an instance of the scene, the method instance() must be used.
func _on_shoot():
var bullet = preload("res://bullet.scn").instance()
add_child(bullet)
This method creates the nodes in hierarchy, configures them (sets all the properties) and returns the root node of the
scene, which can be added to any other node.
The approach has several advantages. As the instance function is pretty fast, adding extra content to the scene can be
done efficiently. New enemies, bullets, effects, etc can be added or removed quickly, without having to load them
again from disk each time. It is important to remember that, as always, images, meshes, etc are all shared between the
scene instances.
Resource extends from Reference. As such, when a resource is no longer in use, it will automatically free itelf. Since,
in most cases, Resources are contained in Nodes, scripts or other resources, when a node is removed or freed, all the
children resources are freed too.
1.10.6 Scripting
Like any object in Godot, not just nodes, Resources can be scripted too. However, there isn’t generally much of a win,
as resources are just data containers.
1.11 Filesystem
1.11.1 Introduction
Filesystem usage is yet another hot topic in engine development. This means, where are assets stored, how are they
accessed, how do multiple programmers edit the same repository, etc.
Initial versions of the engine (and previous iterations before it was named Godot) used a database. Assets were stored
there and assigned an ID. Other approaches were tested, too, with local databases, files with metadata, etc. To say
truth, and after a long time, simplicity proved to be best and Godot stores all assets as files in the flesystem.
1.11.2 Implementation
Godot stores resources to disk. Anything, from a script, to a scene or a PNG image is a resource to the engine. If
a resource contains properties that referece other resources on disk, the path to that resource is included. If it has
sub-resources that are built-in, the resource is saved in a single file together with all the bundled sub-resources. For
example, a font resource is often saved with the character textures bundled inside.
Metadata files were also dropped and the whole engine design tries to avoid them. The reason for this is simple,
existing asset managers and VCSs are just much better than anything we can implement, so Godot tries the best to
play along with SVN, Git, Mercurial, Perforce, etc.
1.11.3 engine.cfg
The mere existence of this file marks that there is a Godot project in that directory and all sub-directories.
This file contains the project configuration in plain text, win.ini style, though it will work to mark the existence of a
project even if the file is empty.
Example of a filesystem:
/engine.cfg
/enemy/enemy.scn
/enemy/enemy.gd
/enemy/enemysprite.png
/player/player.gd
Godot only supports “/” as a directory delimiter. This is done for portability reasons. All operating systems support
this, even Windows, so a path such as c:\\project\\engine.cfg needs to be typed as c:/project/engine.cfg.
For accessing resources, using the host OS filesystem layout can be cumbersome and non portable. To solve this
problem, the specal path ‘”res://“‘ was created.
The path ‘”res://“‘ will always point at the project root (where engine.cfg is located, so in fact ‘”res://engine.cfg“‘ is
always valid).
1.11. Filesystem 49
Godot Engine Documentation, Release latest
This filesystem is read-write only when running the project locally from the editor. When exported or when running
on different devices (such as phones or consoles, or running from DVD), the filesystem will become read-only and
writing will no longer be permitted.
Writing to disk is still needed often, from doing a savegame to downloading content packs. For this, the engine ensures
that there is a special path ‘”user://”‘ that is always writable.
Of course, opening the host filesystem always works, as this is always useful when Godot is used to write tools, but
for shipped projects this is discouraged and may not even be supported in some platforms.
1.11.8 Drawbacks
Not everything is rosy. Using resources and files and the plain filesystem has two main drawbacks. The first is that
moving assets around (renaming them or moving them from a directory to another inside the project) once they are
referenced is not that easy. If this is done, then dependencies will need to be re-satisfied upon load.
The second is that under Windows or OSX, file access is case insensitive. If a developer works in this operating system
and saves a file like “myfile.PNG”, then references it as “myfile.png”, it will work there, but not on any other platform,
such as Linux, Android, etc. It may also not work on exported binaries, which use a compressed package for files.
Because of this, please instruct your team to use a specific naming convention for files when working with Godot!
1.12 SceneTree
1.12.1 Introduction
This is where things start getting abstract, but don’t panic, as there’s not really more depth than this.
In previous tutorials, everything revolves around the concept of Nodes, scenes are made of them, and they become
active once they enter the Scene Tree.
This deserves going a little more into depth. In fact, the scene system is not even a core component of Godot, as it is
possible to skip it and make a script (or C++ code) that talks directly to the [[Servers]]. But making a game that way
would be a lot of work and is reserved for other uses.
1.12.2 MainLoop
The way Godot works internally is as follows. There is the the OS class, which is the only instance that runs at the
beginning. Afterwards, all drivers, servers, scripting languages, scene system, etc are loaded.
When initialization is complete, OS needs to be supplied a MainLoop to run. Up to this point, all this is internals
working (you can check main/main.cpp file in the source code if you are ever interested to see how this works
internally).
The user program, or game, starts in the MainLoop. This class has a few methods, for initialization, idle (frame-
syncronized callback), fixed (physics-synchronized callback), and input. Again, this is really low level and when
making games in Godot, writing your own MainLoop does not even make sense.
1.12.3 SceneTree
One of the ways to explain how Godot works, is that it’s a high level game engine over a low level middleware.
The scene system is the game engine, while the OS and servers are the low level API.
In any case, the scene system provides it’s own main loop to OS, SceneTree.
This is automatically instanced and set when running a scene, no need to do any extra work.
It’s important to know that this class exists because it has a few important uses:
• It contains the root Viewport, when a scene is first opened, it’s added as a child of it to become part of the Scene
Tree (more on that next)
• It contains information about the groups, and has means to call all nodes in a group, or get a list of them.
• It contains some global state functionality, such as setting pause mode, or quitting the process.
When a node is part of the Scene Tree, the SceneTree singleton can be obtained by simply calling Node.get_tree().
The root Viewport is always a top of the scene. From a node, it can be obtained in two different ways:
get_tree().get_root() # access via scenemainloop
get_node("/root") # access via absolute path
This node contains the main viewport, anything that is a child of a Viewport is drawn inside of it by default, so it
makes sense that the top of all nodes is always a node of this type, otherwise nothing would be seen!
While other viewports can be created in the scene (for split-screen effects and such), this one is the only one that is
never created by the user. It’s created automatically inside SceneTree.
When a node is connected, directly or indirectly, to the root viewport, it becomes part of the Scene Tree.
This means that, as explained in previous tutorials, will get the _enter_tree() and _ready() callbacks (as well as
_exit_tree()).
1.12. SceneTree 51
Godot Engine Documentation, Release latest
When nodes enter the Scene Tree, they become active. They get access to everything they need to process, get input,
display 2D and 3D, notifications, play sound, groups, etc. When they are removed from the Scene Tree, they lose it.
Most node operations in Godot, such as drawing 2D, processing or getting notifications are done in tree order. This
means that parents and siblings with less order will get notified before the current node.
4. An extra notification, “ready” ( _ready() callback in GDScript) is provided for convenience, when a node and
all it”™s children are inside the active scene.
5. When a scene (or part of it) is removed, they receive the “exit scene” notification ( _exit_tree() callback in
GDScript) in bottom-to-top order
After a scene is loaded, it is often desired to change this scene for another one. The simple way to do this to use the
SceneTree.change_scene function:
func _my_level_was_completed():
get_tree().change_scene("res://levels/level2.scn")
This is a quick and useful way to switch scenes, but has the drawback that the game will stall until the new scene is
loaded and running. At some point in your game, it may be desired to create proper loading screens with progress bar,
animated indicators or thread (background) loading. This must be done manually using autoloads (see next chapter!)
and [[Background Loading]].
1.13.1 Introduction
Scene Singletons are very useful things, as they represent a very common use case, but it’s not clear at the begining
where their value is.
The scene system is very useful, but by itself it has a few drawbacks:
• There is no “common” place to store information (such as core, items obtained, etc) between two scenes.
• It is possible to make a scene that loads other scenes as children and frees them, while keeping that information,
but then if that is done, it’s not possible to run a scene alone by itself and expect it to work
• It is also possible to store persistent information to disk in ‘user://‘ and have scenes always load it, but sav-
ing/loading that while changing scenes is cumbersome.
So, after using Godot for a while, it becomes clear that it is necessary to have parts of a scene that:
• Are always loaded, no matter which scene is opened from the editor.
• Can keep global variables, such as player information, items, money, etc.
• Can handle switching of scenes and transitions.
• Just have something that acts like a singleton, since GDScript does not support global variables by design.
For this, the option for auto-loading nodes and scripts exists.
1.13.2 Autoload
Autoload can be a scene, or a script that inherits from Node (a Node will be created and the script will be set to it).
They are added to the project in the Scene [STRIKEOUT:> Project Settings]> AutoLoad tab.
Each autoload needs a name, this name will be the node name, and the node will be always added to the root viewport
before any scene is loaded.
This means, that a for a singleton named “playervariables”, any node can access it by requesting:
var player_vars = get_node("/root/playervariables")
This short tutorial will explain how to make a scene switcher by using autoload. For simple scene switching, the
SceneTree.change_scene method suffices (described [[Scene Main Loop]] ), so this method is for more complex
behaviors when switching scenes.
First download the template from here: attachment:autoload.zip, then open it.
Two scenes are present, scene_a.scn and scene_b.scn on an otherwise empty project. Each are identical and contain a
button connected to a callback for going to the opposite scene. When the project runs, it starts n scene_a.scn. However,
this does nothing and pressing the button does not work.
1.13.4 global.gd
First of all, create a global.gd script. The easier way to create a resource from scratch is from the resources tab:
The script should be opened in the script editor. Next step will be adding it to autoload, for this, go to: Scene
[STRIKEOUT:> Project Settings]> AutoLoad and add a new autoload with name “global” that points to this file:
Now, when any scene is run, the script will be always loaded.
So, going back to it, In the _ready() function, the current scene will be fetched. Both the current scene and global.gd
are children of root, but the autoloaded nodes are always first. This means that the last child of root is always the
loaded scene.
Also, make sure that global.gd extends from Node, otherwise it won’t be loaded.
extends Node
func _ready():
var root = get_tree().get_root()
current_scene = root.get_child( root.get_child_count() -1 )
Next, is the function for changing scene. This function will erase the current scene and replace it by the requested one.
func goto_scene(path):
# The way around this is deferring the load to a later time, when
# it is ensured that no code from the current scene is running:
call_deferred("_deferred_goto_scene",path)
func _deferred_goto_scene(path):
As mentioned in the comments above, we really want to avoid the situation of having the current scene being deleted
while being used (code from functions of it being run), so using Object.call_deferred is desired at this point. The result
is that execution of the commands in the second function will happen at an immediate later time when no code from
the current scene is running.
Finally, all that is left is to fill the empty functions in scene_a.gd and scene_b.gd:
#add to scene_a.gd
func _on_goto_scene_pressed():
get_node("/root/global").goto_scene("res://scene_b.scn")
and
#add to scene_b.gd
func _on_goto_scene_pressed():
get_node("/root/global").goto_scene("res://scene_a.scn")
Finally, by running the project it’s possible to switch bewtween both scenes y pressing the button!
(To load scenes with a progress bar, check out the next tutorial, [[Background Loading]])
Engine
2.1.1 Viewports
Introduction
Godot has a small but very useful feature called viewports. Viewports are, as they name implies, rectangles where the
world is drawn. They have three main uses, but can flexibly adapted to a lot more. All this is done via the Viewport
node.
Input
Viewports are also responsible of delivering properly adjusted and scaled input events to all it’s children nodes. Both
the root viewport and sub-viewports do this automatically, but render targets do not. Because of this, the user must do
it manually via the Viewport.input function if needed.
Listener
Godot supports 3D sound (in both 2D and 3D nodes), more on this can be found in another tutorial (one day..). For this
type of sound to be audible, the viewport needs to be enabled as a listener (for 2D or 3D). If you are using a custom
viewport to display your world, don’t forget to enable this!
59
Godot Engine Documentation, Release latest
When using a 2D or 3D Camera / Camera2D, cameras will always display on the closest parent viewport (going
towards the root). For example, in the following hierarchy:
• Viewport
– Camera
Camera will display on the parent viewport, but in the following one:
• Camera
– Viewport
It will not (or may display in the root viewport if this is a subscene).
There can be only one active camera per viewport, so if there is more than one, make sure that the desired one has the
“current” property set, or make it the current camera by calling:
camera.make_current()
Viewports have a “rect” property. X and Y are often not used (only the root viewport really uses them), while WIDTH
AND HEIGHT represent the size of the viewport in pixels. For Sub-Viewports, these values are overridden by the
ones from the parent control, but for render targets this sets their resolution.
It is also possible to scale the 2D content and make it believe the viewport resolution is other than the one specified in
the rect, by calling:
viewport.set_size_override(w,h) #custom size for 2D
viewport.set_size_override_stretch(true/false) #enable stretch for custom size
The root viewport uses this for the stretch options in the project settings.
Worlds
For 3D, a Viewport will contain a World. This is basically, the universe that links physics and rendering together.
Spatial-base nodes will register using the World of the closest viewport. By default, newly created viewports do not
contain a World but use the same as a parent viewport (root viewport does contain one though, which is the one objects
are renderer by default). A world can be set in a viewport using the “world” property, and that will separate all children
nodes of that viewport from interacting with the parent viewport world. This is specially useful in scenarios where,
for example, you might want to show a separate character in 3D imposed over the game (like in Starcraft).
As a helper for situations where you want to create viewports that display single objects and don’t want to create a
world, viewport has the option to use it’s own World. This is very useful when you want to instance 3D characters or
objects in the 2D world.
For 2D, each Viewport always contains it’s own World2D. This suffices in most cases, but in case sharing them may
be desired, it is possible to do so by calling the viewport API manually.
Capture
It is possible to query a capture of the viewport contents. For the root viewport this is effectively a screen capture.
This is done with the following API:
60 Chapter 2. Engine
Godot Engine Documentation, Release latest
After a frame or two (check _process() ), the capture will be ready, get it back by using:
var capture = viewport.get_screen_capture()
If the returned image is empty, capture still didn’t happen, wait a little more, as this API is asyncronous.
Sub-Viewport
If the viewport is a child of a control, it will become active and display anything it has inside. The layout is something
like this:
• Control
– Viewport
The viewport will cover the area of it’s parent control completely.
Render Target
To set as a render target, just toggle the “render target” property of the viewport to enabled. Note that whatever is
inside will not be visible in the scene editor. To display the contents, the render target texture must be used. This can
be requested via code using (for example):
By default, re-rendering of the render target happens when the render target texture has been drawn in a frame. If
visible, it will be rendered, otherwise it will not. This behavior can be changed to manual rendering (once), or always
render, no matter if visible or not.
A few classes are created to make this easier in most common cases inside the editor:
• ViewportSprite (for 2D).
• ViewportQuad (for 3D).
• ViewportFrame (for GUI).
Make sure to check the viewport demos! Viewport folder in the demo.zip available to download, or
https://github.com/okamstudio/godot/tree/master/demos/viewport
Base Resolution
A base screen resolution for the project can be specified in the project settings.
However, what it does is not completely obvious. When running on PC, the engine will attempt to set this resolution
(or use something smaller if it fails). On mobile, consoles or devices with a fixed resolution or full screen rendering,
this resolution will be ignored and the native resolution will be used instead. To compensate for this, Godot offers
many ways to control how the screen will resize and stretch to different screen sizes.
Resizing
There are several types of devices, with several types of screens, which in turn have different pixel density and res-
olutions. Handling all of them can be a lot of work, so Godot tries to make the developer’s life a little easier. The
Viewport node has several functions to handle resizing, and the root node of the scene tree is always a viewport
(scenes loaded are instanced as a child of it, and it can always be accessed by calling get_tree().get_root()
or get_node("/root")).
In any case, while changing the root Viewport params is probably the most flexible way to deal with the problem, it
can be a lot of work, code and guessing, so Godot provides a simple set of parameters in the project settings to handle
multiple resolutions.
62 Chapter 2. Engine
Godot Engine Documentation, Release latest
Stretch Settings
Stretch settings are located in the project settings, it’s just a bunch of configuration variables that provide several
options:
Stretch Mode
• Disabled: The first is the stretch mode. By default this is disabled, which means no stretching happens (the
bigger the screen or window, the bigger the resolution, always matching pixels 1:1).
• 2D: In this mode, the resolution specified in display/width and display/height in the project settings will be
stretched to cover the whole screen. This means that 3D will be unaffected (will just render to higher-res) and
2D will also be rendered at higher-res, just enlarged.
• Viewport: Viewport scaling is different, the root Viewport is set as a render target, and still renders precisely
to the resolution specified in the display/ section of the project settings. Finally, this viewport is copied and
scaled to fit the screen. This mode is useful when working with pixel-precise games, or just for the sake of
rendering to a lower resolution for improving performance.
Stretch Aspect
• Ignore: Ignore the aspect ratio when stretching the screen. This means that the original resolution will be
stretched to fit the new one, even if it’s wider or narrower.
• Keep: Keep aspect ratio when stretching the screen. This means that the original resolution will be kept when
fitting the new one, and black bars will be added to the sides or the top/bottom of the screen.
• Keep Width: Keep aspect ratio when stretching the screen, but if the resulting screen is taller than the specified
resolution, it will be stretched vertically (and more vertical resolution will be reported in the viewport, propor-
tionally). This is usually the best option for creating GUIs or HUDs that scale, so some controls can be anchored
to the bottom ([[Gui Repositioning]]).
• Keep Height: Keep aspect ratio when stretching the screen, but if the resulting screen is wider than the specified
resolution, it will be stretched horizontally (and more horizontal resolution will be reported in the viewport, pro-
portionally). This is usually the best option for 2D games that scroll horizontally (like runners or platformers).
2.1.3 InputEvent
What is it?
Managing input is usually complex, no matter the OS or platform. To ease this a little, a special built-in type is
provided, [[API:InputEvent]]. This datatype can be configured to contain several types of input events. Input Events
travel through the engine and can be received in multiple locations, depending on the purpose.
Every input event is originated from the user/player (though it’s possible to generate an InputEvent and feed then back
to the engine, which is useful for gestures). The OS object for each platform will read events from the device, then
feed the to MainLoop. As [[API::SceneTree]] is the default MainLoop implementation, events are fed to it. Godot
provides a function to get the current SceneTree object : get_tree().
But SceneTree does not know what to do with the event, so it will give it to the viewports, starting by the “root”
[[API:Viewport]] (the first node of the scene tree). Viewport does quite a lot of stuff with the received input, in order:
64 Chapter 2. Engine
Godot Engine Documentation, Release latest
1. First, it will try to feed the input to the GUI, and see if any control can receive it. If so, the [[API:Control]]
will be called the virtual function [[API:Control._input_event()]] and the signal “input_event” will be emitted (this
function is re-implementable by script by inheriting from it). If the control wants to “consume” the event, it
will call [[API:Control.accept_event()]] and the event will not spread any more. 2. If the GUI does not want
the event, the standard _input function will be called in any node with input processing enabled (enable with
[[API:Node.set_process_input()]]) and override [[API:Node._input()]]). If any function consumes the event, it can
call [[API:SceneTree.set_input_as_handled()]], and the event will not spread any more. 3. If so far no one consumed
the event, the unhandled input callback will be called (enable with [[API:Node.set_process_unhandled_input()]])
and override [[API:Node._unhandled_input()]]). If any function consumes the event, it can call [[Scene-
Tree.set_input_as_handled()]], and the event will not spread any more. 4. If no one wanted the event so far, and
a [[API:Camera]] is assigned to the Viewport, a ray to the physics world (in the ray direction from the click) will
be casted. If this ray hits an object, it will call the [[API:CollisionObject._input_event()]] function in the relevant
physics object (bodies receive this callback by default, but areas do not. This can be configured through [[API:Area]]
properties). 5. Finally, if the event was unhandled, it will be passed to the next Viewport in the tree, or it will be
ignored.
Anatomy of an InputEvent
[[API:InputEvent]] is just a base built-in type, it does not represent anything and only contains some basic information,
such as event ID (which is increased for each event), device index, etc.
InputEvent has a “type” member. By assigning it, it can become different types of input event. Every type of In-
putEvent has different properties, according to it’s role.
Example of changing event type.
# create event
var ev = InputEvent()
# set type index
ev.type=InputEvent.MOUSE_BUTTON
# button_index is only available for the above type
ev.button_index=BUTTON_LEFT
Actions
An InputEvent may or may not represent a pre-defined action. Actions are useful because they abstract the input
device when programming the game logic. This allows for:
• The same code to work on different devices with different inputs (ie: keyboard on PC, Joypad on console)
• Input to be reconfigured at run-time.
Actions can be created from the Project Settings menu in the Actions tab. If you read the [[Tutorial 2D]], there is an
explanation on how does the action editor work.
Any event has the methods [[API:InputEvent.is_action()]], [[API:InputEvent.is_pressed()]] and
[[API:InputEvent.is_echo()]].
Alternatively, it may be desired to supply the game back with an action from the game code (a good example of this
is detecting gestures). SceneTree (derived from MainLoop) has a method for this: [[API:MainLoop.input_event(ev)]].
You would normally use it like this:
var ev = InputEvent()
ev.type=InputEvent.ACTION
# set as move_left, pressed
66 Chapter 2. Engine
Godot Engine Documentation, Release latest
ev.set_as_action("move_left",true)
# feedback
get_tree().input_event(ev)
InputMap
Customizing and re-mapping input from code is often desired. If your whole workflow depends on actions, the
[[API:InputMap]] singleton is ideal for reassigning or creating different actions at run-time. This singleton is not
saved (must be modified manually) and it’s state is run from the project settings (engine.cfg). So any dynamic system
of this type needs to store settings in the way the programmer sees best fit.
About
The reason for this small tutorial is to clear up many common mistakes about input coordinates, obtaining mouse
position and screen resolution, etc.
Using hardware coordinates makes sense in the case of writing complex UIs meant to run on PC, such as editors,
MMOs, tools, etc. Yet, make not as much sense outside of that scope.
[STRIKEOUT:The only way to reliably obtain this information is by using functions such as:]
This method is no longer supported: It was too confusing and caused errors for users making 2D games. Screen
would stretch to different resolutions and input would stop making sense. Please use the ‘‘_input‘‘ function
OS.get_video_mode_size()
Input.get_mouse_pos()
However, this is discouraged for pretty much any situation. Please do not use these functions unless you really know
what you are doing.
Godot uses viewports to display content, and viewports can be scaled by several options (see [[tutorial_multires]]
tutorial). Use, then, the functions in nodes to obtain the mouse coordinates and viewport size, for example:
func _input(ev):
# Mouse in viewport coordinates
if (ev.type==InputEvent.MOUSE_BUTTON):
print("Mouse Click/Unclick at: ",ev.pos)
elif (ev.type==InputEvent.MOUSE_MOTION):
print("Mouse Motion at: ",ev.pos)
func _ready():
set_process_input(true)
Alternatively it’s possible to ask the viewport for the mouse position
get_viewport().get_mouse_pos()
2.2 Filesystem
Introduction
This tutorial is aimed to propose a simple workflow on how to organize projects. Since Godot allows the programmer
to use the file-system as he or she pleases, figuring out a way to organize the projects when starting to use the engine
can be a little challenging. Because of this, a simple workflow will be described, which can be used or not, but should
work as a starting point.
Additionally, using version control can be challenging so this proposition will include that too.
Organization
Other game engines often work by having an asset database, were you can browse images, models, sounds, etc. Godot
is more scene-based in nature so most of the time the assets are bundled inside the scenes or just exist as files but are
referenced from scenes.
It is very often necessary to use asset importing in Godot. As the source assets for importing are also recognized as
resources by the engine, this can become a problem if both are inside the project folder, because at the time of export
the exporter will recognize them and export both.
To solve this, it is a good practice to have your game folder inside another folder (the actual project folder). This
allows to have the game assets separated from the source assets, and also allows to use version control (such as svn or
git) for both. Here is an example:
myproject/art/models/house.max
myproject/art/models/sometexture.png
myproject/sound/door_open.wav
myproject/sound/door_close.wav
myproject/translations/sheet.csv
Then also, the game itself is, in this case, inside a game/ folder:
myproject/game/engine.cfg
myproject/game/scenes/house/house.scn
myproject/game/scenes/house/sometexture.tex
myproject/game/sound/door_open.smp
myproject/game/sound/door_close.smp
myproject/game/translations/sheet.en.xl
myproject/game/translations/sheet.es.xl
68 Chapter 2. Engine
Godot Engine Documentation, Release latest
• myproject/ can be put directly inside a VCS (like svn or git) for version control, both game and source assets
are kept track of.
• If a team is working on the project, assets can be re-imported by other project members, because Godot keeps
track of source assets using relative paths.
Scene Organization
Inside the game folder, a question that often arises is how to organize the scenes in the filesystem. Many developers
try asset-type based organization and end up having a mess after a while, so the best answer is probably to organize
them based on how the game works and not based on asset type. Here are some examples.
If you were organizing your project based on asset type, it would look like this:
game/engine.cfg
game/scenes/scene1.scn
game/scenes/scene2.scn
game/textures/texturea.png
game/textures/another.tex
game/sounds/sound1.smp
game/sounds/sound2.wav
game/music/music1.ogg
Which is generally a bad idea. When a project starts growing beyond a certain point, this becomes unmanageable. It’s
really difficult to tell what belongs to what.
It’s generally a better idea to use game-context based organization, something like this:
game/engine.cfg
game/scenes/house/house.scn
game/scenes/house/texture.tex
game/scenes/valley/canyon.scn
game/scenes/valley/rock.scn
game/scenes/valley/rock.tex
game/scenes/common/tree.scn
game/scenes/common/tree.tex
game/player/player.scn
game/player/player.gd
game/npc/theking.scn
game/npc/theking.gd
game/gui/main_screen/main_sceen.scn
game/gui/options/options.scn
This model or similar models allows projects to grow to really large sizes and still be completely manageable. Notice
that everything is based on parts of the game that can be named or described, like the settings screen or the valley.
Since everything in Godot is done with scenes, and everything that can be named or described can be a scene, this
workflow is very smooth and easygoing.
Cache Files
Godot uses a hidden file called ”.fscache” at the root of the project. On it, it caches project files and is used to quickly
know when one is modified. Make sure to not commit this file to git or svn, as it contains local information and might
confuse another editor instance in another computer.
2.2. Filesystem 69
Godot Engine Documentation, Release latest
2.2.2 Paths
Path Separators
For the sake of supporting as many platforms as possible, Godot only accepts unix style path separators (/). These
work everywhere, including Windows.
A path like: “C:\\Projects” will become “c:/Projects”.
Resource Path
As mentioned before. Godot considers that a project exists at any given folder that contains an “engine.cfg” text file,
even if such file is empty.
Accessing project files can be done by opening any path with “res://” as a base. For example, a texture located in the
root of the project folder may be opened from the following path: “res://sometexture.png”.
While the project is running, it is a very common scenario that the resource path will be read-only, due to it being
inside a package, self contained executable, or system wide install location.
Storing persistent files in such scenarios should be done by using the “user://” prefix, for example:
“user://gamesave.txt”.
In some devices (for example, mobile ad consoles) this path is unique for the app. Under desktop operating systems, the
engine uses the typical ~/.Name (check the project name under the settings) in OSX and Linux, and APPDATA/Name
for Windows.
Introduction
Save games can be complicated. It can be desired to store more information than the current level or number of stars
earned on a level. More advanced save games may need to store additional information about an arbitrary number of
objects. This will allow the save function to scale as the game grows more complex.
First we should identify what objects we want to keep between game sessions and what information we want to keep
from those objects. For this tutorial, we will use groups to mark and handle objects to be saved but other methods are
certainly possible.
We will start by adding objects we wish to save to the “Persist” group. As in the [[Scripting_(Continued)]] tutorial,
we can do this through the GUI or through script. Let’s add the relevant nodes using the GUI:
Once this is done when we need to save the game we can get all objects to save them and then tell them all to save
with this script:
70 Chapter 2. Engine
Godot Engine Documentation, Release latest
Serializing
The next step is to serialize the data. This makes it much easier to read and store to disk. In this case, we’re assuming
each member of group Persist is an instanced node and thus has a file path. GDScript has helper functions for this
such as dictionary.to_json() and dictionary.parse_json() so we will use a dictionary. Our node needs to contain a save
function that returns this data. The save function will look like this:
func save():
var savedict = {
filename=get_filename(),
parent=get_parent().get_path(),
posx=get_pos().x, #Vector2 is not supported by json
posy=get_pos().y,
attack=attack,
defense=defense,
currenthealth=currenthealth,
maxhealth=maxhealth,
damage=damage,
regen=regen,
experience=experience,
TNL=TNL,
level=level,
AttackGrowth=AttackGrowth,
DefenseGrowth=DefenseGrowth,
HealthGrowth=HealthGrowth,
isalive=isalive,
last_attack=last_attack
}
return savedict
This gives us a dictionary with the style { variable_name } which will be useful when loading.
As covered in the [[File_System]] tutorial, we’ll need to open a file and write to it and then later read from it. Now
that we have a way to call our groups and get their relevant data, let’s use to_json() to convert it into an easily stored
string and store them in a file. Doing it this way ensures that each line is its own object so we have an easy way to pull
the data out of the file as well.
# Note: This can be called from anywhere inside the tree. This function is path independent.
# Go through everything in the persist category and ask them to return a dict of relevant variables
func save_game():
var savegame = File.new()
savegame.open("user://savegame.save", File.WRITE)
var savenodes = get_tree().get_nodes_in_group("Persist")
for i in savenodes:
var nodedata = i.save()
savegame.store_line(nodedata.to_json())
savegame.close()
Game saved! Loading is fairly simple as well. For that we’ll read each line, use parse_json() to read it back to a dict,
and then iterate over the dict to read our values. But we’ll need to first create the object and we can use filename and
2.2. Filesystem 71
Godot Engine Documentation, Release latest
#We need to revert the game state so we're not cloning objects during loading. This will vary wi
#For our example, we will accomplish this by deleting savable objects.
var savenodes = get_tree().get_nodes_in_group("Persist")
for i in savenodes:
i.queue_free()
# Load the file line by line and process that dictionary to restore the object it represents
var currentline = {} # dict.parse_json() requires a declared dict.
savegame.open("user://Invasionsave.save", File.READ)
while (!savegame.eof_reached()):
currentline.parse_json(savegame.get_line())
#First we need to create the object and add it to the tree and set its position.
var newobject = load(currentline["filename"]).instance()
get_node(currentline["parent"]).add_child(newobject)
newobject.set_pos(Vector2(currentline["posx"],currentline["posy"]))
# Now we set the remaining variables.
for i in currentline.keys():
if (i == "filename" or i == "parent" or i == "posx" or i == "posy"):
continue
newobject.set(i, currentline[i])
savegame.close()
And now we can save and load an arbitrary number of objects laid out almost anywhere across the scene tree! Each
object can store different data depending on what it needs to save.
Some Notes
We may have glossed over a step, but setting the game state to one fit to start loading data can be very complicated.
This step will need to be heavily customized based on the needs of an individual project.
This implementation assumes no Persist objects are children of other Persist objects. Doing so would create invalid
paths. If this is one of the needs of a project this needs to be considered. Saving objects in stages (parent objects first)
so they are available when child objects are loaded will make sure they’re available for the add_child() call. There will
also need to be some way to link children to parents as the nodepath will likely be invalid.
Why?
Because the world today is not the world of yesterday. A capitalist oligarchy runs the world and forces us to consume
in order to keep the gears of this rotten society on track. As such, the biggest market for video game consumption
today is the mobile one. It is a market of poor souls forced to compulsively consume digital content in order to forget
the misery of their every day life, commute, or just any other brief free moment they have that they are not using to
produce goods or services for the ruling class. These individuals need to keep focusing on their video games (because
not doing so will produce them a tremendous existential angst), so they go as far as spending money on them to extend
their experience, and their preferred way of doing so is through in-app purchases and virtual currency.
72 Chapter 2. Engine
Godot Engine Documentation, Release latest
But, imagine if someone was to find a way to edit the saved games and assign the items and currency without effort?
This would be terrible, because it would help players consume the content much faster, and as such run out of it sooner
than expected. If this happens they will have nothing that avoids them to think, and the tremendous agony of realizing
their own irrelevance would again take over their life.
No, we definitely do not want this to happen, so let’s see how to encrypt savegames and protect the world order.
How?
The class File is simple to use, just open a location and read/write data (integers, strings and variants). To create an
encrypted file, a passphrase must be provided, like this:
var f = File.new()
var err = f.open_encrypted_with_pass("user://savedata.bin",File.WRITE,"mypass")
f.store_var( game_state )
f.close()
This will make the file unreadable to users, but will still not avoid them to share savefiles. To solve this, using the
device unique id or some unique user identifier is needed, for example:
var f = File.new()
var err = f.open_encrypted_with_pass("user://savedata.bin",File.WRITE,OS.get_unique_ID())
f.store_var( game_state )
f.close()
2.3 Internationalization
2.3.1 Internationalization
Introduction
Sería excelente que el mundo hablara solo un idioma. Unfortunately for us developers, that is not the case. While not
generally a big requirement when developing indie or niche games, it is also very common that games going into a
more massive market require localization.
Godot offers many tools to make this process more straightforward, so this tutorial is more like a collection of tips and
tricks.
Localization is usually done by specific studios hired for the job and, despite the huge amount of software and file
formats available for this, the most common way to do localization to this day is still with spreadsheets. The process
of creating the spreadsheets and importing them is already covered in the [[Import Translation]] tutorial, so this one
could be seen more like a follow up to that one.
The translations can get updated and re-imported when they change, but they still have to be added to the project. This
is done in Scene [STRIKEOUT:> Project Settings]> Localization:
2.3. Internationalization 73
Godot Engine Documentation, Release latest
Localizing Resources
It is also possible to instruct Godot to open alternative versions of assets (resources) depending on the current language.
For this the “Remaps” tab exists:
Select the resource to be remapped, and the alternatives for each locale.
Some controls such as Button. will automatically fetch a translation each time they are set a key instead of a text.
For example, if a label is assigned “MAIN_SCREEN_GREETING1” and a key to different languages exists in the
translations, this will be automatically converted. This process is done upon load though, so if the project in question
74 Chapter 2. Engine
Godot Engine Documentation, Release latest
has a dialog that allows changing the language in the settings, the scenes (or at least the settings scene) will have to be
re-loaded for new text to have effect.
For code, the Object.tr() function can be used. This will just look-up the text into the translations and convert it if
found:
level.set_text(tr("LEVEL_5_NAME"))
status.set_text(tr("GAME_STATUS_"+str(status_index)))
The same text in different languages can vary greatly in length. For this, make sure to read the tutorial on [[GUI
Repositioning]], as having dynamically adjusted control sizes may help. Containers can be very useful, as well as the
multiple options in Label for text wrapping.
TranslationServer
Godot has a server for handling the low level translation management called the TranslationServer. Translations can
be added or removed during run-time, and the current language be changed too.
Command Line
Language can be tested when running Godot from command line. For example, to test a game in french, the following
arguments can be supplied:
The project name becomes the app name when exporting to different operating systems and platforms. To specify
the project name in more than one language. In the project settings dialog, create a new setting application/name and
append it the locale identifier. For example:
2.3. Internationalization 75
Godot Engine Documentation, Release latest
As always, If you don’t know the code of a language or zone, check the list.
Pause?
In most games, it is always desirable to, at some point, interrupt the game to do something else. Be it taking a break,
to changing options. However this is not as simple as it seems. The game might be stopped, but it might be desirable
that some menus and animations continue working.
Implementing a fine-grained control for what can be paused (and what not) is a lot of work, so a simple framework for
pausing is provided in Godot.
To set pause mode, the pause state must be set. This is done by calling SceneTree.set_pause with a “true” argument:
76 Chapter 2. Engine
Godot Engine Documentation, Release latest
get_tree().set_pause(true)
White-Listing Nodes
Before enabling pause, make sure that nodes that must keep working during pause are white-listed. This is done by
editing the “Pause Mode” property in a node:
By default all nodes have this property in the “Inherit” state. This means, that they will only process (or not) depending
on what this same property is set on the parent node. If the parent is set to “Inherit” , then the grandparent will be
checked and so on. Ultimately, if a state can’t be found in any of the grandparents, the pause state in SceneTree is
used. This means that, by default, when the game is paused every node will be paused.
So the three possible states for a node are:
• Inherit: Process depending on the state of the parent, grandparent, etc. The first parent that has a non-Inherit
state.
• Stop: Stop the node no matter what (and children in Inherit mode). Paused or not this node will not process.
• Process: Process the node no matter what (and children in Inherit mode). Paused or not this node will process.
Example
An example of this is creating a popup or panel with controls inside, and set it’s pause mode to “Process” then just
hide it:
78 Chapter 2. Engine
Godot Engine Documentation, Release latest
Just by setting the root of the pause popup to “Process”, all children and grandchildren will inherit that state. This
way, this branch of the scene tree will continue working when paused.
Finally, make it so when a pause button is pressed (any button will do), enable the pause and show the pause screen.
func _on_pause_button_pressed():
get_tree().set_pause(true)
get_node("pause_popup").show()
To remove the pause, just do the opposite when the pause screen is closed:
func _on_pause_popup_close_pressed():
get_node("pause_popup").hide()
get_tree().set_pause(false)
When switching the main scene of your game (for example going to a new level), you might want to show a loading
screen with some indication that progress is being made. The main load method (ResourceLoader::load or just
load from gdscript) blocks your thread while the resource is being loaded, so It’s not good. This document discusses
the ResourceInteractiveLoader class for smoother load screens.
ResourceInteractiveLoader
The ResourceInteractiveLoader class allows you to load a resource in stages. Every time the method poll
is called, a new stage is loaded, and control is returned to the caller. Each stage is generally a sub-resource that is
loaded by the main resource. For example, if you’re loading a scene that loads 10 images, each image will be one
stage.
Usage
Obtaining a ResourceInteractiveLoader
This method will give you a ResourceInteractiveLoader that you will use to manage the load operation.
Polling
Error ResourceInteractiveLoader::poll();
Use this method to advance the progress of the load. Each call to poll will load the next stage of your resource. Keep
in mind that each stage is one entire “atomic” resource, such as an image, or a mesh, so it will take several frames to
load.
Returns OK on no errors, ERR_FILE_EOF when loading is finished. Any other return value means there was an error
and loading has stopped.
80 Chapter 2. Engine
Godot Engine Documentation, Release latest
get_stage_count
get_stage
returns the current stage being loaded
Error ResourceInteractiveLoader::wait();
Use this method if you need to load the entire resource in the current frame, without any more steps.
Ref ResourceInteractiveLoader::get_resource();
If everything goes well, use this method to retrieve your loaded resource.
Example
This example demostrates how to load a new scene. Consider it in the context of the
[[https://github.com/okamstudio/godot/wiki/tutorial_singletons#scene-switcher]] example.
First we setup some variables and initialize the
current_scene
with the main scene of the game:
var loader
var wait_frames
var time_max = 100 h1. msec
var current_scene
func _ready():
var root = get_tree().get_root()
current_scene = root.get_child( root.get_child_count() -1 )
The function
goto_scene
is called from the game when the scene needs to be switched. It requests an interactive loader, and calls
set_progress(true)
wait_frames = 1
_process is where the loader is polled. poll is called, and then we deal with the return value from that call. OK
means keep polling, ERR_FILE_EOF means load is done, anything else means there was an error. Also note we skip
one frame (via wait_frames, set on the goto_scene function) to allow the loading screen to show up.
Note how use use OS.get_ticks_msec to control how long we block the thread. Some stages might load really
fast, which means we might be able to cram more than one call to poll in one frame, some might take way more than
your value for time_max, so keep in mind we won’t have precise control over the timings.
func _process(time):
if loader == null:
# no need to process anymore
set_process(false)
return
if wait_frames > 0: # wait for frames to let the "loading" animation to show up
wait_frames -= 1
return
var t = OS.get_ticks_msec()
while OS.get_ticks_msec() < t + time_max: # use "time_max" to control how much time we block this
Some extra helper functions. update_progress updates a progress bar, or can also update a paused animation
(the animation represents the entire load process from beginning to end). set_new_scene puts the newly loaded
scene on the tree. Because it’s a scene being loaded, instance() needs to be called on the resource obtained from
the loader.
82 Chapter 2. Engine
Godot Engine Documentation, Release latest
func update_progress():
var progress = float(loader.get_stage()) / loader.get_stage_count()
# update your progress bar?
get_node("progress").set_progress(progress)
# call this on a paused animation. use "true" as the second parameter to force the animation to u
get_node("animation").seek(progress * len, true)
func set_new_scene(scene_resource):
current_scene = scene_resource.instance()
get_node("/root").add_child(current_scene)
ResourceInteractiveLoader can be used from multiple threads. A couple of things to keep in mind if you attempt it:
Use a Semaphore
While your thread waits for the main thread to request a new resource, use a Semaphore to sleep (instead of a busy
loop or anything similar).
If you have a mutex to allow calls from the main thread to your loader class, don’t lock it while you call poll on the
loader. When a resource is finished loading, it might require some resources from the low level APIs (VisualServer,
etc), which might need to lock the main thread to acquire them. This might cause a deadlock if the main thread is
waiting for your mutex while your thread is waiting to load a resource.
Example class
You can find an example class for loading resources in threads here. Usage is as follows:
func start()
Queue a resource. Use optional parameter “p_in_front” to put it in front of the queue.
func cancel_resource(path)
Get the progress of a resource. Returns -1 on error (for example if the resource is not on the queue), or a number
between 0.0 and 1.0 with the progress of the load. Use mostly for cosmetic purposes (updating progress bars, etc), use
is_ready to find out if a resource is actually ready.
func get_resource(path)
Returns the fully loaded resource, or null on error. If the resource is not done loading (is_ready returns false), it
will block your thread and finish the load. If the resource is not on the queue, it will call ResourceLoader::load
to load it normally and return it.
Example:
# initialize
queue = preload("res://resource_queue.gd").new()
queue.start()
# suppose your game starts with a 10 second custscene, during which the user can't interact with the
# For that time we know they won't use the pause menu, so we can queue it to load during the cutscene
queue.queue_resource("res://pause_menu.xml")
start_curscene()
# later when the user presses the pause button for the first time:
pause_menu = queue.get_resource("res://pause_menu.xml").instance()
pause_menu.show()
# to check progress
if queue.is_ready("res://level_1.xml"):
show_new_level(queue.get_resource("res://level_1.xml"))
else:
update_progress(queue.get_process("res://level_1.xml"))
# when the user walks away from the trigger zone in your Metroidvania game:
queue.cancel_resource("res://zone_2.xml")
Note: this code in its current form is not tested in real world scenarios. Find me on IRC (punto on irc.freenode.net) or
e-mail me ([email protected]) for help.
Quitting
Most platforms have the option to request the application to quit. On desktops, this is usually done with the “x” icon on
the window titlebar. On Android, the back button is used to quit when on the main screen (and to go back otherwise).
The MainLoop has a special notification that is sent to all nodes when quit is requested: Main-
Loop.NOTIFICATION_WM_QUIT.
Handling it is done as follows (on any node):
84 Chapter 2. Engine
Godot Engine Documentation, Release latest
func _notification(what):
if (what==MainLoop.NOTIFICATION_WM_QUIT_REQUEST):
get_tree().quit() #default behavior
When developing mobile apps, quitting is not desired unless the user is on the main screen, so the behavior can be
changed.
It is important to note that by default, Godot apps have the built-in behavior to quit when quit is requested, this can be
changed:
get_tree().set_auto_accept_quit(false)
86 Chapter 2. Engine
CHAPTER 3
2D tutorials
3.1 Graphics
Regular 2D nodes, such as Node2D or Control both inherit from CanvasItem, which is the base for all 2D nodes.
CanvasItems can be arranged in trees and they will inherit their transform. This means that, moving the parent, the
children will be moved too.
These nodes are placed as direct or indirect children to a Viewport, and will be displayed through it.
Viewport has a property “canvas_transform” (Viewport.set_canvas_transform(), which allows to transform all the
CanvasItem hierarchy by a custom Matrix32 transform. Nodes such as Camera2D, work by changing that transform.
Changing the canvas transform is useful because it is a lot more efficient than moving the root canvas item (and hence
the whole scene). Canvas transform is a simple matrix that offsets the whole 2D drawing, so it’s the most efficient way
to do scrolling.
Not Enough..
But this is not enough. There are often situations where the game or application may not want everything transformed
by the canvas transform. Examples of this are:
• Parallax Backgrounds: Backgrounds that move slower than the rest of the stage.
• HUD: Head’s up display, or user interface. If the world moves, the life counter, points, etc must stay static.
• Transitions: Effects used for transitions (fades, blends) may also want it to remain at a fixed location.
How can these problems be solved in a single scene tree?
CanvasLayers
The answer is CanvasLayer, which is a node that adds a separate 2D rendering layer for all it’s children and grand-
children. Viewport children will draw by default at layer “0”, while a CanvasLayer will draw at any numeric layer.
Layers with a greater number will be drawn above those with a smaller number. CanvasLayers also have their own
87
Godot Engine Documentation, Release latest
transform, and do not depend of the transform of other layers. This allows the UI to be fixed in-place, while the word
moves.
An example of this is creating a parallax background. This can be done with a CanvasLayer at layer “-1”. The screen
with the points, life counter and pause button can also be created at layer “1”.
Here’s a diagram of how it looks:
CanvasLayers are independent of tree order, and they only depend on their layer number, so they can be instantiated
when needed.
Performance
Even though there shouldn’t be any performance limitation, it is not advised to use excessive amount of layers to
arrange drawing order of nodes. The most optimal way will always be arranging them by tree order. In the future,
nodes will also have a priority or sub-layer index which should aid for this.
Introduction
This tutorial is created after a topic that is a little dark for most users, and explains all the 2D transforms going on for
nodes from the moment they draw their content locally to the time they are drawn into the screen.
88 Chapter 3. 2D tutorials
Godot Engine Documentation, Release latest
Canvas Transform
As mentioned in the previous tutorial [[Canvas Layers]], every CanvasItem node (remember that Node2D and Control
based nodes use CanvasItem as their common root) will reside in a Canvas Layer. Every canvas layer has a transform
(translation, rotation, scale, etc) that can be accessed as a Matrix32.
By default, nodes are drawn in Layer 0, in the built-in canvas. To put nodes in a different layer, a CanvasLayer node
can be used. This was covered in the previous tutorial anyway, just refreshing.
Viewports also have a Global Canvas transform (also a Matrix32 ). This is the master transform and affects all
individual Canvas Layer transforms. Generally this transform is not of much use, but is used in the CanvasItem Editor
in Godot’s editor.
Stretch Transform
Finally, viewports have a Stretch Transform, which is used when resizing or stretching the screen. This transform is
used internally by the [[Tutorial Multires]], but can also be requested to the viewport.
Input events received in the Node._input_event(ev) callback are multiplied by this transform, but lack the ones above.
To convert InputEvent coordinates to local CanvasItem coordinates, the CanvasItem.make_input_local(ev) function
was added for convenience.
Transform Order
For a coordinate in CanvasItem local properties to become an actual screen coordinate, the following chain of trans-
forms must be applied:
3.1. Graphics 89
Godot Engine Documentation, Release latest
Transform Functions
|Type: | Transform|
|CanvasItem | CanvasItem.get_global_transform() |
|CanvasLayer| CanvasItem.get_canvas_transform() |
|CanvasLayer+GlobalCanvas+Stretch | CanvasItem.get_viewport_transform() |
90 Chapter 3. 2D tutorials
Godot Engine Documentation, Release latest
Finally then, to convert a CanvasItem local coordinates to screen coordinates, just multiply in the following order:
var screen_coord = get_viewport_transform() + ( get_global_transform() + local_pos )
Keep in mind, however, that it is generally not desired to work with screen coordinates. The recommended approach
is to simply work in Canvas coordinates (CanvasItem.get_global_transform()), to allow automatic screen resolution
resizing to work properly.
It is often desired to feed custom input events to the scene tree. With the above knowledge, to correctly do this, it must
be done the following way:
var local_pos = Vector2(10,20) # local to Control/Node2D
var ie = InputEvent()
ie.type=InputEvent.MOUSE_BUTTON
ie.button_index=1 #left click
ie.pos = get_viewport_transform() + ( get_global_transform() + local_pos )
get_tree().input_event(ie)
Why?
Godot has nodes to draw sprites, polygons, particles, and all sort of stuff. For far most cases this is enough, but not
always. If something desired is not supported, and before crying in fear, angst and range because a node to draw
that-specific-something does not exist.. it would be good to know that it is possible to easily make any 2D node (be it
Control or Node2D based) draw custom commands. It is really easy to do it too.
But..
Custom drawing manually in a node is really useful. Here are some examples why:
• Drawing shapes or logic that is not handled by nodes (example: making a node that draws a circle, an image
with trails, a special kind of animated polygon, etc).
• Visualizations that are not that compatible with nodes: (example: a tetris board). The tetris example uses a
custom draw function to draw the blocks.
• Managing drawing logic of a large amount of simple objects (in the hundreds of thousands). Using a thousand
nodes is probably not nearly as efficient as drawing, but a thousand of draw calls are cheap. Check the “Shower
of Bullets” demo as example.
• Making a custom UI control. There are plenty of controls available, but it’s easy to run into the need to make a
new, custom one.
OK, How?
Add a script to any CanvasItem derived node, like Control or Node2D. Override the _draw() function.
extends Node2D
func _draw():
#your draw commands here
pass
3.1. Graphics 91
Godot Engine Documentation, Release latest
Draw commands are described in the CanvasItem class reference. There are plenty of them.
Updating
The _draw() function is only called once, and then the draw commands are cached and remembered, so further calls
are unnecessary.
If re-drawing is required because a state or something else changed, simply call CanvasItem.update() in that same node
and a new _draw() call will happen.
Here is a little more complex example. A texture variable that will be redrawn if modified:
extends Node2D
func _set_texture(value):
#if the texture variable is modified externally,
#this callback is called.
texture=value #texture was changed
update() #update the node
func _draw():
draw_texture(texture,Vector2())
In some cases, it may be desired to draw every frame. For this, just call update() from the _process() callback, like
this:
extends Node2D
func _draw():
#your draw commands here
pass
func _process(delta):
update()
func _ready():
set_process(true)
Tools
Drawing your own nodes might also be desired while running them in the editor, to use as preview or visualization of
some feature or behavior.
Remember to just use the “tool” keyword at the top of the script (check the [[GDScript]] reference if you forgot what
this does).
Introduction
Very often it is desired to make a shader that reads from the same screen it’s writing to. 3D APIs such as OpenGL or
DirectX make this very difficult because of internal hardware limitations. GPUs are extremely parallel, so reading and
92 Chapter 3. 2D tutorials
Godot Engine Documentation, Release latest
writing causes all sort of cache and coherency problems. As a result, not even the most modern hardware supports this
properly.
The workaround is to make a copy of the screen, or a part of the screen, to a back-buffer and then read from it while
drawing. Godot provides a few tools that makes this process easy!
Godot [[Shader]] has a special instruction, “texscreen”, it takes as parameter the UV of the screen and returns a vec3
RGB with the color. A special built-in varying: SCREEN_UV can be used to obtain the UV for the current fragment.
As a result, this simple 2D fragment shader:
COLOR=vec4( texscreen(SCREEN_UV), 1.0 );
results in an invisible object, because it just shows what lies behind. The same shader using the visual editor looks like
this:
TexScreen Example
Texscreen instruction can be used for a lot of things. There is a special demo for Screen Space Shaders, that you can
download to see and learn. One example is a simple shader to adjust brightness, contrast and saturation:
uniform float brightness=1.0;
uniform float contrast=1.0;
uniform float saturation=1.0;
vec3 c = texscreen(SCREEN_UV);
c.rgb = mix(vec3(0.0),c.rgb,brightness);
c.rgb = mix(vec3(0.5),c.rgb,contrast);
c.rgb = mix(vec3(dot(vec3(1.0),c.rgb)*0.33333),c.rgb,saturation);
COLOR.rgb=c;
While this seems magical, it’s not. The Texscreen instruction, when first found in a node that is about to be drawn,
does a full-screen copy to a back-buffer. Subsequent nodes that use texscreen() in shaders will not have the screen
copied for them, because this ends up being very inefficient.
As a result, if shaders that use texscreen() overlap, the second one will not use the result of the first one, resulting in
unexpected visuals:
3.1. Graphics 93
Godot Engine Documentation, Release latest
In the above image, the second sphere (top right) is using the same source for texscreen() as the first one below, so the
first one “dissapears”, or is not visible.
To correct this, a BackBufferCopy node can be instanced between both spheres. BackBufferCopy can work by either
specifying a screen region or the whole screen:
Back-Buffer Logic
So, to make it clearer, here’s how the backbuffer copying logic works in Godot:
• If a node uses the texscreen(), the entire screen is copied to the back buffer before drawing that node. This only
happens the first time, subsequent nodes do not trigger this.
• If a BackBufferCopy node was processed before the situation in the point above (even if texscreen() was not
used), this behavior described in the point above does not happen. In other words, automatic copying of the
entire screen only happens if texscreen() is used in a node for the first time and no BackBufferCopy node (not
disabled) was found before in tree-order.
94 Chapter 3. 2D tutorials
Godot Engine Documentation, Release latest
• BackBufferCopy can copy either the entire screen or a region. If set to only a region (not the whole screen) and
your shader uses pixels not in the region copied, the result of that read is [STRIKEOUT:undefined] (most likely
garbage from previous frames). In other words, it’s possible to use BackBufferCopy to copy back a region of
the screen and then use texscreen() on a different region. Avoid this behavior!
Intro
A simple (but flexible enough for most uses) particle system is provided. Particle systems are used to simulate complex
physical effects tsuch as sparks, fire, magic particles, smoke, mist, magic, etc.
The idea is that a “particle” is emitted at a fixed interval and with a fixed lifetime. During his lifetime, every particle
will have the same base behavior. What makes every particle different and provides a more organic look is the “ran-
domness” associated to each parameter. In essence, creating a particle system means setting base physics parameters
and then adding randomness to them.
Particles2D
Particle systems are added to the scene via the Particles2D node. They are enabled by default and start emitting white
points downwards (as affected by the gravity). This provides a reasonable starting point to start adapting it to our
needs.
Texture
A particle system uses a single texture (in the future this might be extended to animated textures via spritesheet). The
texture is set via the relevant texture property:
3.1. Graphics 95
Godot Engine Documentation, Release latest
Physics Variables
Before taking a look at the global parameters for the particle system, let’s first see what happens when the physics
variables are tweaked.
Direction
Spread
Spread is the angle at which particles will randomly be emitted. Increasing the spread will increase the angle. A spread
of 180 will emit in all directions.
Linear Velocity
Linear Velocity is the speed at which particles will be emitted (in pixels/sec). Speed might later be modified by gravity
or other accelerations (as described further below).
Spin Velocity
Spin Velocity is the speed at which particles turn around their center (in degrees/sec).
Orbit Velocity
96 Chapter 3. 2D tutorials
Godot Engine Documentation, Release latest
Gravity can be modified as in direction and strength. Gravity affects every particle currently alive.
Radial Acceleration
If this acceleration is positive, particles are accelerated away from the center. If negative, they are absorbed towards it.
Tangential Acceleration
This acceleration will use the tangent vector to the center. Combined with Radial Acceleration can do nice effects.
Damping
Damping applies friction to the particles, forcing them to stop. It is specially useful for sparks or explosions, which
usually begin with a high linear velocity and then stop as they fade.
Initial Angle
Determines the intial angle of the particle (in degress). This parameter is mostly useful randomized.
Color Phases
Particles can use up to 4 color phases. Each color phase can include transparency.
Phases must provide an offset value from 0 to 1, and alays in ascending order. For example, a color will begin at
offset 0 and end in offset 1, but 4 colors might use diferent offsets, such as 0, 0.2, 0.8 and 1.0 for the different phases:
3.1. Graphics 97
Godot Engine Documentation, Release latest
Global Parameters
Lifetime
The time in seconds that every particle will stay alive. When lifetime ends, a new particle is created to replace it.
Lifetime: 0.5
Lifetime: 4.0
Timescale
It happens often that the effect achieved is perfect, except too fast or too slow. Timescale helps adjust the overall speed.
Timescale everything 2x:
Preprocess
Particle systems begin with 0 particles emitted, then start emitting. This can be an inconvenience when just loading
a scene and systems like a torch, mist, etc begin emitting the moment you enter. Preprocess is used to let the system
process a given amount of seconds before it is actually shown the first time.
98 Chapter 3. 2D tutorials
Godot Engine Documentation, Release latest
Emit Timeout
This variable will switch emission off after given amount of seconds being on. When zero, itś disabled.
Offset
Half Extents
Makes the center (by default 1 pixel) wider, to the size in pixels desired. Particles will emit randomly inside this area.
It is also possible to set an emission mask by using this value. Check the “Particles” menu on the 2D scene editor
viewport and select your favorite texture. Opaque pixels will be used as potential emission location, while transparent
ones will be ignored:
Local Space
By default this option is on, and it means that the space that particles are emitted to is contained within the node. If
the node is moved, all particles are moved with it:
If disabled, particles will emit to global space, meaning that if the node is moved, the emissor is moved too:
Explosiveness
If lifetime is 1 and there are 10 particles, it means every particle will be emitted every 0.1 seconds. The explosiveness
parameter changes this, and forces particles to be emitted all together. Ranges are:
• 0: Emit all particles together.
• 1: Emit particles at equal interval.
Values in the middle are also allowed. This feature is useful for creating explosions or sudden bursts of particles:
Randomness
All physics parameters can be randomiez. Random variables go from 0 to 1. the formula to randomize a parameter is:
initial_value = param_value + param_value*randomness
What is it?
Cut-out is a technique of animating in 2D where pieces of paper (or similar material) are cut in special shapes and laid
one over the other. The papers are animated and photographed, frame by frame using a stop motion technique (more
info here.
With the advent of the digital age, this technique became possible using computers, which resulted in an increased
amount of animation TV shows using digital Cut-out. Notable examples are South Park or Jake and the Never Land
Pirates .
In video games, this technique also become very popular. Examples of this are Paper Mario or Rayman Origins .
3.1. Graphics 99
Godot Engine Documentation, Release latest
Cutout in Godot
Godot provides a few tools for working with these kind of assets, but it’s overall design makes it ideal for the workflow.
The reason is that, unlike other tools meant for this, Godot has the following advantages:
• The animation system is fully integrated with the engine: This means, animations can control much more
than just motion of objects, such as textures, sprite sizes, pivots, opacity, color modulation, etc. Everything can
be animated and blended.
• Mix with Traditional: AnimatedSprite allows traditional animation to be mixed, very useful for complex ob-
jects, such as shape of hands and foot, changing face expression, etc.
• Custom Shaped Elements: Can be created with Polygon2D allowing the mixing of UV animation, deforma-
tions, etc.
• Particle Systems: Can also be mixed with the traditional animation hierarchy, useful for magic effecs, jetpacks,
etc.
• Custom Colliders: Set colliders and influence areas in different parts of the skeletons, great for bosses, fighting
games, etc.
• Animation Tree: Allows complex combinations and blendings of several animations, the same way it works in
3D.
And much more!
Making of GBot!
For this tutorial, we will use as demo content the pieces of the GBot character, created by Andreas Esau.
Get your assets attachment:gbot_resources.zip .
Create an empty Node2D as root of the scene, weĺl work under it:
OK, the first node of the model that we will create will be the hip. Generally, both in 2D and 3D, the hip is the root of
the skeleton. This makes it easier to animate:
Next will be the torso. The torso needs to be a child of the hip, so create a child sprite and load the torso, later
accommodate it properly:
This looks good. Let’s try if our hierarchy works as a skeleton by rotating the torso:
Ouch, that doesn’t look good! The rotation pivot is wrong, this means it needs to be adjusted.
This small little cross in the middle of the Sprite is the rotation pivot:
The Pivot can be adjusted by changing the offset property in the Sprite:
However, there is a way to do it more visually. Pick the object and move it normally. After the motion has begun and
while the left mouse button is being held, press the “v” key without releasing the mouse button. Further motion will
move the object around the pivot. This small tool allows adjusting the pivot easily. Finally, move the pivot to the right
place:
Now it looks good! Let’s continue adding body pieces, starting by the right arm. Make sure to put the sprites in
hierarchy, so their rotations and translations are relative to the parent:
This seems easy, so continue with the right arm. The rest should be simple! Or maybe not:
Right. Remember your tutorials, Luke. In 2D, parent nodes appear below children nodes. Well, this sucks. It seems
Godot does not support cutout rigs after all. Come back next year, maybe for 1.2.. no wait. Just Kidding! It works just
fine.
But how can this problem be solved? We want the whole to appear behind the hip and the torso. For this, we can move
the nodes behind the hip:
But then, we lose the hierarchy layout, which allows to control the skeleton like.. a skeleton. Is there any hope?.. Of
Course!
RemoteTransform2D Node
Godot provides a special node, RemoteTransform2D. This node will transform nodes that are sitting somewhere else
in the hierarchy, by copying it’s transform to the remote node.
This enables to have a visibility order independent from the hierarchy.
Simply create two more nodes as children from torso, remote_arm_l and remote_hand_l and link them to the actual
sprites:
Moving the remote transform nodes will move the sprites, allowing to easily animate and pose the character:
Complete the skeleton by following the same steps for the rest of the parts. The resulting scene should look similar to
this:
The resulting rig should be easy to animate, by selecting the nodes and rotating them you can animate forward kine-
matic (FK) efficiently.
For simple objects and rigs this is fine, however the following problems are common:
• Selecting sprites can become difficult for complex rigs, and the scene tree ends being used due to the difficulty
of clicking over the proper sprite.
• Inverse Kinematics is often desired for extremities.
To solve these problems, Godot supports a simple method of skeletons.
Skeletons
Godot does not really support actual skeletons. What exists is a helper to create “bones” between nodes. This is
enough for most cases, but the way it works is not completely obvious.
As an example, let’s turn the right arm into a skeleton. To create skeletons, a chain of nodes must be selected from top
to bottom:
Then, the option to create a skeleton is located at Edit [STRIKEOUT:> Skeleton]> Make Bones:
This will add bones covering the arm, but the result is not quite what is expected.
It looks like the bones are shifted up in the hierarchy. The hand connects to the arm, and the arm to the body. So the
question is:
• Why does the hand lack a bone?
• Why does the arm connect to the body?
This might seem strange at first, but will make sense later on. In traditional skeleton systems, bones have a position, an
orientation and a length. In Godot, bones are mostly helpers so they connect the current node with the parent. Because
of this, toggling a node as a bone will just connect it to the parent.
So, with this knowledge. Let’s do the same again so we have an actual, useful skeleton.
The first step is creating an endpoint node. Any kind of node will do, but Position2D is preferred because it’s visible
in the editor. The endpoint node will ensure that the last bone has orientation
Now select the whole chain, from the endpoint to the arm and create bones:
The result resembles a skeleton a lot more, and now the arm and forearm can be selected and animated.
Finally, create endpoints in all meaningful extremities and connect the whole skeleton with bones up to the hip:
Finally! the whole skeleton is rigged! On close look, it is noticeable that there is a second set of endpoints in the
hands. This will make sense soon.
Now that a whole skeleton is rigged, the next step is setting up the IK chains. IK chains allow for more natural control
of extremities.
IK Chains
To add in animation, IK chains are a powerful tool. Imagine you want to pose a foot in a specific position in the
ground. Moving the foot involves also moving the rest of the leg bones. Each motion of the foot involves rotating
several other bones. This is quite complex and leads to imprecise results.
So, what if we could just move the foot and let the rest of the leg accommodate to the new foot position?
This type of posing is called IK (Inverse Kinematic).
To create an IK chain, simply select a chain of bones from endpoint to the base for the chain. For example, to create
an IK chain for the right leg select the following:
Then enable this chain for IK. Go to Edit [STRIKEOUT:> Skeleton]> Make IK Chain
Once the IK chain is set-up, simply grab any of the bones in the extremity, any child or grand-child of the base of the
chain and try to grab it and move it. Result will be pleasant, satisfaction warranted!
Animation
The following section will be a collection of tips for creating animation for your rigs. If unsure about how the animation
system in Godot works, refresh it by checking again the [[tutorial_animation]].
2D Animation
When doing animation in 2D, a helper will be present in the top menu. This helper only appears when the animation
editor window is opened:
The key button will insert location/rotation/scale keyframes to the selected objects or bones. This depends on the mask
enabled. Green items will insert keys while red ones will not, so modify the key insertion mask to your preference.
Rest Pose
These kind of rigs do not have a “rest” pose, so it’s recommended to create a reference rest pose in one of the
animations.
Simply do the following steps:
1. Make sure the rig is in “rest” (not doing any specific pose).
2. Create a new animation, rename it to “rest”.
3. Select all nodes (box selection should work fine).
4. Select “loc” and “rot” on the top menu.
5. Push the key button. Keys will be inserted for everything, creating a default pose.
Rotation
Animating these models means only modifying the rotation of the nodes. Location and scale are rarely used, with the
only exception of moving the entire rig from the hip (which is the root node).
As a result, when inserting keys, only the “rot” button needs to be pressed most of the time:
This will avoid the creation of extra animation tracks for the position that will remain unused.
Keyframing IK
When editing IK chains, is is not neccesary to select the whole chain to add keyframes. Selecting the endpoint of the
chain and inserting a keyframe will automatically insert keyframes until the chain base too. This makes the task of
animating extremities much simpler.
RemoteTransform2D works in most cases, but sometimes it is really necessary to have a node above and below others
during an animation. To aid on this the “Behind Parent” property exists on any Node2D:
When creating really complex animations and inserting lots of keyframes, editing the individual keyframe curves for
each can become an endless task. For this, the Animation Editor has a small menu where changing all the curves is
easy. Just select every single keyframe and (generally) apply the “Out-In” transition curve to smooth the animation:
3.1.8 Introduction
Tilemaps are a simple and quick way to make 2D game levels. Basically, you start with bunch of reference tiles (or
pieces) that can be put in a grid, as many times each as desired:
Collision can also be added to the tiles, allowing for both 2D side scroller or top down games.
Making a Tileset
To begin with, a tileset needs to be made. Here are some tiles for it. They are all in the same image because artists will
often prefer this. Having them as separate images also works too.
Create a new project and throw the above png image inside.
We will be creating a TileSet resource. While this resource exports properties, it’s pretty difficult to get complex data
into it and maintain it:
There’s enough properties to get by, and with some effort editing this way can work, but the easiest way to edit and
maintain a tileset is with the export tool!
TileSet Scene
Create a new scene with a regular node or node2d as root. For each new a sprite will be added. Since tiles here are
50x50, enabling snap might be a good idea.
If more than one tile is present in the source image, make sure to use the region property of the sprite to adjust which
of the sprites is the source texture is being edited.
Finally, make sure to name your sprite node correctly, this will ensure that, in subsequent edits to the tileset (for
example, if added collision, changed the region, etc), the tile will still be identified correctly and updated. This
name should be unique.
Sounds like a lot of requirements, so here’s a screenshot that shows where everything of relevance is:
Continue adding all the tiles, adjust the offsets if needed (if you use multiple tiles in a single image) unless there is a
sprite per each tile. Again, as always, remember that their names must be unique.
Collision
To add collision to a tile, create a StaticBody2D child for each sprite. This is a static collision node. Then, as a child
of the StaticBody2D, create a CollisionShape2D or CollisionPolygon. The later is recommended because it’s easier to
edit:
Finally, edit the polygon, this will give the tile a collision. Remember to use snap!. Using snap will make sure
collision polygons are aligned properly, allowing a character to walk seamlessly from tile to tile. Also do not scale or
move the collision and/or collision polygon nodes. leave them at offset 0,0, with scale 1,1 and rotation 0 respect to the
parent sprite.
Keep adding collisions to tiles untile we are done. Note that BG is just a BG so we don’t care about it.
OK! We’re done! Remember to save this scene for future edit, call it “tileset_edit.scn” or something like that.
Exporting a TileSet
With the scene created and opened in the editor, next step will be to create a tileset. Use Scene > Convert To > Tile
Set from the Scene Menu:
Then choose a filename, like “mytiles.res”. Make sure the “Merge With Existing” option is toggled on. This way,
every time the tileset resource file is overwritten, existing tiles are merged and updated (they are referenced by their
unique name, so again, name your tiles properly).
Create a new scene, use any node or node2d as root, then create a TileMap as a child.
Go to the tileset property of this node and assign the one created in previous steps:
Also set the cell size to ‘50’, since that is the size used by the tiles. Quadrant size is a tuning value, which means that
the engine will draw and cull the tilemap in blocks of 16x16 tiles. This value is usually fine and does not need to be
changed, but can be used to tune performance in specific cases (if you know what you are doing).
With all set, make sure the TileMap node is selected. A red grid will appear on screen, allowing to paint on it with the
selected tile on the left pallete.
To avoid moving and selecting the tilemap node accidentally (something common given it’s a huge node), it is recom-
mended that you lock it, using the lock button:
When using a single texture for all the tiles, scaling the tileset (or even moving to a non pixel-aligned location) will
most likely result in filtering artifacts like this:
This can’t be avoided, as it is the way the hardware bilinear filter works. So, to avoid this situation, there are a few
If a game was to be always run in the same device and at the same resolution, positioning controls would be a simple
matter of setting the position and size of each one of them. Unfortunately, it is rarely the case.
Only TVs nowadays have a standard resolution and aspect ratio. Everything else, from computer monitors to tablets,
portable consoles and mobile phones have different resolutions and aspect ratios.
There are several ways to handle this, but for now let’s just imagine that the screen resolution has changed and the
controls need to be re-positioned. Some will need to follow the bottom of the screen, others the top of the screen, or
maybe the right or left margins.
This is done by editing the margin properties of controls. Each control has four margins: left, right, bottom and top.
By default all of them represent a distance in pixels relative to the top-left corner of the parent control or (in case there
is no parent control) the viewport.
When horizontal (left,right) and/or vertical (top,bottom) anchors are changed to END, the margin values become
relative to the bottom-right corner of the parent control or viewport.
Here the control is set to expand it’s bottom-right corner with that of the parent, so when re-sizing the parent, the
control will always cover it, leaving a 20 pixel margin:
Finally, there is also a ratio option, where 0 means left, 1 means right and anything in between is interpolated.
Oh Beautiful GUI!
This tutorial is about advanced skinning of an user interface. Most games generally don’t need this, as they end up just
relying on Label, TextureFrame, TextureButton and TextureProgress.
However, many types of games often need complex user interfaces, like MMOs, traditional RPGs, Simulators, Strat-
egy, etc. These kind of interfaces are also common in some games that include editors to create content, or interfaces
for network connectivity.
Godot user interface uses these kind of controls with the default theme, but they can be skinned to resemble pretty
much any kind of user interface.
Theme
The GUI is skinned through the Theme resource. Theme contains all the information required to change the entire
visual styling of all controls. Theme options are named, so it’s not obvious which name changes what (specialy from
code), but several tools are provided. The ultimate place to look at what each theme option is for each control, which
will always be more up to date than any documentation is the file scene/resources/default_theme/default_theme.cpp.
The rest of this document will explain the different tools used to customize the theme.
A Theme can be applied to any control in the scene. As a result, all children and grand-children controls will use that
same theme too (unless another theme is specified further down the tree). If a value is not found in a theme, it will be
searched in themes higher up in the hierarchy towards the root. If nothing was found, the default theme is used. This
system allows for flexible overriding of themes in complex user interfaces.
Theme Options
var l = Label.new()
l.set_theme(t)
In the example above, a new theme is created. The “font_color” option is changed and then applied to a label. As a
result, the label (and all children and grand children labels) will use that color.
It is possible to override those options without using the theme directly and only for a specific control by using the
override API in Control:
var l = Label.new()
l.add_color_override("font_color",Color(1.0,1.0,1.0))
In the inline help of Godot (help tab) you can check which theme options are overrideable. This is not yet available in
the wiki class reference, but will be soon.
Customizing a Control
If only a few controls need to be skinned. It is often not neccesary to create a new theme. Controls offer their theme
options as special kind of properties. If checked, overriding will take place:
As can be see in the image above, theme options have little check-boxes. If checked, they can be used to override the
value of the theme just for that control.
Creating a Theme
The simplest way to create a theme is to edit a theme resource. Create a Theme from the resource menu, the editor
will appear immediately. Following to this, save it (to, as example, mytheme.thm):
This will create an empty theme that can later be loaded and assigned to controls.
Take some assets attachment:skin_assets.zip, go to the “theme” menu and select “Add Class Item”:
A menu will appear promting the type of control to create. Select “Button”:
Immediately, all button theme options will appear in the property editor, where they can be edited:
Select the “normal” stylebox and create a new “StyleBoxTexture”, then edit it. A texture stylebox basically contains a
texture and the size of the margins that will not stretch when the texture is stretched. This is called “3x3” stretching:
Repeat the steps and add the other assets. There is no hover or disabled image in the example files, so use the same
stylebox as in normal. Set the supplied font as the button font and change the font color to black. Soon, your button
will look different and retro:
Save this theme to the .thm file. Go to the 2D editor and create a few buttons:
Now, go to the root node of the scene and locate the “theme” property, replace it by the theme that was just created. It
should look like this:
So Many Controls..
Yet there are never enough. Creating your own custom controls that act just the way you want them is an obsession
of almost every GUI programmer. Godot provides plenty of them, but they may not work exactly the way you want.
Before contacting the developers with a pull-request to support diagonal scrollbars, at least it will be good to know
how to create these controls easily from script.
Drawing
For drawing, it is recommended to check the [[Custom Draw 2D]] tutorial. The same applies. Some functions are
worth mentioning due to their usefulness when drawing, so they will be detailed next:
Unlike 2D nodes, “size” is very important with controls, as it helps to organize them in proper layouts. For this, the
Control.get_size() method is provided. Checking it during _draw() is vital to ensure everything is kept in-bounds.
Checking Focus
Some controls (such as buttons or text editors) might provide input focus for keyboard or joypad input. Examples
of this are entering text or pressing a button. This is controlled with the Control.set_focus_mode() function. When
drawing, and if the control supports input focus, it is always desired to show some sort of indicator (highight, box, etc)
to indicate that this is the currently focused control. To check for this status, the Control.has_focus() exists. Example
func _draw():
if (has_focus()):
draw_selected()
else:
draw_normal()
Sizing
As mentioned before, size is very important to controls. This allows them to lay out properly, when set into grids,
containers, or anchored. Controls most of the time provide a minimum size to help to properly lay them out. For
example, if controls are placed vertically on top of each other using a VBoxContainer, the minimum size will make
sure your custom control is not squished by the other controls in the container.
To provide this callback, just override Control.get_minimum_size(), for example:
func get_minimum_size():
return Vector2(30,30)
Input
Controls provide a few helpers to make managing input events much esier than regular nodes.
Input Events
There are a few tutorials about input before this one, but it’s worth mentioning that controls have a special input
method that only works when:
• The mouse pointer is over the control.
• The left button was pressed over this control (control always captures input until button si released)
• Control provides keyboard/joypad focus via Control.set_focus_mode.
This function is Control._input_event(event). Simply override it in your control. No processing needs to be set.
extends Control
func _input_event(ev):
if (ev.type==InputEvent.MOUSE_BUTTON and ev.button_index==BUTTON_LEFT and ev.pressed):
print("Left mouse button was pressed!")
For more information about events themselves, check the [[Input Events]] tutorial.
Notifications
Controls also have many useful notifications for which no callback exists, but can be checked with the _notification
callback:
func _notification(what):
if (what==NOTIFICATION_MOUSE_ENTER):
pass # mouse entered the area of this control
elif (what==NOTIFICATION_MOUSE_EXIT):
pass # mouse exited the area of this control
elif (what==NOTIFICATION_FOCUS_ENTER):
pass # control gained focus
elif (what==NOTIFICATION_FOCUS_EXIT):
pass # control lost focus
elif (what==NOTIFICATION_THEME_CHANGED):
pass # theme used to draw the control changed
# update and redraw is recommended if using a theme
elif (what==NOTIFICATION_VISIBILITY_CHANGED):
pass # control became visible/invisible
# check new status with is_visible()
elif (what==NOTIFICATION_THEME_CHANGED):
pass # theme used to draw the control changed
# update and redraw is recommended if using a theme
elif (what==NOTIFICATION_RESIZED):
pass # control changed size, check new size
# with get_size()
elif (what==NOTIFICATION_MODAL_CLOSED):
pass # for modal popups, notification
# that the popup was closed
3.3 Physics
Our world is made of tangible matter. In our world, a piano can’t go through a wall when going into a house. It needs
to use the door. Video games are often like the the real world and Pac-Man can’t go through the walls of his maze
(although he can teleport from the left to the right side of the screen and back).
Anyway, moving sprites around is nice but one day they have to collide properly, so let’s get to the point.
Shapes
The base collidable object in Godot’s 2D world is a Shape2D. There are many types of shapes, all of them inherit this
base class:
• CircleShape2D
• RectangleShape2D
• CapsuleShape2D
• ConvexPolygonShape2D
• ConcavePolygonShape2D
• etc. (there are others check the class list).
Shapes are of type Resource, but they can be created via code easily. For example:
#create a circle
var c = CircleShape2D.new()
c.set_radius(20)
#create a box
var b = RectangleShape2D.new()
b.set_extents(Vector2(20,10))
The main use for shapes is checking collision/intersection and getting resolution information. Shapes are mostly
convex, (except the concavepolygon one, which is just a list of segments to check collision against). This collision
check is done easily with the built-in functions like:
#check if there is a collision between two shapes, each with a transform
if b.collide(b_xform,a,a_xform):
print("OMG Collision!")
Godot will return correct collision and collision info from the different calls to the Shape2D api. Collision between all
shapes and transforms can be done this way, or even obtaining contact information, motion casting, etc.
Transforming Shapes
As seen before in the collide functions, 2D shapes in godot can be transformed by using a regular Matrix32 transform,
meaning the can check collision while scaled, moved and rotated. The only limitation to this is that shapes with curved
sections (such as circle and capsule) can only be scaled uniformly. This means that circle or capsule shapes scaled in
the form of an ellipse will not work properly. This is a limitation on the collision algorithm used (SAT), so make
sure that your circle and capsule shapes are always scaled uniformly!
Even though this sounds good, reality is that collision detection alone is usually not enough in most scenarios. Many
problems start arising as long as the development of the game is in progress:
Games have several dozens, hundreds, thousands! of objects that can collide and be collided. The typical approach is
to test everything against everything in two for loops like this:
for i in colliders:
for j in colliders:
if (i.collides(j)):
do_collision_code()
But this scales really bad. Let’s imagine there are only 100 objects in the game. This means that 100*100=10000
collisions will need to be tested each frame. This is a lot!
Visual Aid
Most of the time, creating a shape via code is not enough. We need to visually place it over a sprite, draw a collision
polygon, etc. It is obvious that we need nodes to create the proper collision shapes in a scene.
Collision Resolution
Imagine we solved the collision issue, we can tell easily and quickly which shapes overlap. If many of them are
dynamic objects that move around, or move according to newtonian physics, solving a collision of multiple objects
can be really difficult code-wise.
To solve all these problems, Godot has a physics and collision engine that is well integrated into the scene system, yet
it allows different levels and layers of functionality. The built-in physics engine can be used for:
• Simple Collision Detection: See Shape2D API.
• Scene Kinematics: Handle shapes, collisions, broadphase, etc as nodes. See Area2D.
• Scene Physics: Rigid bodies and constraints as nodes. See RigidBody2D, and the joint nodes.
Units of Measure
It is often a problem when integrating a 2D Physics engine to a game that such engines are optimized to work using me-
ters as unit of measure. Godot uses a built-in custom 2D physics engine that is designed to function properly in pixels,
so all units and default values used for stabilization are tuned for this, making development more straightforward.
CollisionObject2D
CollisionObject2D is the (virtual) base node for everything that can be collided in 2D. Area2D, StaticBody2D, Kine-
maticBody2D and RigidBody2D all inherit from it. This node contains a list of shapes (Shape2D) and a relative
transform. This means that all collisionable objects in Godot can use multiple shapes at different transforms (off-
set/scale/rotation). Just remember that, as mentioned before, non-uniform scale will not work for circle and capsule
shapes.
StaticBody2D
The simplest node in the physics engine is the StaticBody2D, which provides a static collision. This means that other
objects can collide against it, but StaticBody2D will not move by itself or generate any kind of interaction when
colliding other bodies. It’s just there to be collided.
Creating one of those bodies is not enough, because it lacks collision:
From the previous point, we know that CollisionObject2D derived nodes have an internal lists of shapes and transforms
for collisions, but how to edit them? There are two special nodes for that.
CollisionShape2D
This node is a helper node. It must be created as a direct children of a CollisionObject2D derived node (Area2D.
By itself it does nothing, but when created as a child of the above mentioned nodes, it adds collision shapes to them.
Any amount of CollisionShape2D children can be created, meaning the parent object will simply have mroe collision
shapes. When added/deleted/moved/edited, it updates the list of shapes in the parent node.
At run time, though, this node does not exist (can’t be accessed with get_node() ), since it’s only meant to be an editor
helper. To access the shapes created at runtime, use the CollisionObject2D API directly.
As an example, here’s the scene from the platformer, containing an Area2D with child CollisionObject2D and coin
sprite:
Triggers
CollisionPolygon2D
This one is similar to CollisionShape2D, except that instead of assigning a shape, a polygon can be edited (drawn by
the user) to determine the shape. The polygon can be convex or concave, it doesn’t matter.
Going back, here’s the scene with the StaticBody2D, the static body is the child of a sprite (meaning if the sprite
moves, the collision does too). In turn, the CollisionPolygon is a child of staticbody, meaning it adds collision shapes
to it.
In fact, what CollisionPolygon does is to decompose the polygon in convex shapes (shapes can only be convex,
remember?) and adds them to the CollisionObject2D:
KinematicBody2D
Kinematic bodies are special types of bodies that are meant to be user-controlled. They are not affected by the physics
at all (to other types of bodies, such a character or a rigidbody, these are the same as a staticbody). They have however,
two main uses:
• Simulated Motion: When these bodies are moved manually, either from code or from an AnimationPlayer
(with process mode set to fixed!), the physics will automatically compute an estimate of their linear and angular
velocity. This makes them very useful for moving platforms or other AnimationPlayer-controlled objects (like
a door, a bridge that opens, etc). As an example, the 2d/platformer demo uses them for moving platforms.
• Kinematic Characters: KinematicBody2D also has an api for moving objects (the move() function) while
performing collision tests. This makes them really useful to implement characters that collide against a world,
but that don’t require advanced physics. A special [[tutorial_kinematic_char]].
RigidBody2D
This type of body simulates newtonian physics. It has mass, friction, bounce, and the 0,0 coordinates simulates the
center of mass. When real physics are needed, RigidBody2D is the node to use. The motion of this body is affected
by gravity and/or other bodies.
Rigid bodies are usually active all the time, but when they end up in resting position and don’t move for a while, they
are put to sleep until something else wakes them up. This saves an enormous amount of CPU.
RigidBody2D nodes update their transform constantly, as it is generated by the simulation from a position, linear
velocity and angular velocity. As a result, [STRIKEOUT:this node can’t be scaled]. Scaling the children nodes should
work fine though.
As a plus, as this is very common in games, it is possible to change a RigidBody2D node to behave like a Character
(no rotation), StaticBody or KinematicBody according to different situations (example, an enemy frozen by an ice
beam becomes a StaticBody)
The best way to interact with a RigidBody2D is during the force integration callback. In this very moment, the physics
engine synchronizes state with the scene and allows full modification of the internal parameters (otherwise, as it may
be running in a thread, changes will not take place until next frame). To do this, the following function must be
overridden:
func _integrate_forces(state):
[use state to change the object]
The ‘state‘ parameter is of type Physics2DDirectBodyState. Please do not use this object (state) outside the callback
as it will result in an error.
Contact Reporting
In general, RigidBody2D will not keep track of the contacts, because this can require a huge amount of memory if
thousands of rigid bodies are in the scene. To get contacts reported, simply increase the amount of the “contacts
reported” property from zero to a meaningful value (depending on how many you are expecting to get). The contacts
can be later obtained via the Physics2DDirectBodyState.get_contact_count() and related functions.
Contact monitoring via signals is also available (signals similar to the ones in Area2D, described below) via a boolean
property.
Area2D
1. Override the space parameters for objects entering them (ie. gravity, gravity direction, gravity type, density, etc).
2. Monitor when rigid or kinematic bodies enter or exit the area.
3. Monitor other areas (this is the simplest way to get overlap test)
The second function is the most common. For it to work, the “monitoring” property must be enabled (it is by default).
There are two types of signals emitted by this node:
#Simple, high level notification
body_enter(body:PhysicsBody2D)
body_exit(body:PhysicsBody2D)
area_enter(area:Area2D)
area_exit(body:Area2D)
Areas also by default receive mouse/touchscreen input, providing a lower-level way than controls to i
A few global variables can be tweaked in the project settings for adjusting how 2D physics works:
Leaving them alone is best (except for the gravity, that needs to be adjusted in most games), but there is one specific
parameter that might need tweaking which is the “cell_size”. Godot 2D physics engine used by default a space hashing
algorithm that divides space in cells to compute close collision pairs more efficiently.
If a game uses several colliders that are really small and occupy a small portion of the screen, it might be necessary to
shrink that value (always to a power of 2) to improve efficiency. Likewise if a game uses few large colliders that span
a huge map (of several screens of size), increasing that value a bit might help save resources.
The physics engine may spawn multiple threads to improve performance, so it can use up to a full frame to process
physics. Because of this, when accessing physics variables such as position, linear velocity, etc. they might not be
representative of what is going on in the current frame.
To solve this, Godot has a fixed process callback, which is like process but it’s called once per physics frame (by
default 60 times per second). During this time, the physics engine is in synchronization state and can be accessed
directly and without delays.
To enable a fixed process callback, use the set_fixed_process() function, example:
extends KinematicBody2D
func _fixed_process(delta):
move( direction * delta )
func _ready():
set_fixed_process(true)
It is very often desired to “explore” the world around from our code. Throwing rays is the most common way to do
it. The simplest way to do this is by using the RayCast2D node, which will throw a ray every frame and record the
intersection.
At the moment there isn’t a high level API for this, so the physics server must be used directly. For this, the
Physics2DDirectspaceState class must be used. To obtain it, the following steps must be taken:
Contact Reporting
Remember that not every combination of two bodies can “report” contacts. Static bodies are passive and will not report
contacts when hit. Kinematic Bodies will report contacts but only against Rigid/Character bodies. Area2D will report
overlap (not detailed contacts) with bodies and with other areas. The following table should make it more visual:
Introduction
Yes, the name sounds strange. “Kinematic Character” WTF is that? The reason is that when physics engines came
out, they were called “Dynamics” engines (because they dealt mainly with collision responses). Many attempts were
made to create a character controller using the dynamics engines but it wasn’t as easy as it seems. Godot has one of
the best implementations of dynamic character controller you can find (as it can be seen in the 2d/platformer demo),
but using it requieres a considerable level of skill and understanding of physics engines (or a lot of patience with trial
and error).
Some physics engines such as Havok seem to swear by dynamic character controllers as the best alternative, while
others (PhysX) would rather promote the Kinematic one.
Fixed Process
To manage the logic of a kinematic body or character, it is always advised to use fixed process, which is called the
same amount of times per second, always. This makes physics and motion calculation work in a more predictable way
than using regular process, which might have spikes or lose precision is the frame rate is too high or too low.
extends KinematicBody2D
func _fixed_process(delta):
pass
func _ready():
set_fixed_process(true)
Scene Setup
To have something to test, here’s the scene (from the tilemap tutorial) attachment:kbscene.zip. We’ll be creating a new
scene for the character. Use the robot sprite and create a scene like this:
Let’s add a circular collision shape to the collision body, create a new CircleShape2D in the shape property of Colli-
sionShape2D. Set the radius to 30:
Note: As mentioned before in the physics tutorial, the physics engine can’t handle scale on most types of shapes
(only collision polygons, planes and segments work), so always change the parameters (such as radius) of the
shape instead of scaling it. The same is also true for the kinematic/rigid/static bodies themselves, as their scale
affect the shape scale.
Now create a script for the character, the one used as an example above should work as a base.
Finally, instance that character scene in the tilemap, and make the map scene the main one, so it runs when pressing
play.
Go back to the character scene, and open the script, the magic begins now! Kinematic body will do nothing by default,
but it has a really useful function called move(motion_vector:Vector2). This function takes a Vector2 as an argument,
and tries to apply that motion to the kinematic body. If a collision happens, it stops right at the moment of the collision.
So, let’s move our sprite downwards until it hits the floor:
extends KinematicBody2D
func _fixed_process(delta):
move( Vector2(0,1) ) #move down 1 pixel per physics frame
func _ready():
set_fixed_process(true)
The result is that the character will move, but stop right when hitting the floor. Pretty cool, huh?
The next step will be adding gravity to the mix, this way it behaves a little more like an actual game character:
extends KinematicBody2D
func _fixed_process(delta):
func _ready():
set_fixed_process(true)
Now the character falls smoothly. Let’s make it walk to the sides, left and right when touching the directional keys.
Remember that the values being used (for speed at least) is pixels/second.
This adds simple walking support by pressing left and right:
extends KinematicBody2D
func _fixed_process(delta):
if (Input.is_action_pressed("ui_left")):
velocity.x = -WALK_SPEED
elif (Input.is_action_pressed("ui_right")):
velocity.x = WALK_SPEED
else:
velocity.x = 0
func _ready():
set_fixed_process(true)
Problem?
And.. it doesn’t work very well. If you go to the left against a wall, it gets stuck unless you release the arrow key.
Once it is on the floor, it also gets stuck and it won’t walk. What is going on??
The answer is, what it seems like it should be simple, it isn’t that simple in reality. If the motion can’t be completed,
the character will stop moving. It’s as simple as that. This diagram should illustrate better what is going on:
Basically, the desired motion vector will never complete because it hits the floor and the wall too early in the motion
trajectory and that makes it stop there. Remember that even though the character is on the floor, the gravity is always
turning the motion vector downwards.
Solution!
The solution? This situation is solved by “sliding” by the collision normal. KinematicBody2D provides two useful
functions:
• KinematicBody2D.is_colliding()
• KinematicBody2D.get_collision_normal()
So what we want to do is this:
When colliding, the function move() returns the “remainder” of the motion vector. That means, if the motion vector is
40 pixels, but collision happened at 10 pixels, the same vector but 30 pixels long is returned.
The correct way to solve the motion is, then, to slide by the normal this way:
func _fixed_process(delta):
if (is_colliding()):
var n = get_collision_normal()
motion = n.slide( motion )
velocity = n.slide( velocity )
move( motion )
func _ready():
set_fixed_process(true)
Note that not only the motion has been modified but also the velocity. This makes sense as it helps keep
the new direction too.
The normal can also be used to detect that the character is on floor, by checking the angle. If the normal points up (or
at least, within a certain threshold), the character can be determined to be there.
A more complete demo can be found in the demo zip distributed with the engine, or in the
https://github.com/okamstudio/godot/tree/master/demos/2d/kinematic_char.
Introduction
One of the most common tasks in game development is casting a ray (or custom shaped object) and see what it hits.
This enables complex behaviors, AI, etc. to take place.
This tutorial will explain how to do this in 2D and 3D.
Godot stores all the low level game information in servers, while the scene is just a frontend. As such, ray casting is
generally a lower-level task. For simple raycasts, node such as RayCast and RayCast2D will work, as they will return
every frame what the result of a raycast is.
Many times, though, ray-casting needs to be a more interactive process so a way to do this by code must exist.
Space
In the physics world, Godot stores all the low level collision and physics information in a space. The current
2d space (for 2D Physics) can be obtained by calling CanvasItem.get_world_2d().get_space(). For 3D, it’s Spa-
tial.get_world().get_space().
The resulting space RID can be used in PhysicsServer and Physics2DServer respectively for 3D and 2D.
Acessing Space
Godot physics runs by default in the same thread as game logic, but may be set to run on a separate thread to work
more efficiently. Due to this, the only time accessing space is safe is during the Node._fixed_process(delta) callback.
Accessing it from outside this function may result in an error due to space being locked.
To perform queries into physics space, the Physics2DDirectSpaceState and PhysicsDirectSpaceState must be used.
In code, for 2D spacestate, this code must be used:
func _fixed_process(delta):
var space_rid = get_world_2d().get_space()
var space_state = Physics2DServer.space_get_direct_state(space_rid)
For 3D:
func _fixed_process(delta):
var space_state = get_world().get_direct_space_state()
Raycast Query
For performing a 2D raycast query, the method Physics2DDirectSpaceState.intersect_ray() must be used, for example:
func _fixed_process(delta):
var space_state = get_world().get_direct_space_state()
# use global coordinates, not local to node
var result = space_state.intersect_ray( Vector2(0,0), Vector2(50,100) )
Result is a dictionary, if ray didn’t hit anything, the dictionary will be empty. If it did hit something it will contain
collision information:
if (not result.empty()):
print("Hit at point: ",result.position)
The collision result dictionary, when something hit, has this format:
{
position:Vector2 # point in world space for collision
normal:Vector2 # normal in world space for collision
collider:Object # Object collided or null (if unassociated)
collider_id:ObjectID # Object it collided against
rid:RID # RID it collided against
shape:int # shape index of collider
metadata:Variant() # metadata of collider
}
Collision Exceptions
It is a very common case to attempt casting a ray from a character or another game scene to try to infer properties of
the world around it. The problem with this is that the same character has a collider, so the ray can never leave the
origin (it will keep hitting it’s own collider), as evidenced in the following image.
To avoid self-intersection, the intersect_ray() function can take an optional third parameter which is an array of excep-
tions. This is an example of how to use it from a KinematicBody2D or any other collisionobject based node:
extends KinematicBody2D
func _fixed_process(delta):
var space_state = get_world().get_direct_space_state()
var result = space_state.intersect_ray( get_global_pos(), enemy_pos, [ self ] )
The extra argument is a list of exceptions, can be objects (need Godot 1.1beta2+ for this) or RIDs.
Casting a ray from screen to 3D physics space is useful for object picking. There is not much of a need to do this
because CollisionObject has an “input_event” signal that will let you know when it was clicked, but in case there is
any desire to do it manually, here’s how.
To cast a ray from the screen, the Camera node is needed. Camera can be in two projection modes, perspective and
orthogonal. Because of this, both the ray origin and direction must be obtained. (origin changes in orthogonal, while
direction changes in perspective):
func _input(ev):
if ev.type==InputEvent.MOUSE_BUTTON and ev.pressed and ev.button_index==1:
Of course, remember that during _input(), space may be locked, so save your query for _fixed_process().
3D tutorials
4.1 Graphics
4.1.1 Introduction
Creating a 3D game can be challenging. That extra Z coordinate makes many of the common techniques that helped
to make 2D games simple no longer work. To aid in this transition, it is worth mentioning that Godot uses very similar
APIs for 2D and 3D. Most nodes are the same and are present in both 2D and 3D versions. In fact, it is worth checking
the 3D platformer tutorial, or the 3D kinematic character tutorials, which are almost identical to their 2D counterparts.
In 3D, math is a little more complex than in 2D, so also checking the [[Vector Math]] in the wiki (which were specially
created for game developers, not mathematicians or engineers) will help pave the way into efficiently developing 3D
games.
Spatial Node
Node2D is the base node for 2D. Control is the base node for everything GUI. Following this reasoning, the 3D engine
uses the Spatial node for everything 3D.
Spatial nodes have a local transform, which is relative to the parent node (as long as the parent node is also or inherits
of type Spatial). This transform can be accessed as a 4x3 Transform, or as 3 Vector3 members representing location,
euler rotation (x,y and z angles) and scale.
159
Godot Engine Documentation, Release latest
3D Content
Unlike 2D, where loading image content and drawing is straightforward, 3D is a little more difficult. The content
needs to be created with special 3D tool (usually referred to as DCCs) and exported to an exchange file format in order
to be imported in Godot (3D formats are not as standardized as images).
DCC-Created Models
There are two pipelines to import 3D models in Godot. The first and most common one is through the [[Import 3D]]
importer, which allows to import entire scenes (just as they look in the DCC), including animation, skeletal rigs, blend
shapes, etc.
The second pipeline is through the [[Import Meshes]] importer. This second method allows importing simple .OBJ
files as mesh resources, which can be then put inside a MeshInstance node for display.
Generated Geometry
It is possible to create custom geometry by using the Mesh resource directly, simply create your arrays and use the
Mesh.add_surface function. A helper class is also available, SurfaceTool, which provides a more straightforward API
and helpers for indexing, generating normals, tangents, etc.
In any case, this method is meant for generating static geometry (models that will not be updated often), as creating
vertex arrays and submitting them to the 3D API has a significant performance cost.
Immediate Geometry
If, instead, there is a requirement to generate simple geometry that will be updated often, Godot provides a special
node, ImmediateGeometry which provides an OpenGL 1.x style immediate-mode API to create points, lines, triangles,
etc.
2D in 3D
While Godot packs a powerful 2D engine, many types of games use 2D in a 3D environment. By using a fixed camera
(either orthogonal or perspective) that does not rotate, nodes such as Sprite3D and AnimatedSprite3D can be used to
create 2D games that take advantage of mixing with 3D backgrounds, more realistic parallax, lighting/shadow effects,
etc.
The disadvantage is, of course, that added complexity and reduced performance in comparison to plain 2D, as well as
the lack of reference of working in pixels.
Environment
Besides editing a scene, it is often common to edit the environment. Godot provides a WorldEnvironment node
that allows changing the background color, mode (as in, put a skybox), and applying several types of built-in post-
processing effects. Environments can also be overriden in the Camera.
3D Viewport
Editing 3D scenes is done in the 3D tab. This tab can be selected manually, but it will be automatically enabled when
a Spatial node is selected.
Default 3D scene navigation controls are similar to Blender (aiming to have some sort of consistency in the free
software pipeline..), but options are included to customize mouse buttons and behavior to be similar to other tools in
Editor Settings:
Coordinate System
Godot uses the metric system for everything. 3D Physics and other areas are tuned for this, so attempting to use a
different scale is usually a bad idea (unless you know what you are doing).
When working with 3D assets, it’s always best to work in the correct scale (set your DCC to metric). Godot allows
scaling post-import and, while this works in most cases, in rare situations it may introduce floating point precision
issues (and thus, glitches or artifacts) in delicate areas such as rendering or physics. So, make sure your artists always
work in the right scale!
The Y coordinate is used for “up”, though for most objects that need alignment (like lights, cameras, capsule collider,
vehicle, etc), the Z axis is used as a “pointing towards” direction. This convention roughly means that:
• X is sides
• Y is up/down
• Z is front/back
Moving objects in the 3D view is done through the manipulator gizmos. Each axis is represented by a color: Red,
Green, Blue represent X,Y,Z respectively. This convention applies to the grid and other gizmos too (and also to the
shader language, ordering of components for Vector3,Color,etc).
View Menu
The view options are controlled by the ‘[view]‘ menu. Pay attention to this little menu inside the window because it is
often overlooked!
Default Lighting
Customizing this (and other default view options) is also possible via the settings menu:
which opens this window, allowing to customize ambient light color and default light direction:
Cameras
No matter how many objects are placed in 3D space, nothing will be displayed unless a Camera is also added to the
scene. Cameras can either work in orthogonal or perspective projections:
Cameras are associated and only display to a parent or grand-parent viewport. Since the root of the scene tree is a
viewport, cameras will display on it by default, but if sub-viewports (either as render target or picture-in-picture) are
desired, they need their own children cameras to display.
When dealing with multiple cameras, the following rules are followed for each viewport:
• If no cameras are present in the scene tree, the first one that enters it will become the active camera. Further
cameras entering the scene will be ignored (unless they are set as current).
• If a camera has the “current” property set, it will be used regardless of any other camera in the scene. If the
property is set, it will become active, replacing the previous camera.
• If an active camera leaves the scene tree, the first camera in tree-order will take it’s place.
Lights
There is no limitation on the number of lights and types in Godot. As many as desired can be added (as long as
performance allows). Shadow maps are, however, limited. The more they are used, the less the quality overall.
It is possible to use [[Light Baking]], to avoid using large amount of real-time lights and improve performance.
4.1.2 Materials
Introduction
Materials can be applied to most visible 3D objects, they basically are a description to how light reacts to that object.
There are many types of materials, but the main ones are the FixedMaterial and ShaderMaterial. Tutorials for each of
them exist [[Fixed Material]] and [[Shader Material]].
This tutorial is about the basic properties shared between them.
Flags
Materials, no matter which type they are, have a set of flags associated. Each has a different use and will be explained
as follows.
Visible
Toggles whether the material is visible. If unchecked, the object will not be shown.
Godot by default only shows geometry faces (triangles) when facing the camera. To do this it needs them to be in
view in clockwise order. This saves a lot of GPU power by ensuring that not visible triangles are not drawn.
Some flat objects might need to be drawn all the times though, for this the “double sided” flag will make sure that no
matter the facing, the triangle will always be drawn. It is also possible to invert this check and draw
counter-clockwise looking faces too, though it’s not very useful except for a few cases (like drawing outlines).
Unshaded
Objects are always black unless light affects them, and their shading changes according to the type and direction of
lights. When this flag is turned on, the diffuse color is displayed right the same as it appears in the texture or parameter:
On Top
When this flag is turned on, te object will be drawn after everything else has been drawn and without a depth test. This
is generally only useful for HUD effects or gizmos.
Ligthmap on UV2
When using lightmapping (see the [[Light Baking]] tutorial), this option determines that the lightmap should be ac-
cessed on the UV2 array instead of regular UV.
Parameters
Blend Mode
Objects are usually blended in Mix mode. Other blend modes (Add and Sub) exist for special cases (usually particle
effects, light rays, etc) but materials can be set to them:
Line Width
When drawing lines, the size of them can be adjusted here per material.
This is a tricky but very useful setting. By default, opaque objects are drawn using the depth buffer and translucent
objects are not (but are sorted by depth). This behavior can be changed here. The options are:
• Always: Draw objects with depth always, even those with alpha. This often results in glitches like the one in
the first image (which is why it’s not the default).
• Opaque Only: Draw objects with depth only when they are opaque, and do not se depth for alpha. This is the
default because it’s fast, but it’s not the most correct setting. Objects with transparency that self-intersect will
always look wrong, specially those that mix opaque and transparent areas, like grass tree leaves, etc. Objects
with transparency also can’t cast shadows, this is evident i the second image.
• Alpha Pre-Pass: The same as above, but a depth pass is performed for the opaque areas of objects with trans-
parency. This makes objects with transparency look much more correct. In the third image it is evident how the
leaves cast shadows between them and into the floor. This setting is turned off by default because, while on PC
this is not very costly, mobile devices suffer a lot when this setting is turned on, so use it with care.
• Never: Never use the depth buffer for this material. This is mostly useful in combination with the “On Top” flag
explained above.
Introduction
Fixed materials (originally Fixed Pipeline Materials) are the most common type of materials, using the most common
material options found in 3D DCCs (such as Maya, 3DS Max or Blender). The big advantage of using them is that 3D
artists are very familiar with this layout. They also allow to try out different things quickly without the need of writing
shaders. Fixed Materials inherit from Material, which also has several options. If you haven’t read it before, reading
the [[Materials]] tutorial is recommended.
Options
Here is the list of all the options available for fixed materials:
Fixed Flags
These are a set of flags that control general aspects of the material.
Use Alpha
This flag needs to be active for transparent materials to blend with what is behind, otherwise display will always be
opaque. Do not enable this flag unless the material really needs it, because it can severely affect performance and
quality. Materials with transparency will also not cast shadows (unless they contain opaque areas and the “opaque
pre-pass” hint is turned on, see the [[Materials]] tutorial for more information).
Vertex color painting is a very common technique to add detail to geometry. 3D DCCs all support this, and many even
support baking occlusion to it. Godot allows this information to be used in the fixed material by modulating the diffuse
color when enabled.
Point Size
Point size is used to set the point size (in pixels) for when rendering points. This feature is mostly used in tools and
HUDs
Discard Alpha
When alpha is enabled (see above) the invisible pixels are blended with what is behind them. In some combinations
(of using alpha to render depth) it may be possible that invisible pixels cover other objects.
If this is the case, enable this option for the material. This option is often used in combination with “opaque pre-pass”
hint (see the [[Materials]] tutorial for more information).
Parameters
Regular shader materials allow custom lighting code. Fixed materials come with four predefined shader types:
• Lambert: The standard diffuse light, where the amount of light is proportional to the angle with the light
emissor.
• Wrap: A variation on Lambert, where the “coverage” of the light can be changed. This is useful for many types
of materials such as wood, clay, hair, etc.
• Velvet: This is similar to Lambert, but adds light scattering in the edges. It’s useful for leathers and some types
of metals.
• Toon: Standard toon shading with a coverage parameter. The specular component also becomes toon-ized.
Detail is a second diffuse texture which can be mixed with the first one (more on textures later!). Detail blend and mix
control how these are added together, here’s an example of what detail textures are for:
Normal Depth
Normal depth controls the inensity of the normal-mapping as well as the direction. On 1 (the default) normalmapping
applies normaly, on -1 the map is inverted and on 0 is disabled. Intermediate or greater values are accepted. Here’s
how it’s supposed to look:
Glow
This value controls how much of the color is sent to the glow buffer. It can be greater than 1 for a stronger effect. For
glow to work, a WorldEnvironment must exist with Glow activated.
Blend Mode
Objects are usually blended in Mix mode. Other blend modes (Add and Sub) exist for special cases (usually particle
effects, light rays, etc) but materials can be set to them:
When drawing points or lines, the size of them can be adjusted here per material.
Textures
Almost all of the parameters above can have a texture assigned to them. There are four options to where they can get
their UV coordinates:
• UV Coordinates (UV Array): This is the regular UV coordinate array that was imported with the model.
• UV x UV XForm: UV Coordinates multiplied by the UV Xform matrix.
• UV2 Coordinates: Some imported models might have come with a second set of UV coordinates. These are
common for detail textures or for baked light textures.
• Sphere: Spherical coordinates (difference of the normal at the pixel by the camera normal).
The value of every pixel of the texture is multiplied by the original parameter. This means that if a texture is loaded
for diffuse, it will be multiplied by the color of the diffuse color parameter. Same applies to all the others except for
specular exponent, which is replaced.
[[https://creativecommons.org/licenses/by/3.0/legalcode]] license.
Introduction
For the most common cases, [[Fixed Material]] are enough to create the desired textures or look and feel. Shader
materials are a step beyond that adds a huge amount of flexibility. With them, it is possible to:
• Create procedural texures.
• Create complex texture blendings.
• Create animated materials, or materials that change with time.
• Create refractive effects or other advanced effects.
• Create special lighting shaders for more exotic materials.
• Animate vertices, like tree leaves or grass.
• And much more!
Traditionally, most engines will ask you to learn GLSL, HLSL or CG, which are pretty complex for the skillset of most
artists. Godot uses a simplified version of a shader language that will detect errors as you type, so you can see your
edited shaders in real-time. Additionally, it is possible to edit shaders using a visual graph editor (NOTE: Currently
disabled! work in progress!).
Creating a ShaderMaterial
Create a new ShaderMaterial in some object of your choice. Go to the “Shader” property, then create a new “Shader”:
Edit the newly created shader, and the shader editor will open:
There are three code tabs open, the first is for the vertex shader, the second for the fragment and the third for the
lighting. The shader language is documented in it’s [[Shader]] so a small example will be presented next.
Create a very simple fragment shader that writes a color:
uniform color col;
DIFFUSE = col.rgb;
Code changes take place in real-time. If the code is modified, it will be instantly recompiled and the object will be
updated. If a typo is made, the editor will notify of the compilation failure:
Finally, go back and edit the material, and the exported uniform will be instantly visible:
This allows to very quickly create custom, complex materials for every type of object.
4.1.5 Lighting
Introduction
Lights emit light that mix with the materials and produces a visible result. Light can come from several types of
sources in a scene:
• From the Material itself, in the form of the emission color (though it does not affect nearby objects unless
baked).
• Light Nodes: Directional, Omni and Spot.
• Ambient Light in the Environment.
• Baked Light (read [[Light Baking]]).
The emission color is a material property, as seen in the previous tutorials about materials (go read them if you didn’t
at this point!).
Light Nodes
As mentioned before, there are three types of light nodes: Directional, Ambient and Spot. Each has different uses and
will be described in detail below, but firs let’s take a look at the common parameters for lights:
Directional Light
This is the most common type of light and represents the sun. It is also the cheapest light to compute and should be
used whenever possible (although it’s not the cheapest shadow-map to compute, but more on that later). Directional
light nodes are represented by a big arrow, which represent the direction of the light, however the position of the node
does not affect the lighting at all, and can be anywhere.
Basically what faces the light is lit, what doesn’t is dark. Most lights have specific parameters but directional lights
are pretty simple in nature so they don’t.
Omni Light
Omni light is a point that throws light all around it up to a given radius (distance) that can be controlled by the user.
The light attenuates with the distance and reaches 0 at the edge. It represents lamps or any other light source that
comes from a point.
The attenuation curve for these kind of lights in nature is computed with an inverse-quadratic function that never
reaches zero and has almost infinitely large values near the emissor.
This makes them considerably inconvenient to tweak for artists, so Godot simulates them with an artist-controlled
exponential curve instead.
Spot Light
Spot lights are similar to Omni lights, except they only operate between a given angle (or “cutoff”). They are useful
to simulate flashlights, car lights, etc. This kind of light is also attenuated towards the opposite direction it points to.
Ambient Light
Ambient light can be found in the properties of a WorldEnvironment (remember only one of such can be instanced per
scene). Ambient light consists of a uniform light and energy. This light is applied the same to every single pixel of the
rendered scene, except to objects that used baked light.
Baked Light
Baked Light stands for pre-computed ambient light. It can serve multiple purposes, such as baking light emissors that
are not going to be used in real-time, and baking light bounces from real-time lights to add more realism to a scene
(see Baked Light]] tutorial for more information).
Introduction
Simply throwing a light is not enough to realistically illuminate a scene. It should be, in theory, but given the way
video hardware works, parts of objects that should not be reached by light are lit anyway.
Most people (including artists), see shadows as something projected by light, as if they were created by the light itself
by darkening places that are hidden from the light source.
This is actually not correct and it’s important to understand that shadows are places where light simply does not reach.
As a rule (and without counting indirect light) if a light is turned off, the places where shadow appear should remain
the same. In other words, shadows should not be seen as something “added” to the scene, but as an area that “remains
dark”.
All light types in Godot can use shadow mapping, and all support several different techniques that trade quality by
performance. Shadow mapping uses a texture storing the “depth view” of the light and checks against it in real-time
for each pixel it renders.
The bigger the resolution of the shadow map texture, the more detail the shadow has, but more video memory and
bandwidth consumed (which means frame-rate goes down).
Directional lights can affect a really big area. The bigger the scene, the bigger the affected area. Given the shadow
map resolution stays the same, the same amount of shadow pixels cover a bigger area, resulting in blocky shadows.
Multiple techniques exist to deal with resolution problems, but the most common one is PSSM (Parallel Split Shadow
Maps):
These techniques divide the view in 2 or 4 sections, and a shadow is rendered for each. This way, close obects can use
larger shadow while further away objects will use one in less detail, but in proportion this seems to make the shadow
map size increase while it’s actually kept thr same. Of course, this technique is not free, the more splits the more the
performance goes down. On mobile, generally it is convenient to not sure more tan 2 splits.
An alternative technique is PSM (Perspective Shadow Mapping). This technique is much cheaper than PSSM (as
cheap as orthogonal), but it only really works for a few camera angles respect to the light. In other words, PSM is
only useful for games where the camera direction and light direction are both fixed, and the light is not parallel to the
camera (which is when PSM completely breaks).
Omnidirectional lights are also troubesome. How to represent 360 degrees of light with a single texture? There are two
alternatives, the first one is to use DPSM (Dual Paraboloid Shadow Mapping). This technique is fast, but it requires
DISCARD to be used (which makes it not very usable on mobile). DPSM can also look rather bad if the geometry is
not tesselated enough, so more vertices might be neccesary if it doesnt’r look tight. The second option is to simply not
use a shadow map, and use a shadow cubemap. This is faster, but requires six passes to render all directions and is not
supported on the current (GLES2) renderer.
Spot light shadows are generally the simpler, just needing a single texture and no special techniques.
img/shadow_spot.png
Shadows Parameters
The fact that shadows are actually a texture can generate several problems. The most common is Z fighting (lines at
the edge of the objects that cast the shadows. There are two ways to fix this, the first is to tweak the offset parameters,
and the second is to use a filtered shadow algorithm, which generally looks better and has not as many glitches, but
consumes more GPU time.
Adjusting Z-Offset
So, you have decided to go with non-filtered shadows because they are faster, you want a little more detail or maybe
you just like the sexy saw-like shadow outlines because they remind you of your favorite previous-gen games. Truth
is this can kind of be a pain, but most of the time it can be solved to nice results. There is no magic number and
whathever resut you come up will be different from scene to scene, it just takes a while of tweaking. Let’s go step by
step.
First step is to turn on the shadows, let’s assume that both Z-Offset and Z-Slope-Scale are at 0. You will be greeted by
this:
Holy crap, shadow is all around the place and extremely glitchy! this happens because the shadow is fighting with the
same geometry that is casting it. This is called “self-shadowing”. To avoid this meaningless fight, you realize you need
to make peace between the shadow and the geometry, so you push back the shadow a little by adjusting the shadow
Z-Offset. This improves things a lot:
But it’s not quite perfect, self shadowing did not dissapear completely. So close to perfection but still not there.. so in
a turn of greed you increase the Z-Offset even more!
And it gets rid of those self-shadowings! hooray! except something is wrong.. oh, right. Being pushed back too
much, the shadows start disconnecting from their casters, which looks pretty awful. Ok, you go back to the previous
Z-offset.
This is when Z-Slope-Scale comes to save the day. This setting makes shadow caster objects thinner, so the borders
don’t self-shadow:
Aha! Finally something that looks acceptable. It’s perfectly acceptable and you can perfectly ship a game that looks
like this (imagine you are looking at Final Fantasy quality art btw, not this horribe attempt at 3D modelling). There
may be very tiiny bits left of self shadowing that no one cares about, so your inextinguishable greed kicks in again and
you raise the Z-Slope Scale again:
Well, that was too much, shadows casted are way too thin and don’t look good anymore. Well, though luck, the
previous setting was good anyway, let’s accept that perfection does not exist and move on to something else.
Important!
If you are using shadow maps with directional lights, make sure that the view distance of the camera is set to an
optimal range. This means, if the distance between your camera and the visible end of the scene is 100, then set the
view distance to that value. If a greater than necessary value is used, the shadow maps will lose detail as they will try
to cover a bigger area.
So, always make sure to use the optimal range!
Shadow Filtering
Raw shadows are blocky. Increasing their resolution just makes smaller blocks, but they are still blocks.
Godot offers a few ways to filter them (shadow in the example is low-resolution on purpose!):
PCF5 and PCF13 are simple texture-space filtering. Will make the texture a little more acceptable but still needs
considerable resolution for it to look good.
ESM is a more complex filter and has a few more tweaking parameters. ESM uses shadow blurring (amount of blur
passes and multiplier can be adjusted).
Introduction
Normally, an artist does all the 3D modelling, then all the texturing, looks at his or her awesome looking model in
the 3D DCC and says “looks fantastic, ready for integration!” then goes into the game, lighting is setup and the game
runs.
So where does all this HDR stuff thing come from? The idea is that instead of dealing with colors that go from black
to white (0 to 1), we use colors whiter than white (for example, 0 to 8 times white).
To be more practical, imagine that in a regular scene, the intensity of a light (generally 1.0) is set to 5.0. The whole
scene will turn very bright (towards white) and look horrible.
After this the luminance of the scene is computed by averaging the luminance of every pixel of it, and this value is
used to bring the scene back to normal ranges. This last operation is called tone-mapping. Finally, we are at a similar
place from where we started:
Except the scene is more contrasted, because there is a higher light range in play. What is this all useful for? The idea
is that the scene luminance will change while you move through the world, allowing situations like this to happen:
Additionally, it is possible to set a threshold value to send to the glow buffer depending on the pixel luminance. This
allows for more realistic light bleeding effects in the scene.
The problem with this technique is that computer monitors apply a gamma curve to adapt better to the way the human
eye sees. Artists create their art on the screen too, so their art has an implicit gamma curve applied to it.
The color space where images created in computer monitors exist is called “sRGB”. Every visual content that people
has on their computers or downloads from the internet (such as pictures, movies, porn, etc) is in this colorspace.
The mathematics of HDR require that we multiply the scene by different values to adjust the luminance and exposure
to different light ranges, and this curve gets in the way as we need colors in linear space for this.
Working in HDR is not just pressing a switch. First, imported image assets must be converted to linear space on
import. There are two ways to do this:
This is the most compatible way of using linear-space assets and it will work everywhere including all mobile devices.
The main issue with this is loss of quality, as sRGB exists to avoid this same problem. Using 8 bits per channel to
represent linear colors is inefficient from the point of view of the human eye. These textures might be later compressed
too, which makes the problem worse.
In any case though, this is the easy solution that works everywhere.
This is the most correct way to use assets in linear-space, as the texture sampler on the GPU will do the conversion
after reading the texel using floating point. This works fine on PC and consoles, but most mobile devices do no support
After all the rendering is done, the linear-space rendered image must be converted back to sRGB. To do this, simply
enable sRGB conversion in the current Environment (more on that below).
Keep in mind that sRGB [STRIKEOUT:> Linear and Linear]> sRGB conversions must always be both enabled.
Failing to enable one of them will result in horrible visuals suitable only for avant garde experimental indie games.
Parameters of HDR
HDR is found in the Environment resource. These are found most of the time inside a WorldEnvironment node, or set
in a camera. There are many parameters for HDR:
ToneMapper
The ToneMapper is the heart of the algorithm. Many options for tonemappers are provided:
• Linear: Simplest tonemapper. It does it’s job for adjusting scene brightness, but if the differences in light are
too big, it will cause colors to be too saturated.
• Log: Similar to linear, but not as extreme.
• Reinhardt: Classical tonemapper (modified so it will not desaturate as much)
• ReinhardtAutoWhite: Same as above, but uses the max scene luminance to adjust the white value.
Exposure
The same exposure parameter as in real cameras. Controls how much light enters the camera. Higher values will result
in a brighter scene and lower values will result in a darker scene.
White
Glow Threshold
Determine above which value (from 0 to 1 after the scene is tonemapped), light will start bleeding.
Glow Scale
Min Luminance
Lower bound value of light for the scene at which the tonemapper stops working. This allows dark scenes to remain
dark.
Max Luminance
Upper bound value of light for the scene at which the tonemapper stops working. This allows bright scenes to remain
saturated.
Auto-exposure will change slowly and will take a while to adjust (like in real cameras). Bigger values means faster
adjustment.
Introduction
Godot follows a balanced performance philosophy. In performance world, there are always trade-offs, which consist
in trading speed for usability and flexibility. Some practical examples of this are:
• Rendering objects efficiently in high amounts is easy, but when a large scene must be rendered it can become
inefficient. To solve this, visibility computation must be added to the rendering, which makes rendering less
efficient, but at the same less objects are rendered, so efficiency overall improves.
• Configuring the properties of every material for every object that needs to be renderer is also slow. To solve this,
objects are sorted by material to reduce the costs, but at the same time sorting has a cost.
• In 3D physics a similar situation happens. The best algorithms to handle large amounts of physics objects (such
as SAP) are very slow at insertion/removal of objects and ray-casting. Algorithms that allow faster insertion and
removal, as well as ray-casting will not be able to handle as many active objects.
And there are many more examples of this! Game engines strive to be general purpose in nature, so balanced algo-
rithms are always favored over algorithms that might be the fast in some situations and slow in others.. or algorithms
that are fast but make usability more difficult.
Godot is not an exception and, while it is designed to have backends swappable for different algorithms, the default
ones (or more like, the only ones that are there for now) prioritize balance and flexibility over performance.
With this clear, the aim of this tutorial is to explain how to get the maximum performance out of Godot.
Rendering
3D rendering is one of the most difficult areas to get performance from, so this section will have a list of tips.
Godot renderer is a little different to what is out there. It’s designed to minimize GPU state changes as much as
possible. FixedMaterial does a good job at reusing materials that need similar shaders but, if custom shaders are used,
make sure to reuse them as much as possible. Godot’s priorities will be like this:
• Reusing Materials: The less amount of different materials in the scene, the faster the rendering will be. If a
scene has a huge amount of objects (in the hundreds or thousands) try reusing the materials or in the worst case
use atlases.
• Reusing Shaders: If materials can’t be reused, at least try to re-use shaders (or FixedMaterials with different
parameters but same configuration).
If a scene has, for example, 20.000 objects with 20.000 different materials each, rendering will be really slow. If the
same scene has 20.000 objects, but only uses 100 materials, rendering will be blazing fast.
It is a common thought that the lower the polygons in a model, the faster it will be rendered. This is really relative and
depends on many factors.
On a modern PC and consoles, vertex cost is low. Very low. GPUs originally only rendered triangles, so all the
vertices:
Nowadays, all this is handled inside the GPU, so the performance is extremely high. 3D artists usually have the wrong
feeling about polycount performance because 3D DCCs (such as Blender, Max, etc) need to keep geometry in CPU
memory in order for it to be edited, reducing actual performance. Truth is, a model rendered by a 3D engine is much
more optimal than how 3D DCCs display them.
On mobile devices, the story is different. PC and Console GPUs are brute-force monsters that can pull as much
electricity as they need from the power grid. Mobile GPUs are limited to a tiny battery, so they need to be a lot more
power efficient.
To be more efficient, mobile GPUs attempt to avoid overdraw. This means, the same pixel on the screen being rendered
(as in, with lighting calculation, etc) more than once. Imagine a town with several buildings, GPUs don’t really know
what is visible and what is hidden until they draw it. A house might be drawn and then another house in front of it
(rendering happened twice for the same pixel!). PC GPUs normally don’t care much about this and just throw more
pixel processors to the hardware to increase performance (but this also increases power consumption).
On mobile, pulling more power is not an option, so a technique called “Tile Based Rendering” is used (almost every
mobile hardware uses a variant of it), which divide the screen into a grid. Each cell keps the list of triangles drawn to it
and sorts them by depth to minimize overdraw. This technique improves performance and reduces power consumption,
but takes a toll on vertex performance. As a result, less vertices and triangles can be processed for drawing.
Generally, this is not so bad, but there is a corner case on mobile that must be avoided, which is to have small objects
with a lot of geometry within a small portion of the screen. This forces mobile GPUs to put a lot of strain on a single
screen cell, considerably decreasing performance (as all the other cells must wait for it to complete in order to display
the frame).
To make it short, do not worry about vertex count so much on mobile, but avoid concentration of vertices in small
parts of the screen. If, for example, a character, NPC, vehicle, etc is far away (so it looks tiny), use a smaller level of
detail (LOD) model instead.
An extra situation where vertex cost must be considered is objects that have extra processing per vertex, such as:
• Skinning (skeletal animation)
• Morphs (shape keys)
• Vertex Lit Objects (common on mobile)
Texture Compression
Godot offers to compress textures of 3D models when imported (VRAM compression). Video Ram compression is
not as efficient in size as PNG or JPG when stored, but increase performance enormously when drawing.
This is because the main goal of texture compression is bandwidth reduction between memory and the GPU.
In 3D, the shapes of objects depend more on the geometry than the texture, so compression is generally not noticeable.
In 2D, compression depends more on shapes inside the textures, so the artifacting resulting from the compression is
more noticeable.
As a warning, most Android devices do not support texture compression of textures with transparency (only opaque),
so keep this in mind.
Transparent Objects
As mentioned before, Godot sorts objects by material and shader to improve performance. This, however, can not
be done on transparent objects. Transparent objects are rendered from back to front to make blending with what is
behind work. As a result, please try to keep transparent objects to a minimum! If an object has a small section with
transparency, try to make that section a separate material.
As also mentioned before, using objects with less vertices can improve performance in some cases. Godot has a
very simple system to use level of detail, GeometryInstance based objects have a visibility range that can be defined.
Having several GeometryInstance objects in different ranges works as LOD.
If several identical objects have to be drawn in the same place or nearby, try using MultiMesh instead. MultiMesh
allows drawing of dozens of thousands of objects at very little performance cost, making it ideal for flocks, grass,
particles, etc.
Bake Lighting
Small lights are usually not a performance issue. Shadows a little more. In general, if several lights need to affect
a scene, it’s ideal to bake it ([[Light Baking]]). Baking can also improve the scene quality by adding indirect light
bounces.
If working on mobile, baking to texture is recommended, since this method is even faster.
Godot 3D skeleton support is currently quite rudimentary. Skeleton node and class were disigned mainly to support
importing skeletal animations as set of transformation matrices.
Skeleton node
Skeleton node can be directly added anywhere you want on scene. Usually mesh is a child of Skeleton, as it easier to
manipulate this way, as Transforms within skeleton are relative to where Skeleton is. But you can specify Skeleton
node in every MeshInstance.
Being obvious, Skeleton is intended to deform meshes, and consists of structures called “bones”. Each “bone” is
represented as Transform, which is applied to a group of vertices within a mesh. You can directly control a group
of vertices from Godot. For that please reference MeshDataTool class, method set_vertex_bones. This class is very
powerful but not documented.
The “bones” are organized in hierarchy, every bone, except for root bone(s) have parent. Every bone have associated
name you can use to refer to it (e.g. “root” or “hand.L”, etc). Also bones are all numbered, these numbers are bone
IDs. Bone parents are referred by their numbered IDs.
For the rest of the article we consider the following scene
main (Spatial) - script is always here
== skel (Skeleton)
==== mesh (MeshInstance)
This scene is imported from Blender. It contains arm mesh with 2 bones - upperarm and lowerarm, with lowerarm
parented to upperarm
Skeleton class
You can view Godot internal help for descriptions of every function. Basically all operations on bones are done using
their numeric ID. You can convert from name to numeric ID and vise versa.
To find number of bones in skeleton we use get_bone_count() function
extends Spatial
var skel
func _ready():
skel = get_node("skel")
var id = skel.find_bone("upperarm")
print("bone id:", id)
var parent = skel.get_bone_parent(id)
print("bone parent id:", id)
extends Spatial
var skel
func _ready():
skel = get_node("skel")
var id = skel.find_bone("upperarm")
print("bone id:", id)
Now, we want to do something interesting with ID except for printing it. Also, we might need additional information
- to find bone parents to complete chain, etc. This all is done with get/set_bone_* functions.
To find bone parent we use get_bone_parent(id) function
extends Spatial
var skel
func _ready():
skel = get_node("skel")
var id = skel.find_bone("upperarm")
print("bone id:", id)
var parent = skel.get_bone_parent(id)
print("bone parent id:", id)
Bone transforms is the thing why we’re here at all. There are 3 kind of transforms - local, global, custom.
To find bone local Transform we use get_bone_pose(id) function
extends Spatial
var skel
func _ready():
skel = get_node("skel")
var id = skel.find_bone("upperarm")
print("bone id:", id)
var parent = skel.get_bone_parent(id)
print("bone parent id:", id)
var t = skel.get_bone_pose(id)
print("bone transform: ", t)
So we see 3x4 matrix there, with first column of 1s. What can we do about that? it is Transform, so we can do
everything we can do with Transform, basically translate, rotate and scale. Also we can multiply transforms to have
complex transforms. Remember, “bones” in Godot are just Transforms over a group of vertices. Also we can copy
Transforms of other objects there. So lets rotate our “upperarm” bone:
extends Spatial
var skel
var id
func _ready():
skel = get_node("skel")
id = skel.find_bone("upperarm")
print("bone id:", id)
var parent = skel.get_bone_parent(id)
print("bone parent id:", id)
var t = skel.get_bone_pose(id)
print("bone transform: ", t)
set_process(true)
func _process(dt):
var t = skel.get_bone_pose(id)
t = t.rotated(Vector3(0.0, 1.0, 0.0), 0.1 * dt)
skel.set_bone_pose(id, t)
Now we can rotate individual bones. The same happens for scale and translate - try these on your own and see results.
What we used now was local pose. By default all bones are not modified. But this Transform tells us nothing about
relationship between bones. This information is needed for quite a number of tasks. How can we get it? here comes
global transform:
To find bone global Transform we use get_bone_global_pose(id) function
We will find global Transform for lowerarm bone
extends Spatial
var skel
func _ready():
skel = get_node("skel")
var id = skel.find_bone("lowerarm")
print("bone id:", id)
var parent = skel.get_bone_parent(id)
print("bone parent id:", id)
var t = skel.get_bone_global_pose(id)
print("bone transform: ", t)
As you see, this transform is not zeroed. While being called global, it is actually relative to Skeleton origin. For root
bone, origin is always at 0 if not modified. Lets print origin for our lowerarm bone:
extends Spatial
var skel
func _ready():
skel = get_node("skel")
var id = skel.find_bone("lowerarm")
print("bone id:", id)
var parent = skel.get_bone_parent(id)
print("bone parent id:", id)
var t = skel.get_bone_global_pose(id)
print("bone origin: ", t.origin)
You will see a number. What does this number mean? It is a rotation point of Transform. So it is base part of the
bone. In Blender you can go to Pose mode and try there to rotate bones - they will rotate around their origin. But what
about tip? We can’t know things like bone length, which we need for many things, without knowing tip location. For
all bones in chain except for last one we can calculate tip location - it is simply a child bone origin. Yes, there are
situations when this is not true, for non-connected bones. But that is OK for us for now, as it is not important regarding
Transforms. But the leaf bone tip is nowhere to be found. Leaf bone is a bone without children. So you don’t have
any information about its tip. But this is not a showstopper. You can overcome this by either adding extra bone to the
chain or just calculating leaf bone length in Blender and store the value in your script.
Now as you know basics we can apply these to make full FK-control of our arm (FK is forward-kinematics)
To fully control our arm we need the following parameters:
• Upperarm angle x, y, z
• Lowerarm angle x, y, z
Set up Camera so that arm is properly visible. Rotate DirectionLight so that arm is properly lit
while in scene play mode.
Now we need to setup way to change them. Just lets use keys for that.
Please create 7 actions under project settings:
func _ready():
skel = get_node("arm/Armature/Skeleton")
set_process(true)
var bone = "upperarm"
var coordinate = 0
func set_bone_rot(bone, ang):
var b = skel.find_bone(bone)
var rest = skel.get_bone_rest(b)
var newpose = rest.rotated(Vector3(1.0, 0.0, 0.0), ang.x)
var newpose = newpose.rotated(Vector3(0.0, 1.0, 0.0), ang.y)
var newpose = newpose.rotated(Vector3(0.0, 0.0, 1.0), ang.z)
skel.set_bone_pose(b, newpose)
func _process(dt):
if Input.is_action_pressed("select_x"):
coordinate = 0
elif Input.is_action_pressed("select_y"):
coordinate = 1
elif Input.is_action_pressed("select_z"):
coordinate = 2
elif Input.is_action_pressed("select_upperarm"):
bone = "upperarm"
elif Input.is_action_pressed("select_lowerarm"):
bone = "lowerarm"
elif Input.is_action_pressed("increment"):
if bone == "lowerarm":
lowerarm_angle[coordinate] += 1
elif bone == "upperarm":
upperarm_angle[coordinate] += 1
elif Input.is_action_pressed("decrement"):
if bone == "lowerarm":
lowerarm_angle[coordinate] -= 1
elif bone == "upperarm":
upperarm_angle[coordinate] -= 1
set_bone_rot("lowerarm", lowerarm_angle)
set_bone_rot("upperarm", upperarm_angle)
Pressing keys 1/2 select upperarm/lowerarm, select axis by pressing x, y, z, rotate using numpad “+”/”-“
This way you fully control your arm in FK mode using 2 bones. You can add additional bones and/or improve “feel”
of the interface by using coefficients for the change. I recommend you play with this example a lot before going to
next part.
You can clone the demo code for this chapter using
{{include(Inverse Kinematics)}}
TBD
Before continuing on, I’d recommend reading some theory, the simplest article I find is this:
http://freespace.virgin.net/hugo.elias/models/m_ik2.htm
Talking in Godot terminology, the task we want to solve here is position our 2 angles we talked about above so, that the
tip of lowerarm bone is as close to target point, which is set by target Vector3() as possible using only rotations. This
task is very calculation-intensive and never resolved by analytical equation solve. lso, it is underconstrained problem,
which means there is unlimited number of solutions to the equation.
For easy calculation, for this chapter we consider target is also child of Skeleton. If it is not the case for your setup
you can always reparent it in your script, as you will save on calculations if you do.
In the picture you see angles alpha and beta. In this case we don’t use poles and constraints, so we need to add our
own. On the picture the angles are 2D angles living in plane which is defined by bone base, bone tip and target.
The rotation axis is easily calculated using cross-product of bone vector and target vector. The rotation in this case
will be always in positive direction. If t is Transform which we get from get_bone_global_pose() function, the bone
vector is
t.basis[2]
4.1.11 Implementation
Now, we need to apply our transformations from IK bone to the base of chain. So we apply rotation to IK bone then
move from our IK bone upt to its parent, then apply rotation again, then move to the parent of current bone again, etc.
So we need to limit our chain somewhat.
export var IK_limit = 2
Executing this script will just pass through bone chain printing bone transforms.
extends Spatial
func _process(dt):
pass_chain(dt)
Now we need to actually work with target. The target should be placed somewhere accessible. Since “arm” is imported
scene, we better place target node within our top level scene. But for us to work with target easily its Transform should
be on the same level as Skeleton.
To cope with this problem we create “target” node under our scene root node and at script run we will reparent it
copying global transform, which will achieve wanted effect.
Create new Spatial node under root node and rename it to “target”. Then modify _ready() function to look like
this:
var skel
var target
func _ready():
skel = get_node("arm/Armature/Skeleton")
target = get_node("target")
var ttrans = target.get_global_transform()
remove_child(target)
skel.add_child(target)
target.set_global_transform(ttrans)
set_process(true)
4.2 Physics
4.3 Import
Introduction
Godot supports a flexible and powerful [[3D Scene importer]], that allows for full scene importing. For a lot of artists
and developers this is more than enough. However, many do not like this workflow as much and prefer to import
individual 3D Meshes and build the scenes inside the Godot 3D editor themselves. (Note that for more advanced
features such as skeletal animation, there is no option to the 3D Scene Importer).
The 3D mesh import workflow is simple and works using the OBJ file format. The imported meshes result in a .msh
binary file which the user can put into a [[API:MeshInstance]], which in turn can be placed somewhere in the edited
scene.
Importing
This dialog allows the import of one more more OBJ files into a target path. OBJ files are converted to .msh files. Files
are imported without any material on them, material has to be added by the user (see the [[Fixed materials]] tutorial).
If the external OBJ file is changed it will be re-imported, while keeping the newly assigned material.
Options
A few options are present. Normals is needed for regular shading, while Tangents is needed if you plan to use normal-
mapping on the material. In general, OBJ files describe how to be shaded very well, but an option to force smooth
shading is available.
Finally, there is an option to weld vertices. Given OBJ files are text-based, it is common to find some of these with
vertices that do not mach, which results in strange shading. The weld vertices option merges vertices that are too close
to keep proper smooth shading.
Usage
Mesh resources (what this importer imports) are used inside MeshInstance nodes. Simply set them to the Mesh
property of them.
Introduction
Most game engines just import 3D objects, which may contain skeletons or animations and then all further work is
done in the engine UI, like object placement, full scene animations, etc. In Godot, given the node system is very
similar to how 3D DCC (Such as Maya, 3DS Max or Blender) tools work, full 3D scenes can be imported in all their
glory. Additionally, by using a simple language tag system, it is possible to specify that objects are imported as several
things, such as collidable, rooms and portals, vehicles and wheels, LOD distances, billboards, etc.
This allows for some interesting features:
• Importing simple scenes, rigged objects, animations, etc.
• Importing full scenes. Entire scenarios can be created and updated in the 3D DCC and imported to Godot each
time they change, then only little editing is needed from the engine side.
• Full cutscenes can be imported, including multiple character animation, lighting, camera motion, etc.
• Scenes can be further edited and scripted in the engine, where shaders and environment effects can be added,
enemies can be instanced, etc. The importer will update geometry changes if the source scene changes but keep
the local changes too (in real-time while using the Godot editor!)
• Textures can be all batch-imported and updated when the source scene changes.
This is achieved by using a very simple language tag that will be explained in detail later.
Most game engines use the FBX format for importing 3D scenes, which is definitely one of the most standardized
in the industry. However, this format requires the use of a closed library from Autodesk which is distributed with a
more restrictive licensing terms than Godot. The plan is, sometime in the future, to implement an external conversion
binary, but meanwhile FBX is not really supported.
Autodesk added built-in collada support to Maya and 3DS Max, but It’s really broken and should not be used. The best
way to export this format is by using the OpenCollada plugins. They work really well, although they are not always
up-to date with the latest version of the software.
Blender also has built-in collada support, but It’s really broken and should not be used either.
Godot provides a Python Plugin that will do a much better job at exporting the scenes.
That opens what is probably the biggest of all the import dialogs:
p=. |image1|
Many options exist in there, so each section will be explained as follows:
To import, two options are needed. The first is a source .dae file (.dae stands for Collada. More import formats will
eventually added, but Collada is the most complete open format as of this writing).
A target folder needs to be provided, so the importer can import the scene there. The imported scene will have the
same filename as the source one, except for the .scn extension, so make sure you pick good names when you export!
The textures will be copied and converted. Textures in 3D applications are usually just PNG or JPG files. Godot
will convert them to video memory texture compression format (s3tc, pvrtc, ericsson, etc) by default to improve
performance and save resources.
Since the original textures, 3d file and textues are usually not needed, it’s recommended you keep them outside the
project. For some hints on how to do this the best way, you can check the [[Version control & Project organization]]
tutorial.
Two options for textures are provided. They can be copied to the same place as the scene, or they can be copied to a
common path (configurable in the project settings). If you choose this, make sure no two textures are names the same.
3D rigging tips
Before going into the options, here are some tips for making sure your rigs import properly
• Only up to 4 weights are imported per vertex, if a vertex depends of more than 4 bones, only the 4 most important
bones (the one with the most weight) will be imported. For most models this usually works fine, but just keep it
in mind.
• Do not use non-uniform scale in bone animation, as this will likely not import properly. Try to accomplish the
same effect with more bones.
• When exporting from Blender, make sure that objects modified by a skeleton are children of it. Many objects
can be modified by a single skeleton, but they all should be direct children.
• The same way, when using Blender, make sure that the relative transform of children nodes to the skeleton is
zero (no rotation, no translation, no scale. All zero and scale at 1.0). The position of both objects (the little
orange dot) should be at the same place.
3D import options
This section contains many options to change the way import workflow works. Some (like HDR) will be better
explained in other sections, but in general a pattern can be visible in the options and that is, many of the options end
with “-something”. For example:
• Remove Nodes (-noimp)
• Set Alpha in Materials (-alpha)
• Create Collisions (-col).
This means that the object names in the 3D DCC need to have those options appended at the end for the importer to
tell what they are. When imported, Godot will convert them to what they are meant to be.
Note: Maya users must use “_” (underscore) instead of “-” (minus).
Here is an example of how a scene in the 3D dcc looks (using blender), and how it is imported to Godot:
Notice that:
• The camera was imported normally.
• A Room was created (-room).
• A Portal was created (-portal).
• The Mesh got static collision added (-col).
• The Light was not imported (-noimp).
Options in detail
Following is a list of most import options and what they do in more detail.
Remove nodes (-noimp) Node names that have this at the end will be removed at import time, mo matter their type.
Erasing them afterwards is most of the times pointless because the will be restored if the source scene changes.
Import animations Some scene formats (.dae) support one or more animations. If this is checked, an Animation-
Player node will be created, containing the animations.
Compress geometry This option (disabled [STRIKEOUT:or more like, always enabled] at the moment at the time
of writing this) will compress geometry so it takes less space and renders faster (at the cost of less precision).
Force generation of tangent arrays The importer detects when you have used a normalmap texture, or when the
source file contains tangent/binormal information. These arrays are needed for normalmapping to work, and most
exporters know what they do when they export this. However, it might be possible to run into source scenes that do
not have this information which, as a result, make normal-mapping not work. If you notice that normal-maps do not
work when importing the scene, turn this on!
SRGB -> linear of diffuse textures When rendering using HDR (High Dynamic Range) it might be desirable to
use linear-space textures to achieve a more real-life lighting. Otherwise, colors may saturate and contrast too much
when exposure changes. This option must be used together with the SRGB option in WorldEnvironment. The texture
import options also have the option to do this conversion, but if this one is turned on, conversion will always be done
to diffuse textures (usually what is desired). For more information, read the [[HDR]].
Set alpha in materials (-alpha) When working with most 3D dccs, its pretty obvious when a texture is transpar-
ent and has opacity and this rarely affects the workflow or final rendering. However, when dealing with real-time
rendering, materials with alpha blending are usually less optimal to draw, so they must be explicitly marked as such.
Originally Godot detected this based on whether if the source texture had an alpha channel, but most image manipu-
lation apps like Photoshop or Gimp will export this channel anyway even if not used. Code was added later to check
manually if there really was any transparency in the texture, but artists will anyway and very often lay uvmaps into
opaque parts of a texture and leave unused areas (where no UV exists) transparent, making this detection worthless.
Finally, it was decided that it’s best to import everything as opaque and leave artists to fix materials that need trans-
parency when it’s obvious that they are not looking right (see the [Fixed Ma
As a helper, since every 3D dcc allows naming the materials and keeping their name upon export, the (-alpha) modifier
in their name will hint the 3D scene importer in Godot that this material will use the alpha channel for transparency.
Set vert. color in materials (-vcol) Most 3D DCCs support vertex color painting. This is generally applied as
multiplication or screen blending. However, it is also often the case that your exporter will export this information as
all 1s, or export it as something else and you will not realize it. Since most of the cases this option is not desired, just
add this to any material to confirm that vertex colors are desired.
Create collisions (-col, -colonly) These will only work for Mesh nodes, If the “-col” option is detected, a child static
collision node will be added, using the same geometry as the mesh.
However, it is often the case that the visual geometry is too complex or too un-smooth for collisions, which end up
not working well. To solve this, the “-colonly” modifier exists, which will remove the mesh upon import and create a
StaticBody collision instead. This helps the visual mesh and actual collision to be separated.
Create rooms (-room) This is used to create a room. As a general rule, any node that is a child of this node will be
considered inside the room (including portals). For more information about rooms/portals, look at the [[Portals and
Rooms]] tutorial.
There are two ways in which this modifier can be used. The first is using a Dummy/Empty node in the 3D app with
the “-room” tag. For this to work, the “interior” of the room must be closed (geometry of the childrens should contain
walls, roof, floor, etc and the only holes to the outside should be covered with portals). The importer will then create
a simplified version of the geometry for the room.
The second way is to use the “-room” modifier on a mesh node. This will use the mesh as the base for the BSP tree that
contains the room bounds. Make sure that the mesh shape is closed, all normals point outside and that the geometry
is not self-intersecting, otherwise the bounds may be computed wrong (BSP Trees are too picky and difficult to work
with, which is why they are barely used anymore..).
Anyway, the room will need portals, which are described next.
Create portals (-portal) Portals are the view to look outside a room. They are always some flat shape on the surface
of a room. If the portal is left alone, it is used to activate occlusion when looking inside<->outside the room. Again,
more information on the [[Portals and Rooms]] tutorial.
Basically, the conditions to make and import a portal from the 3D DCC are:
• It should be a child of a room.
• It should lay on the surface of the room (this doesn’t need to be super exact, just make it as close as you can by
eye and Godot will adjust it)
• It must be a flat, convex shape, any flat and convex shape is ok, no matter the axis or size.
• Normals for the flat shape faces must all point towards the OUTSIDE of the room.
To connect to rooms, simply make two identical portals for both rooms and place them overlapped. This does not need
to be perfectly exact, again, as Godot will fix it.
[..] The rest of the tags in this section should be rather obvious, or will be documented/changed in the future.
Double-sidedness
Collada and other formats support specifying the double-sidedness of the geometry (in other words, when not double-
sided, back-faces are not drawn). Godot supports this option per Material, not per Geometry.
When exporting from 3D DCCs that work with per-object double-sidedness (such as Blender of Maya), make sure that
the double sided objects do not share a material with the single sided ones or the importer will not be able to discern.
Animation options
Some things to keep in mind when importing animations. 3D DCCs allow animating with curves for every x,y,z
component, doing IK constraints and other stuff. When imported for real-time, animations are sampled (at small
intervals) so all this information is lost. Sampled animations are fast to process, but can use considerable amounts of
memory.
Because of this, the “Optimize” option exists but, in some cases, this option might get to break an animation, so make
it sure to disable if you see this.
Some animations are meant to be cycled (like walk animations) if this is the case, animation names that end in “-cycle”
or “-loop” are automatically set to loop.
Import script
Creating a script to parse the imported scene is actually really simple. This is great for post processing, changing
materials, doing funny stuff with the geometry, etc.
Create a script that basically looks like this:
tool #needed so it runs in editor
extends EditorScenePostImport
func post_import(scene):
#do your stuff here
pass # scene contains the imported scene starting from the root node
The post-import function takes the imported scene as parameter (the parameter is actually the root node of the scene).
Update logic
Other types of resources (like samples, meshes, fonts, images, etc.) are re-imported entirely when changed and user
changes are not kept.
Because of 3D Scenes can be really complex, they use a different update strategy. The user might have done local
changes to take advantage of the engine features and it would be really frustrating if everything is lost on re-import
because the source asset changed.
This led to the implementation of a special update strategy. The idea behind is that the user will not lose anything he
or she did, and only added data or data that can’t be edited inside Godot will be updated.
It works like this:
Strategy Upon changes on the source asset (ie: .dae), and on re-import, the editor will remember the way the scene
originally was, and will track your local changes like renaming nodes, moving them or reparenting them. Finally, the
following will be updated:
• Mesh Data will be replaced by the data from the updated scene.
• Materials will be kept if they were not modified by the user.
• Portal and Room shapes will be replaced by the ones from the updated scene.
• If the user moved a node inside Godot, the transform will be kept. If the user moved a node in the source asset,
the transform will be replaced. Finally, if the node was moved in both places, the transform will be combined.
In general, if the user deletes anything from the imported scene (node, mesh, material, etc), updating the source asset
will restore what was deleted. This is a good way to revert local changes to anything. If you really don’t want a node
anymore in the scene, either delete it from both places or add the “-noimp” tag to it in the source asset.
Fresh re-import It can also happen that the source asset changed beyond recognition and a full fresh re-import is
desired. If so, simply re-open the 3d scene import dialog from the Import -> Re-Import menu and perform re-import.
Networking
5.1.1 Introduction
It is often desired to use SSL connections for communications to avoid “man in the middle” attacks. Godot has a
connection wrapper, StreamPeerSSL, which can take a regular connection and add security around it. The HTTPClient
class also supports HTTPS by using this same wrapper.
For SSL to work, certificates need to be provided. A .crt file must be specified in the project settings:
225
Godot Engine Documentation, Release latest
The first approach is the simplest, just generate a private and public key pair, and put the public pair in the .crt file
(again, in PEM format). The private key should go to your server.
OpenSSL has some documentation about this. This approach also does not require domain validation nor requires
you to spend a considerable amount of money in purchasing certificates from a CA.
The second approach consists of using a certificate authority (CA) such as Verisign, Geotrust, etc. This is a more
cumbersome process, but it’s more “official” and ensures your identity is clearly represented.
Unless you are working with large companies or corporations, or need to connect to someone else’s servers (ie,
connecting to Google or some other REST API provider via HTTPS) this method is not as useful.
Also, when using a CA issued cert, you must enable domain validation, to ensure the domain you are connecting to
is the one intended, otherwise any website can issue any certificate in the same CA and it will work.
If you are using Linux, you can use the supplied certs file, generally located in:
/etc/ssl/certs/ca-certificates.crt
This file allows HTTPS connections to virtually any website (ie, Google, Microsoft, etc) .
Or just pick any of the more specific certificates there if you are connecting to a specific one.
Here’s an example of using the HTTPClient class. It’s just a script, so it can be run by executing:
c
c:\\godot> godot -s http_test.gd
# HTTPClient demo
# This simple class can do HTTP requests, it will not block but it needs to be polled
func _init():
var err=0
var http = HTTPClient.new() # Create the Client
# Some headers
var headers=[
"User-Agent: Pirulo/1.0 (Godot)",
"Accept: */*"
]
if (http.has_response()):
#If there is a response..
if (http.is_response_chunked()):
#Does it use chunks?
print("Respose is Chunked!")
else:
#Or just plain Content-Length
var bl = http.get_response_body_length()
print("Response Length: ",bl)
while(http.get_status()==HTTPClient.STATUS_BODY):
#While there is body left to be read
http.poll()
var chunk = http.read_response_body_chunk() # Get a chunk
if (chunk.size()==0):
#got nothing, wait for buffers to fill a bit
OS.delay_usec(1000)
else:
rb = rb + chunk # append to read bufer
#done!
quit()
Editor plugins
Coming soon™.
231
Godot Engine Documentation, Release latest
Miscellaneous
7.1 Math
Introduction
This small tutorial aims to be a short and practical introduction to vector math, useful for 3D but also 2D games.
Again, vector math is not only useful for 3D but also 2D games. It is an amazing tool once you get the grasp of it and
makes programming of complex behaviors much simpler.
It often happens that young programmers rely too much on the incorrect math for solving a wide array of problems,
for example using only trigonometry instead of vector of math for 2D games.
This tutorial will focus on practical usage, with immediate application to the art of game programming.
Typically, we define coordinates as an (x,y) pair, x representing the horizontal offset and y the vertical one. This makes
sense given the screen is just a rectangle in two dimensions. As an example, here is a position in 2D space:
233
Godot Engine Documentation, Release latest
A position can be anywhere in space. The position (0,0) has a name, it’s called the origin. Remember this term well
because it has more implicit uses later. The (0,0) of a n-dimensions coordinate system is the origin.
In vector math, coordinates have two different uses, both equally important. They are used to represent a position but
also a vector. The same position as before, when imagined as a vector, has a different meaning.
When imagined as a vector, two properties can be inferred, the direction and the magnitude. Every position in space
can be a vector, with the exception of the origin. This is because coordinates (0,0) can’t represent direction (magnitude
0).
Direction
Direction is simply towards where the vector points to. Imagine an arrow that starts at the origin and goes towards a
[STRIKEOUT:position]. The tip of the arrow is in the position, so it always points outwards, away from the origin.
Imagining vectors as arrows helps a lot.
Magnitude
Finally, the length of the vector is the distance from the origin to the position. Obtaining the length from a vector is
easy, just use the Pithagorean Theorem.
var len = sqrt( x*x + y*y )
But.. Angles?
But why not using an angle? After all, we could also think of a vector as an angle and a magnitude, instead of a
direction and a magnitude. Angles also are a more familiar concept.
To say truth, angles are not that useful in vector math, and most of the time they are not dealt with directly. Maybe
they work in 2D, but in 3D a lot of what can usually be done with angles does not work anymore.
Still, using angles is still not an excuse, even for 2D. Most of what takes a lot of work with angles in 2D, is still much
more natural easier to accomplish with vector math. In vector math, angles are useful only as measure, but take little
part in the math. So, give up the trigonometry already, prepare to embrace vectors!
In any case, obtaining an angle from a vector is easy and can be accomplished with trig.. er what was that? I mean,
the atan2(x,y) function.
Vectors in Godot
To make examples easier, it is worth explaining how vectors are implemented in GDScript. GDscript has both Vector2
and Vector3, for 2D and 3D math respectively. Godot uses Vector classes as both position and direction. They also
contain x and y (for 2D) and x, y and z (for 3D) member variables.
h1. create a vector with coordinates (2,5)
var a = Vector2(2,5)
h1. create a vector and assign x and y manually
var b = Vector2()
b.x=7
b.y=8
When operating with vectors, it is not necessary to operate on the members directly (in fact this is much slower).
Vectors support regular arithmetic operations:
#add a and b
var c = a+b
h1. will result in c vector, with value (9,13)
Regular arithmetic operations such as addition, subtraction, multiplication and division are supported.
Vector multiplication and division can also be mixed with single-digit numbers, also named scalars.
Perpendicular Vectors
Rotating a 2D vector 90° degrees to either side, left or right, is really easy, just swap x and y, then
negate either x or y (direction of rotation depends on which is negated).
Example:
var v = Vector2(0,1)
#rotate right (clockwise)
var v_right = Vector2(-v.y,v.x)
#rotate left (counter-clockwise)
var v_right = Vector2(v.y,-v.x)
This is a handy trick that is often of use. It is impossible to do with 3D vectors, because there are an infinite amount
of perpendicular vectors.
Unit Vectors
Ok, so we know what a vector is. It has a direction and a magnitude. We also know how to use them in Godot. The
next step is learning about unit vectors. Any vector with magnitude of length 1 is considered a unit vector. In 2D,
imagine drawing a circle of radius one. That circle contains all unit vectors in existence for 2 dimensions:
So, what is so special about unit vectors? Unit vectors are amazing. In other words, unit vectors have several, very
useful properties.
Can’t wait to know more about the fantastic properties of unit vectors, but one step at a time. So, how is a unit vector
created from a regular vector?
Normalization
Taking any vector and reducing it’s magnitude to 1.0 while keeping it’s direction is called normalization. Normal-
ization is performed by dividing the x and y (and z in 3D) components of a vector by it’s magnitude:
var a = Vector2(2,4)
var m = sqrt( a.x*a.x + a.y*a.y )
a.x/=m
a.y/=m
As you might have guessed, if the vector has magnitude 0 (meaning, it’s not a vector but the origin also called null
vector), a division by zero occurs and the universe goes through a second big bang, except in reverse polarity and then
back. As a result, humanity is safe but Godot will print an error. Remember! Vector(0,0) can’t be normalized!.
Of course, Vector2 and Vector3 already provide a method to do this:
a = a.normalized()
Dot Product
OK, the dot product is the most important part of vector math. Without the dot product, Quake would have never
been made. This is the most important section of the tutorial, so make sure to grasp it properly. Most people trying
to understand vector math give up here because, despite how simple it is, they can’t make head or tails from it. Why?
Here’s why, it’s because..
The dot product takes two vectors and returns a scalar:
var s = a.x*b.x + a.y*b.y
Yes, pretty much that. Multiply x from vector a by x from vector b. Do the same with y and add it together. In 3D it’s
pretty much the same:
var s = a.x*b.x + a.y*b.y + a.z*b.z
I know, it’s totally meaningless! you can even do it with a built-in function:
var s = a.dot(b)
The order of two vectors does not matter, ‘a.dot(b)‘ returns the same value as ‘b.dot(a)‘.
This is where despair begins and books and tutorials show you this formula:
And you realize it’s time to give up making 3D games or complex 2D games. How can something so simple be so
complex? Someone else will have to make the next Zelda or Call of Duty. Top down RPGs don’t look so bad after all.
Yeah I hear someone did pretty will with one of those on Steam...
So this is your moment, this is your time to shine. DO NOT GIVE UP! At this point, this tutorial will take a sharp
turn and focus on what makes the dot product useful. This is, why it is useful. We will focus one by one in the use
cases for the dot product, with real-life applications. No more formulas that don’t make any sense. Formulas will
make sense once you learn why do they exist for.
Siding
The first useful and most important property of the dot product is to check what side stuff is looking at. Let’s imagine
we have any two vectors, a and b. Any direction or magnitude (neither origin). Does not matter what they are, but
let’s imagine we compute the dot product between them.
var s = a.dot(b)
The operation will return a single floating point number (but since we are in vector world, we call them scalar, will
keep using that term from now on). This number will tell us the following:
• If the number is greater than zero, both are looking towards the same direction (the angle between them is < 90°
degrees).
• If the number is less than zero, both are looking towards opposite direction (the angle between them is > 90°
degrees).
• If the number is zero, vectors are shaped in L (the angle between them is 90° degrees).
So let’s think of a real use-case scenario. Imagine Snake is going through a forest, and then there is an enemy nearby.
How can we quickly tell if the enemy has seen discovered Snake? In order to discover him, the enemy must be able to
see Snake. Let’s say, then that:
• Snake is in position A.
• The enemy is in position B.
• The enemy is facing towards direction vector F.
So, let’s create a new vector BA that goes from the guard (B) to Snake (A), by subtracting the two:
var BA = A-B
Ideally, if the guard was looking straight towards snake, to make eye to eye contact, it would do it in the same direction
as vector BA.
If the dot product between F and BA is greater than 0, then Snake will be discovered. This happens because we will
be able to tell that the guard is facing towards him:
if ( BA.dot(F) > 0 ):
print("!")
Ok, so now we know that dot product between two vectors will let us know if they are looking towards the same side,
opposite sides or are just perpendicular to each other.
This works the same with all vectors, no matter the magnitude so unit vectors are not the exception. However, using
the same property with unit vectors yields an even more interesting result, as an extra property is added:
• If both vectors are facing towards the exact same direction (parallel to each other, angle between them is 0°),
the resulting scalar is 1.
• If both vectors are facing towards the exact opposite direction (parallel to each other, but angle between them is
180°), the resulting scalar is -1.
This means that dot product between unit vectors is always between the range of 1 and -1. So Again..
• If their angle is 0° dot product is 1.
• If their angle is 90°, then dot product is 0.
• If their angle is 180°, then dot product is -1.
Uh.. this is oddly familiar.. seen this before.. where?
Let’s take two unit vectors. The first one is pointing up, the second too but we will rotate it all the way from up (0°) to
down (180° degrees)..
What is this useful for? Well obtaining the angle directly is probably not as useful, but just being able to tell the angle
is useful for reference. One example is in the Kinematic Character demo, when the character moves in a certain
direction then we hit an object. How to tell if what we hit is the floor?
By comparing the normal of the collision point with a previously computed angle.
The beauty of this is that the same code works exactly the same and without modification in 3D. Vector math is, in a
great deal, dimemsion-amount-independent, so adding or removing an axis only adds very little complexity.
Planes
The dot product has another interesting property with unit vectors. Imagine that perpendicular to that vector (and
through the origin) passes a [STRIKEOUT:plane]. Planes divide the entire space into positive (over the plane) and
negative (under the plane), and (contrary to popular belief) you can also use their math in 2D:
Unit vectors that are perpendicular to a surface (so, they describe the orientation of the surface) are called unit normal
vectors. Though, usually they are just abbreviated as *normalsÄ. Normals appear in planes, 3D geometry (to deter-
mine where each face or vertex is siding), etc. A normal is a unit vector, but it’s called normal because of it’s usage.
(Just like we call Origin to (0,0)!).
It’s as simple as it looks. The plane passes by the origin and the surface of it is perpendicular to the unit vector (or
normal). The side towards the vector points to is the positive half-space, while the other side is the negative half-space.
In 3D this is exactly the same, except that the plane is an infinite surface (imagine an infinite, flat sheet of paper that
you can orient and is pinned to the origin) instead of a line.
Distance to Plane
Now that it’s clear what a plane is, let’s go back to the dot product. The dot product between a unit vector and any
point in space (yes, this time we do dot product between vector and position), returns the distance from the point to
the plane:
var distance = normal.dot(point)
But not just the absolute distance, if the point is in the negative half space the distance will be negative, too:
For 3D math, Godot provides a Plane built-in type that handles this.
Basically, N and D can represent any plane in space, be it for 2D or 3D (depending on the amount of dimensions of N)
and the math is the same for both. It’s the same as before, but D id the distance from the origin to the plane, travelling
in N direction. As an example, imagine you want to reach a point in the plane, you will just do:
This will stretch (resize) the normal vector and make it touch the plane. This math might seem confusing, but it’s
actually much simpler than it seems. If we want to tell, again, the distance from the point to the plane, we do the same
but adjusing for distance:
var distance = N.dot(point) - D
Constructing a Plane in 2D
Planes clearly don’t come out of nowhere, so they must be built. Constructing them in 2D is easy, this can be done
from either a normal (unit vector) and a point, or from two points in space.
In the case of a normal and a point, most of the work is done, as the normal is already computed, so just calculate D
from the dot product of the normal and the point.
var N = normal
var D = normal.dot(point)
For two points in space, there are actually two planes that pass through them, sharing the same space but with normal
pointing to the opposite directions. To compute the normal from the two points, the direction vector must be obtained
first, and then it needs to be rotated 90° degrees to either side:
#calculate vector from a to b
var dvec = (point_b - point_a).normalized()
#rotate 90 degrees
var normal = Vector2(dvec.y,-dev.x)
#or alternatively
# var normal = Vector2(-dvec.y,dev.x)
# depending the desired side of the normal
The rest is the same as the previous example, either point_a or point_b will work since they are in the same plane:
var N = normal
var D = normal.dot(point_a)
# this works the same
# var D = normal.dot(point_b)
Doing the same in 3D is a little more complex and will be explained further down.
Here is a simple example of what planes are useful for. Imagine you have a convex polygon. For example, a rectangle,
a trapezoid, a triangle, or just any polygon where faces that don’t bend inwards.
For every segment of the polygon, we compute the plane that passes by that segment. Once we have the list of planes,
we can do neat things, for example checking if a point is inside the polygon.
We go through all planes, if we can find a plane where the distance to the point is positive, then the point is outside the
polygon. If we can’t, then the point is inside.
Pretty cool, huh? But this gets much better! With a little more effort, similar logic will let us know when two convex
polygons are overlapping too. This is called the Separating Axis Theorem (or SAT) and most physics engines use this
to detect collision.
The idea is really simple! With a point, just checking if a plane returns a positive distance is enough to tell if the
point is outside. With another polygon, we must find a plane where all the **other* polygon points* return a positive
distance to it. This check is performed with the planes of A against the points of B, and then with the planes of B
against the points of A:
for p in planes_of_A:
var all_out = true
for v in points_of_B:
if ( p.distance_to(v) < 0):
all_out=false
break
if (all_out):
# a separating plane was found
# do not continue testing
overlapping=false
break
if (overlapping):
#only do this check if no separating plane
#was found in planes of A
for p in planes_of_B:
var all_out = true
for v in points_of_A:
if ( p.distance_to(v) < 0):
all_out=false
break
if (all_out):
overlapping=false
break
if (overlapping):
print("Polygons Collided!")
As you can see, planes are quite useful, and this is the tip of the iceberg. You might be wondering what happens with
non convex polygons. This is usually just handled by splitting the concave polygon into smaller convex polygons, or
using a technique such as BSP (which is not used much nowadays).
Cross Product
Quite a lot can be done with the dot product! But the party would not be complete without the cross product. Remem-
ber back at the beginning of this tutorial? Specifically how to obtain a perpendicular (rotated 90 degrees) vector by
swapping x and y, then negating either of them for right (clockwise) or left (counter-clockwise) rotation? That ended
up being useful for calculating a 2D plane normal from two points.
As mentioned before, no such thing exists in 3D because a 3D vector has infinite perpendicular vectors. It would also
not make sense to obtain a 3D plane from 2 points, as 3 points are needed instead.
To aid in this kind stuff, the brightest minds of humanity’s top mathematicians brought us the cross product.
The cross product takes two vectors and returns another vector. The returned third vector is always perpendicular
to the first two. The source vectors, of course, must not be the same, and must not be parallel or opposite, else the
resulting vector will be (0,0,0):
var c = a.cross(b)
However, unlike the dot product, doing ‘a.cross(b)‘ and ‘b.cross(a)‘ will yield different results. Specifically, the re-
turned vector will be negated in the second case. As you might have realized, this coincides with creating perpendicular
vectors in 2D. In 3D, there are also two possible perpendicular vectors to a pair of 2D vectors.
Also, the resulting cross product of two unit vectors is not a unit vector. Result will need to be renormalized.
Area of a Triangle
Cross product can be used to obtain the surface area of a triangle in 3D. Given a triangle consisting of 3 points, A, B
and C:
Take any of them as a pivot and compute the adjacent vectors to the other two points. As example, we will use B as a
pivot:
var BA = A-B
var BC = C-B
Compute the cross product between BA and BC to obtain the perpendicular vector P:
var P = BA.cross(BC)
The length (magnitude) of P is the surface area of the parallelogram built by the two vectors BA and BC, therefore the
surface area of the triangle is half of it.
var area = P.length()/2
With P computed from the previous step, normalize it to get the normal of the plane.
var N = P.normalized()
And obtain the distance by doing the dot product of P with any of the 3 points of the ABC triangle:
var D = P.dot(A)
img/tutovec20.png
Normals of triangles often are sided towards the direction they can be viewed from, so in this case, the normal of
triangle ABC would point towards the front camera:
img/tutovec21.png
Collision Detection in 3D
This is another bonus bit, a reward for being patient and keeping up with this long tutorial. Here is another piece
of wisdom. This maybe is not something with a direct use case (Godot already does collision detection pretty well)
but It’s a really cool algorithm to understand anyway, because it’s used by almost all physics engines and collision
detection libraries :)
Remember that converting a convex shape in 2D to an array of 2D planes was useful for collision detection? You could
detect if a point was inside any convex shape, or if two 2D convex shapes were overlapping.
Well, this works in 3D too, if two 3D polyhedral shapes are colliding, you won’t be able to find a separating plane. If
a separating plane is found, then the shapes are definitely not colliding.
To refresh a bit a separating plane means that all vertices of polygon A are in one side of the plane, and all vertices of
polygon B are in the other side. This plane is always one of the face-planes of either polygon A or polygon B.
In 3D though, there is a problem to this approach, because it is possible that, in some cases a separating plane can’t be
found. This is an example of such situation:
img/tutovec22.png
To avoid it, some extra planes need to be tested as separators, these planes are the cross product between the edges of
polygon A and the edges of polygon B
img/tutovec23.png
for p in planes_of_A:
var all_out = true
for v in points_of_B:
if ( p.distance_to(v) < 0):
all_out=false
break
if (all_out):
# a separating plane was found
# do not continue testing
overlapping=false
break
if (overlapping):
#only do this check if no separating plane
#was found in planes of A
for p in planes_of_B:
var all_out = true
for v in points_of_A:
if (all_out):
overlapping=false
break
if (overlapping):
for ea in edges_of_A:
for eb in edges_of_B:
var n = ea.cross(eb)
if (n.length()==0):
continue
var max_A=-1e20 # tiny number
var min_A=1e20 # huge number
for v in points_of_A:
var d = n.dot(v)
if (d>max_A):
max_A=d
if (dmax_B):
max_B=d
if (dmax_B or min_B>max_A):
# not overlapping!
overlapping=false
break
if (not overlapping):
break
if (overlapping):
print("Polygons Collided!")
This was all! Hope it was helpful, and please give feedback and let know if something in this tutorial is not clear! You
should be now ready for the next challenge.. [[Transforms]]!
Introduction
Before reading this tutorial, it is advised to read the previous one about [[Vector Math]] as this one is a direct continu-
ation.
This tutorial will be about transformations and will cover a little about matrices (but not in-depth).
Transformations are most of the time applied as translation, rotation and scale so they will be considered as priority
here.
Imagine we have a spaceship somewhere in space. In Godot this is easy, just move the ship somewhere and rotate it:
Ok, so in 2D this looks simple, a position and an angle for a rotation. But remember, we are grown ups here and don’t
use angles (plus, angles are not really even that useful when working in 3D).
We should realize that at some point, someone designed this spaceship. Be it for 2D in a drawing such as Paint.net,
Gimp, Photoshop, etc. or in 3D through a 3D DCC tool such as Blender, Max, Maya, etc.
When it was designed, it was not rotated. It was designed in it’s own coordinate system.
This means that the tip of the ship has a coordinate, the fin has another, etc. Be it in pixels (2D) or vertices (3D).
So, let’s recall again that the ship was somewhere in space:
How did it get there? What moved it and rotated it from the place it was designed to it’s current position? The answer
is... a transform, the ship was transformed from their original position to the new one. This allows the ship to be
displayed where it is.
So, a transform is too generic of a term. To solve this puzzle, we will overimpose the ship’s original design position at
So, we can see that the “design space” has been transformed too. How can we best represent this transformation? Let’s
use 3 vectors for this (in 2D), a unit vector pointing towards X positive, a unit vector pointing towards Y positive and
a translation.
Let’s call the 3 vectors “X”, “Y” and “Origin”, and let’s also overimpose them over the ship so it makes more sense:
Ok, this is nicer, but it still does not make sense. What do X,Y and Origin have to do with how the ship got there?
Well, let’s take the point from top tip of the ship as reference:
And let’s apply the following operation to it (and to all the points in the ship too, but we’ll track the top tip as our
reference point):
var new_pos = pos - origin
Doing this to the selected point will move it back to the center:
This was expected, but then let’s do something more interesting. Use the dot product of X and the point, and add it to
the dot product of Y and the point:
var final_pos = x.dot(new_pos) + y.dot(new_pos)
Then what we have is.. wait a minute, it’s the ship in it’s design position!
How did this black magic happen? The ship was lost in space, and now it’s back home!
It might seem strange, but it does have plenty of logic. Remember, as we have seen in the
[[tutorial_vector_math#distance-to-plane]], what happened is that the distance to X axis, and the distance to Y axis
were computed. Calculating distance in a direction or plane was one of the uses for the dot product. This was enough
to obtain back the design coordinates for every point in the ship.
So, what he have been working with so far (with X, Y and Origin) is an Oriented Coordinate System*. X an Y are the
**Basis*, and *Origin* is the offset.
Basis
The Origin we know what it is. It’s where the 0.0 (origin) of the design coordinate system ended up after being
transformed to a new position. This is why it’s called Origin, But in practice, it’s just an offset to the new position.
The Basis is more interesting. The basis is the X and Y of the new, transformed, OCS are pointing towards. It’s telling
what is in change of drawing 2D and 3D “Hey, the original X and Y axes or your design are right here, pointing
towards these directions”.
So, let’s change the representation of the basis. Instead of 2 vectors, let’s use a matrix.
The vectors are up there in the matrix, horizontally. The next problem now is that.. what is this matrix thing? Well,
we’ll assume you’ve never heard of a matrix.
Transforms in Godot
This tutorial will not explain matrix math (and their operations) in depth, only it’s practical use. There is plenty of
material for that, which should be a lot simpler to understand after completing this tutorial. We’ll just explain how to
use transforms.
Matrix32
Matrix32 is a 3x2 matrix. It has 3 Vector2 elements and it’s used for 2D. The “X” axis is the element 0, “Y” axis is
the element 1 and “Origin” is element 2. It’s not divided in basis/origin for convenience, due to it’s simplicity.
var m = Matrix32()
var x = m[0] # 'X'
var y = m[1] # 'Y'
var o = m[2] # 'Origin'
Most operations will be explained with this datatype (Matrix32), but the same logic applies to 3D.
Identity
It’s easy to guess that an identity matrix is just a matrix that aligns the transform to it’s parent coordinate system. It’s
an OCS that hasn’t been translated, rotated or scaled. All transform types in Godot are created with identity.
Operations
Rotation
Translation
There are two ways to translate a Matrix32, the first one is just moving the origin:
# Move 2 units to the right
var m = Matrix32()
Scale
A matrix can be scaled too. Scaling will multiply the basis vectors by a vetor (X vector by x component of the scale,
Y vector by y component of the scale). It will leave the origin alone:
These kind of operations in matrices are accumulative. It means every one starts relative to the previous one. For those
that have been living on this planet long enough, a good reference of how transform works is this:
A matrix is used similarly to a turtle. The turtle most likely had a matrix inside (and you are likely learning this may
years after discovering Santa is not real).
Transform
Transform is the act of switching between coordinate systems. To convert a position (either 2D or 3D) from “designer”
coordinate system to the OCS, the “xform” method is used.
var new_pos = m.xform(pos)
Inverse Transform
To do the opposite operation (what we did up there with the rocket), the “xform_inv” method is used:
var new_pos = m.xform_inv(pos)
Or pre-multiplication:
var new_pos = pos * m
Orthonormal Matrices
However, if the Matrix has been scaled (vectors are not unit length), or the basis vectors are not orthogonal (90°), the
inverse transform will not work.
In other words, inverse transform is only valid in orthonormal matrices. For this, these cases an affine inverse must be
computed.
The transform, or inverse transform of an identity matrix will return the position unchanged:
# Does nothing, pos is unchanged
pos = Matrix32().xform(pos)
Affine Inverse
The affine inverse is a matrix that does the inverse operation of another matrix, no matter if the matrix has scale or the
axis vectors are not orthogonal. The affine inverse is calculated with the affine_inverse() method:
var mi = m.affine_inverse()
var pos = m.xform(pos)
pos = mi.xform(pos)
#pos is unchanged
Matrix Multiplication
Matrices can be multiplied. Multiplication of two matrices “chains” (concatenates) their transforms.
However, as per convention, multiplication takes place in reverse order.
Example:
var m = more_transforms * some_transforms
Multiplication by Inverse
Multiplication by Identity
Matrix tips
When using a transform hierarchy, remember that matrix multiplication is reversed! To obtain the global transform for
a hierarchy, do:
var global_xform = parent_matrix * child_matrix
For 3 levels:
# due to reverse order, parenthesis are needed
var global_xform = gradparent_matrix + (parent_matrix + child_matrix)
To make a matrix relative to the parent, use the affine inverse (or regular inverse for orthonormal matrices).
# transform B from a global matrix to one local to A
var B_local_to_A = A.affine_inverse() * B
OK, hopefully this should be enough! Let’s complete the tutorial by moving to 3D matrices
As mentioned before, for 3D, we deal with 3 Vector3 vectors for the rotation matrix, and an extra one for the origin.
Matrix3
Godot has a special type for a 3x3 matrix, named Matrix3. It can be used to represent a 3D rotation and scale. Sub
vectors can be accessed as:
var m = Matrix3()
var x = m[0] h1. Vector3
var y = m[1] h1. Vector3
var z = m[2] h1. Vector3
Rotation in 3D
Rotation in 3D is more complex than in 2D (translation and scale are the same), because rotation is an implicit 2D
operation. To rotate in 3D, an axis, must be picked. Rotation, then, happens around this axis.
The axis for the rotation must be a normal vector. As in, a vector that can point to any direction, but length must be
one (1.0).
#rotate in Y axis
var m3 = Matrix3()
m3 = m3.rotated( Vector3(0,1,0), PI/2 )
Transform
To add the final component to the mix, Godot provides the Transform type. Transform has two members:
• basis (of type Matrix3
• origin (of type Vector3
Any 3D transform can be represented with Transform, and the separation of basis and origin makes it easier to work
translation and rotation separately.
An example:
var t = Transform()
pos = t.xform(pos) #transform 3D position
pos = t.basis.xform(pos) h1. (only rotate)
pos = t.origin + pos (only translate)
7.2 Shaders
Introduction
This tutorial will help you to use Godot shaders to deform a plane mesh so it appears like a basic terrain. Remember
that this solution has pros and cons.
Pros:
The heightmap
We will use a few functions of GIMP image editor to produce a simple heightmap. Start GIMP and create a square
image of 512x512 pixels.
|image0|
You are now in front of a new, blank, square image.
|image1|
Then, use a filter to render some clouds on this new image.
|image2|
Parameter this filter to whatever you want. A white pixel corresponds to the highest point of the heightmap, a black
pixel corresponds to the lowest one. So, darker regions are valleys and brighter are mountains. If you want, you can
check “tileable” to render a heightmap that can be cloned and tiled close together with another one. X and Y size
don’t matter a lot as long as they are big enough to provide a decent ground. A value of 4.0 or 5.0 for both is nice.
Click on the “New Seed” button to roll a dice and GIMP will create a new random heightmap. Once you are happy
with the result, click “OK”.
|image3|
You can continue to edit your image if you wish. For our example, let’s keep the heightmap as is, and let’s export it to
a PNG file, say “heightmap.png”. Save it in your Godot project folder.
Now, we will need a plane mesh to import in Godot. Let’s run Blender.
|image4|
Remove the start cube mesh, then add a new plane to the scene.
|image5|
Zoom a bit, then switch to Edit mode (Tab key) and in the Tools buttongroup at the left, hit “Subdivide” 5 or 6 times.
|image6|
Your mesh is now subdivided, which means we added vertices to the plane mesh that we will later be able to move.
Job’s not finished yet: in order to texture this mesh a proper UV map is necessary. Currently, the default UV map
contains only the 4 corner vertices we had at the beginning. However, we now have more, and we want to be able to
texture over the whole mesh correctly.
If all the vertices of your mesh are not selected, select them all (hit “A”). They must appear orange, not black. Then,
in the Shading/UVs button group a the left, click the “Unwrap” button (or simply hit “U”) and select “Smart UV
Project”. Keep the default options and hit “Ok”.
|image7|
Now, we need to switch our view to “UV/Image editor”.
|image8|
Select all the vertices again (“A”) then in the UV button, select “Export UV Layout”.
|image9|
Export the layout as a PNG file. Name it “plane.png” and save it in your Godot project folder. Now, let’s export our
mesh as an OBJ file. Top of the screen, click “File/Export/Wavefront (obj)”. Save your object as “plane.obj” in your
Godot project folder.
Shader magic
|image11|
I like to check “Normals” in the import popup so the import will also consider faces normals, which can be useful
(even if we don’t use them in this tutorial). Your mesh is now displayed in the FileSystem in “res://”.
|image12|
Create a MeshInstance node. In the Inspector, load the mesh we just imported. Select “plane.msh” and hit ok.
|image13|
Great! Our plane is now rendered in the 3D view.
|image14|
It is time to add some shader stuff. In the Inspector, in the “Material Override” line, add a “New ShaderMaterial”.
Edit it by clicking the “>” button just right to it.
|image15|
You have two ways to create a shader: by code (MaterialShader), or using a shader graph (MaterialShaderGraph).
The second one is a bit more visual, but we will not cover it for now. Create a “New MaterialShader”.
|image16|
Edit it by clicking the “>” button just right to it. The Shaders editor opens.
|image17|
The Vertex tab is for the Vertex shader, and the Fragment tab is for the Fragment shader. No need to explain what
both of them do, right? If so, head to the [[Shader]] page. Else, let’s start with the Fragment shader. This one is used
to texture the plane using an image. For this example, we will texture it with the heightmap image itself, so we’ll
actually see mountains as brighter regions and canyons as darker regions. Use this code:
This shader is very simple (it actually comes from the [[Shader]] page). What it basically does is take 2 parameters
that we have to provide from outside the shader (“uniform”):
• the texture file
• a color Then, we multiply every pixel of the image given by tex(source, UV).rgb by the color defined
col and we set it to DIFFUSE variable, which is the rendered color. Remember that the UV variable is a shader
variable that returns the 2D position of the pixel in the texture image, according to the vertex we are currently
dealing with. That is the use of the UV Layout we made before. The color col is actually not necessary to
display the texture, but it is interesting to play and see how it does, right?
However, the plane is displayed black! This is because we didn’t set the texture file and the color to use.
|image18|
In the Inspector, click the “Previous” button to get back to the ShaderMaterial. This is where you want to set the
texture and the color. In “Source”, click “Load” and select the texture file “heightmap.png”. But the mesh is still
black! This is because our Fragment shader multiplies each pixel value of the texture by the col parameter.
However, this color is currently set to black (0,0,0), and as you know, 0*x = 0 ;) . Just change the col parameter to
another color to see your texture appear:
|image19|
Good. Now, the Vertex Shader.
The Vertex Shader is the first shader to be executed by the pipeline. It deals with vertices.
Click the “Vertex” tab to switch, and paste this code:
This shader uses two “uniform” parameters. The source parameter is already set for the fragment shader. Thus, the
same image will be used in this shader as the heightmap. The height_range parameter is a parameter that we will
use to increase the height effect.
At line 3, we save the x and z position of the SRC_VERTEX, because we do not want them to change : the plane
must remain square. Remember that Y axis corresponds to the “altitude”, which is the only one we want to change
with the heightmap.
At line 4, we compute an h variable by multiplying the pixel value at the UV position and the height_range. As
the heightmap is a greyscale image, all r, g and b channels contain the same value. I used g, but any of r, g and b have
the same effect.
At line 5, we set the current vertex’ position at (xz.x, h, xz.y) position. Concerning xz.y remember that its type is
“vec2”. Thus, its components are x and y. The y component simply contains the z position we set at line 3.
Finally, at line 6, we multiply the vertex by the model/view matrix in order to set its position according to camera
position. If you try to comment this line, you’ll see that the mesh behaves weird as you move and rotate the camera.
That’s all good, but our plane remains flat. This is because the height_range value is 0. Increase this value to
observe the mesh distort and take to form of the terrain we set before:
|image20|
Asset pipeline
8.1 General
If you have read the previous tutorials on [[Resources]] and [[File System]], at this point you know that regular image
files (.png, .jpg, etc) are treated as regular resources in Godot.
Unlike texture resources (.tex files), image files contain no extra information on tiling (texture repeat), mipamps or
filtering. Editing this information and saving the texture back will have not any effect, since such formats can’t contain
that information.
Image loader
Loading of images is done by the image loader. The behavior of the loader for all image files can be changed in the
Project Settings dialog (Scene -> Project Settings). There is a section with values that correspond to the every image
file when loaded:
271
Godot Engine Documentation, Release latest
Filter
Filter is used when the image is stretched more than it’s original size, so a texel in the image is bigger than a pixel on
the screen. Turning off the fiter produces a retro-like look:
Repeat
Repeat is mainly used for 3D textures, so it’s off by default (textures are imported with the scenes and usually are not
in the project as image files). When using UV coordinates (something not as common in 2D), and the UV value goes
beyond the 0,0,1,1 rect, the texture repeats instead of clamping to the edge.
Mipmaps
When the mipmaps option is enabled, Godot will generate mip-maps. Mipmaps are versions of the image shrunk by
half in both axis, recursively, until the image is 1 pixel of size. When the 3D hardware needs to shrink the image, it
finds the largest mipmap it can scale from, and scales from there. This improves performance and image quality.
When Mip-Maps are disabled, images start distorting badly when shrunk excessively:
Alpha blending
The blending equation used by applications like Photoshop is too complex for real-time. There are better approxi-
mations such as pre-multiplied alpha, but they impose more stress in the asset pipeline. In the end, we are left with
textures that have artifacts in the edges, because apps such as Photoshop store white pixels in completely transparent
areas. Such white pixels end up showing thanks to the texture filter (when active).
Godot has an option to fix the edges of the image (by painting invisible pixels the same color as the visible neighbours):
To do this, open the image from the resources tab, or edit it from the property editor from another node or resource,
then go to the object options and select “Fix Border Alpha”, then save it.
Since fixing this in so many images can be a little annoying, both Texture Import and Image Export can also perform
this operation.
Texture import
Sometimes, it might be desired to change the above settings per image. Unfortunately, the image loader settings are
global. Texture flags also can’t be saved in a regular .png or .jpg file.
For such cases, the image can be imported as a texture (.tex), where the individual flags can be changed. Godot also
Image export
It is also possible to convert images to other formats (WebP or RAM compression) on export, as well as instructing
the exporter to create an Atlas for a set of images. It is also possible to ask the exporter to scale all images (or selected
groups).
More information on the [[Exporting images]] page.
8.2 Import
What is it for?
When Godot was created, it was probably after several failed and not so failed engine attempts (well, each attempt
failed a little less.. and so on). One of the most difficult areas of creating game engines is managing the import process.
That means, getting the assets that artists make into the game, in a way that functions optimally.
Artists use certain tools and formats, and programmers would rather have their data into a different format. This is
because artists put their focus on creating assets with the best quality possible, while programmers have to make sure
they actually run at decent speed (or run at all), use a certain amount of memory, and don’t take ages loading from
disk.
One would think that just writing a converter/importer would be enough, but this is not all there is to it. The same way
programmers iterate several times over their code, artists keep making changes to their assets. This generates some
bottleneck, because someone has to keep re-importing that artwork right? And importing assets is often something
that has to be agreed by both parties, as the programmer needs to decide how the artwork is imported and the artists
needs to see how it looks.
The goal to establishing an import process is that both can agree on how the rules under which the assets are going to
be imported the first time, and the system will apply those rules automatically each time the asset is re-imported.
Godot does not do the re-import process automatically, though. It gives the team the option to do it at any time ( a red
icon on the top right of the screen, allows the ability to do it at any desired time).
The aim of the import system is that it works well enough for most common cases and projects. What is there has
been tested and seems to cover most needs.
However, as mentioned before, this is on of the most difficult areas of writing a game engine. It may happen often
(specially on large projects, ports, or projects with unusual requirement) that what is provided is not enough. It’s easy
to say that the engine is open source and that the programmer should make their own if they don’t like what is there,
but that would be making a huge disservice to the users and not the right attitude. Because of that, we made sure to
provide as many tools and helpers as possible to support a custom import process, for example:
• Access to the internals of almost all data structures is provided to the scripting and C++ API, as well as saving
and loading in all supported file formats.
• Some importers (like the 3D asset importer) support scripts to modify the data being imported.
• Support for creating custom import plugins is also provided, even for replacing the existing ones.
• If all else fails, Godot supports for adding custom resource loaders, to load data in alternative formats, without
intermediate conversion.
Both the import system and the custom tools provided will improve over time as more use cases are revealed to us.
Importing assets
To begin, it is a good idea to define where the original assets created by the artists (before they are imported) will be
located. Normally, Godot does not mind much about the location, but if the project has several developers, it is a good
idea to understand the simple rule for it to work for everyone.
First of all, it would be really good for this location to not be inside the project path (where engine.cfg is located,
or any sub-folder). Godot expects regular resources in there, and may consider many of the files used as source art
as regular resources. This would lead to it bundling all of them when the project is exported, something which is
undesired.
Now that it is clear that this location must be outside the project folder, the rule that Godot uses to reference external
assets can be explained. When an asset is imported, the engine stores a relative path from the project path to the asset
(In windows, this works as long as they are on the same drive, otherwise an absolute path is stored). This ensures that
the same asset can be re-imported in another computer.
The usual approach to this, when using a VCS such as Subversion, Perforce or GIT, is to create the project in a
subfolder, so both it and the source assets can be commited to a same repository. For example:
Repository layout:
source_assets/sfx/explosion.wav
source_assets/sfx/crash.wav
source_assets/fonts/myfont.ttf
source_assets/translation/strings.csv
source_assets/art/niceart.psd
game/engine.cfg
In the above example, artists, musican, translators, etc. can work in the source_assets/ folder, then import the assets to
the game/ folder. When the repository is updated, anyone can re-import the assets if they changed.
Import dialogs
Godot provides for importing several types of assets, all of them can be accessed from the import dialog:
Each of the dialog shares a similar function, a source file (or several of them) must be provided, as well as a target
destination inside the project folders. Once imported, Godot saves this information as metadata in the imported asset
itself.
More information about each specific type of asset can be found in specific sections, such as Importing Textures.
Godot tracks changes in the source assets constantly. If at least one asset has been found to be modified (md5 is
different than when it was imported), a small red indicator will appear in the top right corner of the screen.
From that moment onward, the user can choose to re-import at any given time by clicking on the red-icon. When this
action is done, a dialog will pop-up showing which resources can be re-imported (all selected by default).
Accepting that dialog will immediately re-import the resources and will update any of them currently in use in the
editor (like a texture, model or audio file).
Manually re-importing
The re-import process is automatic, but it may be desired at some point to change the settings of an already imported
file, so it can be re-imported differently. For this, the Import Settings window is provided.
This screen allows the user to re-open the corresponding import-window to re-import that asset again, with the ability
to change any of the settings.
In most cases you don’t want images imported when dealing with 2D and GUI. Just copy them to the filesystem. Read
the tutorial on [[Image_Files|dealing with image files]] before continuing! For 3D, textures are always imported by the
3D scene importer, so importing those is only useful when importing a texture used for 3D that doesn’t come with the
3D scene (for example, in a shader). The flags and options are the same as here, so reading the rest of the document
might help too.
So, if you have read the previous tutorial on the texture exporter, the texture importer gives you more finer grained
control on how textures are imported. If you want to change flags such as repeat, filter, mip-maps, fix edges, etc. *PER
texture*, importing them is the best way to accomplish this (since you can’t save such flags in a standard image file).
Lack of MipMaps
Images in 3D hardware are scaled with a (bi)linear filter, but this method has limitations. When images are shrunk too
much, two problems arise:
• Aliasing: Pixels are skipped too much, and the image shows discontinuities. This decrases quality.
• Cache Misses: Pixels being read are too far apart, so texture cache reads a lot more data than it should. This
decreases performance.
(Todo, find image sample of why it looks bad)
To solve this, mipmaps are created. Mipmaps are versions of the image shrunk by half in both axis, recursively, until
the image is 1 pixel of size. When the 3D hardware needs to shrink the image, it finds the largest mipmap it can scale
from, and scales from there. This improves performance and image quality.
Godot automatically creates mipmaps upon load for standard image files. This process is time consuming (although
not much) and makes load times a little worse. Pre-importing the textures allows the automatic generation of mipmaps.
Unwanted MipMaps
Remember the previous point about mipmaps? Yes, they are cool, but mobile GPUs only support them if the textures
are in power of 2 dimensions (ie 256x256 or 512x128). In these platforms, Godot will stretch and enlarge the texture
to the closest power of 2 size and then generate the mipmaps. This process takes more of a performance hit and it
might degrade the quality a little more.
Because of this, there are some scenarios when it may be desirable to not use them, and just use a linear filter. One of
them is when working with graphical user interfaces (GUIs). Usually they are made of large images and don’t stretch
much. Even if the screen resolution is in a larger or smaller value than original art, the amount of stretch is not as
much and the art can retain the quality. Pre-importing the textures also allows the disabling of mipmap generation.
Blending artifacts
The blending equation used by applications like Photoshop is too complex for realtime. There are better approxi-
mations such as pre-multiplied alpha, but they impose more stress in the asset pipeline. In the end, we are left with
textures that have artifacts in the edges, because apps such as Photoshop store white pixels in completely transparent
areas. Such white pixels end up showing thanks to the texture filter.
Godot has an option to fix the edges of the image (by painting invisible pixels the same color as the visible neighbours):
However, this must be done every time the image changes. Pre-Importing the textures makes sure that every time the
original file changes, this artifact is fixed upon automatic re-import.
Texture flags
Textures have flags. The user can choose for them to repeat or clamp to edges (when UVs exceed the 0,0,1,1 boundary).
The magnifying filter can also be turned off (for a Minecraft-like effect). Such values can not be edited in standard file
formats (png, jpg, etc), but can be edited and saved in Godot .tex files. Then again, the user may not want to change
the values every time the texture changes. Pre-Importing the textures also takes care of that.
Texture compression
Asides from the typical texture compression, which saves space on disk (.png, jpg, etc), there are also texture com-
pression formats that save space in memory (more specifically video memory. This allows to have much better looking
textures in games without running out of memory, and decrease memory bandwidth when reading them so they are a
big plus.
Video texture compression formats are several and non standard. Apple uses PVRTC. PC GPUs, consoles and nVidia
Android devices use S3TC (BC), other chipsets use other formats. OpenGL ES 3.0 standardized on ETC format, but
we are still a few years away from that working everywhere.
Still, when using this option, Godot converts and compresses to the relevant format depending on the target platform
(as long as the user pre-imported the texture and specified video ram compression!).
This kind of compression is often not desirable for many types 2D games and UIs because it has visible visual artifacts.
This is specially noticeable on games that use the trendy vectory social game artwork. However, again, the fact that it
saves space and improves performance may make up for it.
The 3D scene importer always imports textures with this option turned on.
Atlases
Remember how mobile GPUs have this limitation of textures having to be in power of 2 sizes to be able to generate
mimpmaps for optimum stretching? What if we have a lot of images in different random sizes? All will have to be
scaled and mipmapped when loaded (using more CPU and memory) or when imported (using more memory). This is
probably still ok, but there is a tool that can help improve this situation.
Atlases are big textures that fit a lot of small textures inside efficiently. Godot supports creating atlases in the importer,
and the imported files are just small resources that reference a region of the bigger texture.
Atlases can be a nice solution to save some space on GUI or 2D artwork by packing everything together. The current
importer is not as useful for 3D though (3D Atlasses are created differently, and not all 3D models can use them).
As a small plus, atlases can decrease the amount of “state changes” when drawing. If a lot of objects that are drawn
using several different textures are converted to atlas, then the texture rebinds per object will go from dozens or
hundreds to one. This will give the performance a small boost.
Still wondering whether to use the texture importer or not? Remember that in the end, artists will often use Photoshop
anyway, so it may be wiser to just let the import subsystem to take care of importing and converting the PSD files
instead of asking the artist to save a png and copy it to the project every time.
Texture importer
Finally! It’s time to take a look at the texture importer. There are 3 options in the import menu. They are pretty much
(almost) the same dialog with a different set of defaults.
When selected, the texture import dialog will appear. This is the default one for 2D textures:
Source texture(s)
One or more source images can be selected from the same folder (this importer can do batch-conversion). This can be
from inside or outside the project.
Target path
A destination folder must be provided. It must be inside the project, as textures will be converted and saved to it.
Extensions will be changed to .tex (Godot resource file for textures), but names will be kept.
Texture format
This combo allows to change the texture format (compression in this case):
Each of the four options described in this table together with their advantages and disadvantages ( |image5| = Best,
|image6| =Worst ):
Texture options
Provided are a small amount of options for fine grained import control:
• Streaming Format - This does nothing as of yet, but a texture format for streaming different mipmap levels is
planned. Big engines have support for this.
• Fix Border Alpha - This will fix texture borders to avoid the white auras created by white invisible pixels (see
the rant above).
• Alpha Bit Hint - Godot auto-detects if the texture needs alpha bit support for transparency (instead of full
range), which is useful for compressed formats such as BC. This forces alpha to be 0 or 1.
• Compress Extra - Some VRAM compressions have alternate formats that compress more at the expense of
quality (PVRTC2 for example). If this is ticked, texture will be smaller but look worse.
• No MipMaps - Force imported texture to NOT use mipmaps. This may be desirable in some cases for 2D (as
explained in the rant above), though it’s NEVER desirable for 3D.
• Repeat - Texture will repeat when UV coordinates go beyond 1 and below 0. This is often desirable in 3D, but
may generate artifacts in 2D.
• Filter - Enables linear filtering when a texture texel is larger than a screen pixel. This is usually turned on,
unless it’s required for artistic purposes (minecraft look, for example).
What is a font?
Fonts in modern operating systems are created as scalable vector graphics. They are stored as a collection of curves
(usually one for each character), which are independent of the screen resolution, and stored in standardized file formats,
such as TTF (TrueType) or OTF (OpenType).
Rendering such fonts to bitmaps is a complex process, which employs different methods to convert curves to pixels
depending on context and target size. Due to this, this rendering process must be done by using the CPU. Game
engines use the GPU to render, and 3D APIs don’t really support the means to do this efficiently, so fonts have to be
converted to a format that is friendly to the GPU when imported to a project.
Converting fonts
This conversion process consists of rendering a vector font to a given point size and storing all the resulting characters
in a bitmap texture. The bitmap texture is then used by the GPU to draw a small quad for each character and form
readable strings.
The drawback of this process is that fonts must be pre-imported in the specific sizes that they will use in the project.
However, given that that bitmap fonts compress really well, this is not as bad as it sounds.
Importing a font
Fonts are imported via the Font import dialog. The dialog will ask for a font, a size, some options and a target resource
fie to save.
The dialog is fully dynamic, which means that any change will be reflected in the font preview window. The user ccan
tweak almost every parameter and get instant feedback on how the font will look.
Since the resulting font is a bitmap, a few more options were added to make the imported font look even nicer. These
options were added to please graphic designers, who love putting gradients, outlines and shadows in fonts, as well as
changing all the inter-spaces available :). The options which will be explained in the next section.
Extra spacing
Fonts can be added a shadow. For this, the font is drawn again below on a different color and the blurred with a
gaussian kernel of different sizes. The resulting shadow can be adjusted with an exponential function to make it softer
or more like an outline. A second shadow is also provided to create some added effects, like a bump or outline+shadow.
Gradients
Gradients are also another of the visual effects that graphic designers often use. To show how much we love them, we
added those too. Gradients can be provided as a simple curve between two colors, or a special png file with a hand
drawn gradient.
Internationalization
Colors, shadows and gradients are beautiful, but it’s time we get to serious business. Developing games for Asian
markets is a common practice in today’s globalized world and app stores.
Here’s when things get tricky with using bitmap fonts. Asian alphabets (Chinese, Japanese and Korean) contains
dozens of thousands of characters. Generating bitmap fonts with every single of them is pretty expensive, as the
resulting textures are huge. If the font size is small enough, it can be done without much trouble, but when the fonts
become bigger, we run out of video ram pretty quickly!
To solve this, Godot allows the user to specify a text file (in UTF-8 format) where it expects to find all the characters
that will be used in the project. This seems difficult to provide at first, and more to keep up to date, but it becomes
rather easy when one realizes that the .csv with the translations can be used as such source file (see the [[Import-
ing_translations]] section). As Godot re-imports assets when their dependencies change, both the translation and font
files will be updated and re-imported automatically if the translation csv changes.
Another cool trick for using a text file as limit of which characters can be imported is when using really large fonts.
For example, the user might want to use a super large font, but only to show numbers. For this, he or she writes a
numbers.txt file that contains “1234567890”, and Godot will only limit itself to import data, thus saving a lot of video
memory.
Why importing?
Importing Audio Samples into the game engine is a process that should be easier than it really is. Most readers are
probably thinking “Why not just copying the .wav files to a folder inside the project and be over with it?”.
It’s not usually that simple. Most game engines use uncompressed audio (in memory at least) for sound effects. The
reason for this is because it’s really cheap to play back and resample. Compressed streamed audio (such as .ogg files)
takes a large amount of processor to decode so no more than one or two are streamed simultaneously. However, with
sound effects, one expects a dozen of them to be playing at the same time in several situations.
Because of this, sound effects are loaded uncompressed into memory, and here is where the problems begin.
As is usual with graphics, the situation where programmers don’t really know about audio and audio engineers don’t
know about programming is also common in the industry. This leads to a scenario where a project ends up wasting
resources unnecessarily.
To be more precise, sfx artists tend to work with audio formats that give them a lot of room for tweaking the audio
with a low noise floor minimum aliasing, such as 96khz, 24 bits. In many cases, they work in stereo too. Added to
that, many times they add effects with an infinite or really long fadeout, such as reverb, which take a long time to fade
out. Finally, many DAWs also add silence at the beginning when normalizing to wav.
This results in extremely large files to integrate more often than desired, with sound effects taking dozens of megabytes.
First of all, it is important to know that Godot has an internal reverb generator. Sound effects can go to four different
setups (small, medium and large room as well as hall), with different send amounts. This saves sfx artists the need to
add reverb to the sound effects, reducing their size greatly and ensuring correct trimming. Say no to SFX with baked
reverb!
Another common problem is that, while it’s useful for working inside a DAW, high dynamic range (24 bits) and high
sampling rate (96khz) is completely unnecessary for use in a game, as there is no audible difference. If positional
sound is going to be used (for 2D and 3D), the panning and stereo reverb will be provided by the engine, so there is
little need for stereo sound. How does this affect the resource usage? Look at the following comparison:
Format 1 Second of Audio Frame Size
24 bits, 96 khz, Stereo 576kb 12
16 bits, 44 khz, Mono 88kb 2
16 bits, IMA-ADPCM 22kb 1/2
As seen, for being no audible difference, the 16 bits, 44khz takes 6 times less memory than the 24 bits, 96khz, Stereo
version. The IMA-ADPCM version takes 24 times less memory than what was exported from the DAW.
Trimming
One last issue that happens often is that the waveform files received have silences at the beginning and at the end.
These are inserted by DAWs when saving to a waveform, increase their size unnecessarily and add latency to the
moment they are played back. Trimming them solves this, but it takes effort for the sfx artist, as they have to do it in a
separate application. In the worst case, they may not even know the silences are being added.
Godot has a simple screen for importing audio samples to the engine. SFX artists only have to save the .wav files to a
folder outside the project, and the import dialog will fix the files for inclusion, as well as doing it automatically every
time they are modified and re-imported.
In this screen, the quality of the audio can be limited to what is needed, and trimming is done automatically. As a plus,
several samples can be loaded and batch-converted, just like textures.
Looping
Godot supports looping in the samples (Tools such as Sound Forge or Audition can add loop points to .wav files). This
is useful for sound effects such as engines, machine guns, etc. Ping-pong looping is also supported.
As an alternative, the import screen has a “loop” option that enables looping for the entire sample when importing.
The world is full of different markets and cultures and, to maximize profits™, nowadays games are released in several
languages. To solve this, internationalized text must be supported in any modern game engine.
In regular desktop or mobile applications, internationalized text is usually located in resource files (or .po files for
GNU stuff). Games, however, can use several orders of magnitude more text than applications, so they must support
efficient methods for dealing with loads of multi-language text.
There are two approaches to generate multi language games and applications. Both are based on a key:value system.
The first is to use one of the languages as key (usually english), the second is to use a specific identifier. The first
approach is probably easier for development if a game is released first in english, later in other languages, but a
complete nightmare if working with many languages at the same time.
In general, games use the second approach and a unique ID is used for each string. This allows to revise the text while
it’s being translated to others. the unique ID can be a number, a string, or a string with a number (it’s just a unique
string anyway).
Translators also, most of the time prefer to work with spreadsheets (either as a Microsoft Excel file or a shared Google
Spreadsheet).
Translation format
To complete the picture and allow efficient support for translations, Godot has a special importer that can read .csv
files. Both Microsoft Excel and Google Spreadsheet can export to this format, so the only requirement is that the files
have a special format. The csv files must be saved in utf-8 encoding and the format is as follows:
Import dialog
The import dialog takes a .csv file in the previously described format and generates several compressed translation
resource files inside the project.
Selecting a .csv file autodetects the languages from the first row. and determines which column represents which
language. It is possible to change that manually, by selecting the language for each column.
The import dialog also can add the translation to the list of translations to load when the game runs, specified in
engine.cfg (or the project properties). Godot allows to load and remove translations at runtime, too.
8.3 Export
Why Exporting?
Originally, Godot did not have any means to export projects. The developers would compile the proper binaries and
build the packages for each platform manually.
When more developers (and even non-programmers) started using it, and when our company started taking more
projects at the same time, it became evident that this was a bottleneck.
On PC
Distributing a game project on PC with Godot is rather easy. Just drop the godot.exe (or godot) binary together in the
same place as the engine.cfg file, zip it and you are done. This can be taken advantage to make custom installers.
It sounds simple, but there are probably a few reasons why the developer may not want to do this. The first one is that
it may not be desirable to distribute loads of files. Some developers may not like curious users peeking at how the
game was made, others may just find it inelegant, etc.
Another reason is that, for distribution, the developer might use a specially compiled binary, which is smaller in size,
more optimized and does not include tools inside (like the editor, debugger, etc).
Finally, Godot has a simple but efficient system for creating DLCs as extra package files.
On Mobile
The same scenario in mobile is a little worse. To distribute a project in those devices, a binary for each of those
platforms is built, then added to a native project together with the game data.
This can be troublesome because it means that the developer must be familiarized with the SDK of each platform
before even being able to export. In other words, while learning each SDK is always encouraged, it can be frustrating
to be forced to do it at an undesired time.
There is also another problem with this approach, which is the fact that different devices prefer some data in different
formats to run. The main example of this is texture compression. All PC hardware uses S3TC (BC) compression and
that has been standardized for more than a decade, but mobile devices use different formats for texture compression,
such as PVRCT (iOS) or ETC (Android)
Export Dialog
After many attempts at different export workflows, the current one has worked the best. At the time of this writing,
not all platforms are supported yet, but that will change soon.
To open the export dialog, just click the “Export” Button:
The dialog will open, showing all the supported export platforms:
The default options are often enough to export, so tweaking them is not necessary until it’s needed. However, many
platforms require additional tools (SDKs) to be installed to be able to export. Additionally, Godot needs exports
templates installed to create packages. The export dialog will complain when something is missing and will not allow
the user to export for that platform until he or she resolves it:
At that time, the user is expected to come back to the wiki and follow instructions on how to properly set up that
platform.
Export Templates
Apart from setting up the platform, the export templates must be installed to be able to export projects. They can be
downloaded as a .tpz (a renamed .zip) file from the wiki.
Once downloaded, they can be installed using the “Install Export Templates” option in the editor:
Export Mode
When exporting, Godot makes a list of all the files to export and then creates the package. There are 3 different modes
for exporting:
• Export every single file in the project
• Export only resources (+custom filter), this is default.
• Export only selected resources (+custom filter)
• Export every single file - This mode exports every single file in the project. This is good to test if something
is being forgotten, but developers often have a lot of unrelated stuff around in the dev dir, which makes it a bad
idea.
• Export only resources - Only resources are exported. For most projects, this is enough. However many
developers like to use custom datafiles in their games. To compensate for this, filters can be added for extra
extensions (like, .txt,.csv, etc).
• Export only selected resources - Only select resources from a list are exported. This is probably overkill for
most projects, but in some cases it is justified (usually huge projects). This mode offers total control of what is
exported. Individual resources can be selected and dependency detection is performed to ensure that everything
needed is added. As a plus, this mode allows to “Bundle" scenes and dependencies into a single file,
which is really useful for games distributed on optical media.
This feature will pop up automatically once a platform is properly configured and a supported device is connected
to the computer. Since things can go wrong at many levels (platform may not be configured correctly, SDK may
incorrectly installed, device may be improperly configured, kitty ate the USB cable, etc.), it’s good to let the user
know that it exists.
Some platforms (at the time of this writing, only Android and Blackberry 10) can detect when a USB device is
connected to the computer, and offer the user to automatically export, install and run the project (in debug mode) on
the device. This feature is called, in industry buzz-words, “One Click Deploy” (though, it’s technically two clicks...).
2. Configure device (make sure it’s in developer mode, likes the computer, usb is recognized, usb cable is plugged,
etc).
3. Connect the device..
4. And Voila!
It is often desired to do an operation to all or a group of images upon export. Godot provides some tools for this.
Examples of such operations are:
• Converting all images from a lossless format to a lossy one (ie: png -> web) for greater compression.
• Shrinking all images to half the size, to create a low resolution build for smaller screens.
• Create an atlas for a group of images and crop them, for higher performance and less memory usage.
In this dialog the image extensions for conversion can be selected, and operations can be performed that apply to all
images (except those in groups -next section for that-):
• Convert Image Format: Probably the most useful operation is to convert to Lossy (WebP) to save disk space.
For lossy, a Quality bar can set the quality/vs size ratio.
• Shrink: This allows to shrink all images by a given amount. It’s useful to export a game to half or less resolution
for special devices.
• Compress Formats: Allows to select which image exensions to convert.
On export, Godot will perform the desired operation. The first export might be really slow, but subsequent exports will
be fast, as the converted images will be cached.
This section is similar to the previous one, except it can operate on a selected group of images. When a image is in a
group, the settings from the global export options are overridden by the ones from the group. An image can only be in
one group at the same time. So if the image is in another group different to the current one being edited, it will not be
selectable.
Atlas
As a plus, an atlas can be created from a group. When this mode is active, a button to preview the resulting atlas
becomes available. Make sure that atlases don’t become too big, as some hardware will not support textures bigger
than 2048x2048 pixels. If this happens, just create another atlas.
The atlas can be useful to speed up drawing of some scenes, as state changes are minimized when drawing from it
(through unlike other engines, Godot is designed so state changes do not affect it as much). Textures added to an
atlas get cropped (empty spaces around the image are removed), so this is another reason to use them (save space). If
unsure, though, just leave that option disabled.
The simplest way to distribute a game for PC is to copy the executables (godot.exe on windows, godot on the rest), zip
the folder and send it to someone else. However, this is often not desired.
Godot offers a more elegant approach for PC distribution when using the export system. When exporting for PC
(Linux, Windows, Mac), the exporter takes all the project files and creates a “data.pck” file. This file is bundled with
a specially optimized binary that is smaller, faster and lacks tools and debugger.
Optionally, the files can be bundled inside the executable, though this does not always works properly.
Exporting for android has much less requirements than compiling Godot for it. As follows are the steps to setup the
SDK and the engine.
Download and install Java 6 or OpenJDK 6, Android needs this version and it seems that jarsigner (what is used to
sign APKs) from greater versions do not work.
Create a debug.keystore
Android needs a debug keystore file to install to devices and distribute non-release APKs. If you have used the SDK
before and have built projects, ant or eclipse probably generated one for you (In Linux and OSX, you can find it in the
~/.android folder).
If you can’t find it or need to generate one, the keytool command from the JDK can be used for this purpose:
keytool -keyalg RSA -genkeypair -alias androiddebugkey -keypass android -keystore debug.keystore -storepass an-
droid -dname “CN=Android Debug,O=Android,C=US” -validity 9999
ADB is the command line tool used to communicate with Android devices. It’s installed with the SDK, but you may
need to install one (any) of the Android API levels for it to be installed in the SDK directory.
Setting it up in Godot
Enter the Editor Settings screen. This screens contains the editor settings for the user account in the computer (It’s
independent from the project).
Scroll down to the section where the Android settings are located:
Exporting for iOS is done manually at the moment. These are the steps to load your game in an XCode project, where
you can deploy to a device, publish, etc.
Requirements
Using the Godot editor, [[Exporting_for_pc|export your project for Windows]], to obtain the data.pck file. Replace the
empty data.pck in the XCode project with the new one, and run/archive.
If you want to test your scenes on the iOS device as you edit them, you can add your game directory to the project
(instead of data.pck), and add a property “godot_path” to Info.plist, with the name of your directory as its value.
Alternatively you can add all the files from your game directly, with “engine.cfg” at the root.
Sometimes your game becomes too big and deploying to the device takes too long every time you run. In that case
you can deploy only the engine executable, and serve the game files from your computer.
On your PC, open the editor, and click the righ-most icon on the top-center group of icons, and select “Enable File
Server”. The icon turns red. Your PC will open a port and accept connections to serve files from your project’s
directory (so enable your local firewall accordingly).
On XCode, click on your app name (top left, next to the “Stop” button), and select “Edit Scheme”. Go to the “Argu-
ments” tab, and add 2 arguments, “-rfs” and the IP of your PC.
When you run, your device will connect to the host and open the files remotely. Note that the directory with the game
data (“platformer”) is no longer added to the project, only the engine executable.
Special iOS services can be used in Godot. Check out the [[Services for iOS]] page.
Classes reference
9.1 @GDScript
Category: Core
307
Godot Engine Documentation, Release latest
• PI = 3.141593 — Constant that represents how many times the diameter of a circumference fits around it’s
perimeter.
9.1.4 Description
This contains the list of built-in gdscript functions. Mostly math functions and other utilities. Everything else is
expanded by objects.
Category: Core
• Performance Performance
• Globals Globals
• IP IP
• Geometry Geometry
• ResourceLoader ResourceLoader
• ResourceSaver ResourceSaver
• PathRemap PathRemap
• OS OS
• Reference Marshalls
• TranslationServer TranslationServer
• TranslationServer TS
• Input Input
• InputMap InputMap
• VisualServer VisualServer
• VisualServer VS
• AudioServer AudioServer
• AudioServer AS
• PhysicsServer PhysicsServer
• PhysicsServer PS
• Physics2DServer Physics2DServer
• Physics2DServer PS2D
• SpatialSoundServer SpatialSoundServer
• SpatialSoundServer SS
• SpatialSound2DServer SpatialSound2DServer
• SpatialSound2DServer SS2D
• MARGIN_LEFT = 0 — Left margin, used usually for Control or StyleBox derived classes.
• MARGIN_TOP = 1 — Top margin, used usually for Control or StyleBox derived classes.
• MARGIN_RIGHT = 2 — Right margin, used usually for Control or StyleBox derived classes.
• MARGIN_BOTTOM = 3 — Bottom margin, used usually for Control or StyleBox derived classes.
• VERTICAL = 1 — General vertical alignment, used usually for Separator, ScrollBar, Slider, etc.
• HORIZONTAL = 0 — General horizontal alignment, used usually for Separator, ScrollBar, Slider, etc.
• HALIGN_LEFT = 0 — Horizontal left alignment, usually for text-derived classes.
• HALIGN_CENTER = 1 — Horizontal center alignment, usually for text-derived classes.
• HALIGN_RIGHT = 2 — Horizontal right alignment, usually for text-derived classes.
• VALIGN_TOP = 0 — Vertical top alignment, usually for text-derived classes.
• VALIGN_CENTER = 1 — Vertical center alignment, usually for text-derived classes.
• VALIGN_BOTTOM = 2 — Vertical bottom alignment, usually for text-derived classes.
• SPKEY = 16777216 — Scancodes with this bit applied are non printable.
• KEY_ESCAPE = 16777217 — Escape Key
• KEY_TAB = 16777218 — Tab Key
• KEY_BACKTAB = 16777219 — Shift-Tab key
• KEY_BACKSPACE = 16777220
• KEY_RETURN = 16777221
• KEY_ENTER = 16777222
• KEY_INSERT = 16777223
• KEY_DELETE = 16777224
• KEY_PAUSE = 16777225
• KEY_PRINT = 16777226
• KEY_SYSREQ = 16777227
• KEY_CLEAR = 16777228
• KEY_HOME = 16777229
• KEY_END = 16777230
• KEY_LEFT = 16777231
• KEY_UP = 16777232
• KEY_RIGHT = 16777233
• KEY_DOWN = 16777234
• KEY_PAGEUP = 16777235
• KEY_PAGEDOWN = 16777236
• KEY_SHIFT = 16777237
• KEY_CONTROL = 16777238
• KEY_META = 16777239
• KEY_ALT = 16777240
• KEY_CAPSLOCK = 16777241
• KEY_NUMLOCK = 16777242
• KEY_SCROLLLOCK = 16777243
• KEY_F1 = 16777244
• KEY_F2 = 16777245
• KEY_F3 = 16777246
• KEY_F4 = 16777247
• KEY_F5 = 16777248
• KEY_F6 = 16777249
• KEY_F7 = 16777250
• KEY_F8 = 16777251
• KEY_F9 = 16777252
• KEY_F10 = 16777253
• KEY_F11 = 16777254
• KEY_F12 = 16777255
• KEY_F13 = 16777256
• KEY_F14 = 16777257
• KEY_F15 = 16777258
• KEY_F16 = 16777259
• KEY_KP_ENTER = 16777344
• KEY_KP_MULTIPLY = 16777345
• KEY_KP_DIVIDE = 16777346
• KEY_KP_SUBSTRACT = 16777347
• KEY_KP_PERIOD = 16777348
• KEY_KP_ADD = 16777349
• KEY_KP_0 = 16777350
• KEY_KP_1 = 16777351
• KEY_KP_2 = 16777352
• KEY_KP_3 = 16777353
• KEY_KP_4 = 16777354
• KEY_KP_5 = 16777355
• KEY_KP_6 = 16777356
• KEY_KP_7 = 16777357
• KEY_KP_8 = 16777358
• KEY_KP_9 = 16777359
• KEY_SUPER_L = 16777260
• KEY_SUPER_R = 16777261
• KEY_MENU = 16777262
• KEY_HYPER_L = 16777263
• KEY_HYPER_R = 16777264
• KEY_HELP = 16777265
• KEY_DIRECTION_L = 16777266
• KEY_DIRECTION_R = 16777267
• KEY_BACK = 16777280
• KEY_FORWARD = 16777281
• KEY_STOP = 16777282
• KEY_REFRESH = 16777283
• KEY_VOLUMEDOWN = 16777284
• KEY_VOLUMEMUTE = 16777285
• KEY_VOLUMEUP = 16777286
• KEY_BASSBOOST = 16777287
• KEY_BASSUP = 16777288
• KEY_BASSDOWN = 16777289
• KEY_TREBLEUP = 16777290
• KEY_TREBLEDOWN = 16777291
• KEY_MEDIAPLAY = 16777292
• KEY_MEDIASTOP = 16777293
• KEY_MEDIAPREVIOUS = 16777294
• KEY_MEDIANEXT = 16777295
• KEY_MEDIARECORD = 16777296
• KEY_HOMEPAGE = 16777297
• KEY_FAVORITES = 16777298
• KEY_SEARCH = 16777299
• KEY_STANDBY = 16777300
• KEY_OPENURL = 16777301
• KEY_LAUNCHMAIL = 16777302
• KEY_LAUNCHMEDIA = 16777303
• KEY_LAUNCH0 = 16777304
• KEY_LAUNCH1 = 16777305
• KEY_LAUNCH2 = 16777306
• KEY_LAUNCH3 = 16777307
• KEY_LAUNCH4 = 16777308
• KEY_LAUNCH5 = 16777309
• KEY_LAUNCH6 = 16777310
• KEY_LAUNCH7 = 16777311
• KEY_LAUNCH8 = 16777312
• KEY_LAUNCH9 = 16777313
• KEY_LAUNCHA = 16777314
• KEY_LAUNCHB = 16777315
• KEY_LAUNCHC = 16777316
• KEY_LAUNCHD = 16777317
• KEY_LAUNCHE = 16777318
• KEY_LAUNCHF = 16777319
• KEY_UNKNOWN = 33554431
• KEY_SPACE = 32
• KEY_EXCLAM = 33
• KEY_QUOTEDBL = 34
• KEY_NUMBERSIGN = 35
• KEY_DOLLAR = 36
• KEY_PERCENT = 37
• KEY_AMPERSAND = 38
• KEY_APOSTROPHE = 39
• KEY_PARENLEFT = 40
• KEY_PARENRIGHT = 41
• KEY_ASTERISK = 42
• KEY_PLUS = 43
• KEY_COMMA = 44
• KEY_MINUS = 45
• KEY_PERIOD = 46
• KEY_SLASH = 47
• KEY_0 = 48
• KEY_1 = 49
• KEY_2 = 50
• KEY_3 = 51
• KEY_4 = 52
• KEY_5 = 53
• KEY_6 = 54
• KEY_7 = 55
• KEY_8 = 56
• KEY_9 = 57
• KEY_COLON = 58
• KEY_SEMICOLON = 59
• KEY_LESS = 60
• KEY_EQUAL = 61
• KEY_GREATER = 62
• KEY_QUESTION = 63
• KEY_AT = 64
• KEY_A = 65
• KEY_B = 66
• KEY_C = 67
• KEY_D = 68
• KEY_E = 69
• KEY_F = 70
• KEY_G = 71
• KEY_H = 72
• KEY_I = 73
• KEY_J = 74
• KEY_K = 75
• KEY_L = 76
• KEY_M = 77
• KEY_N = 78
• KEY_O = 79
• KEY_P = 80
• KEY_Q = 81
• KEY_R = 82
• KEY_S = 83
• KEY_T = 84
• KEY_U = 85
• KEY_V = 86
• KEY_W = 87
• KEY_X = 88
• KEY_Y = 89
• KEY_Z = 90
• KEY_BRACKETLEFT = 91
• KEY_BACKSLASH = 92
• KEY_BRACKETRIGHT = 93
• KEY_ASCIICIRCUM = 94
• KEY_UNDERSCORE = 95
• KEY_QUOTELEFT = 96
• KEY_BRACELEFT = 123
• KEY_BAR = 124
• KEY_BRACERIGHT = 125
• KEY_ASCIITILDE = 126
• KEY_NOBREAKSPACE = 160
• KEY_EXCLAMDOWN = 161
• KEY_CENT = 162
• KEY_STERLING = 163
• KEY_CURRENCY = 164
• KEY_YEN = 165
• KEY_BROKENBAR = 166
• KEY_SECTION = 167
• KEY_DIAERESIS = 168
• KEY_COPYRIGHT = 169
• KEY_ORDFEMININE = 170
• KEY_GUILLEMOTLEFT = 171
• KEY_NOTSIGN = 172
• KEY_HYPHEN = 173
• KEY_REGISTERED = 174
• KEY_MACRON = 175
• KEY_DEGREE = 176
• KEY_PLUSMINUS = 177
• KEY_TWOSUPERIOR = 178
• KEY_THREESUPERIOR = 179
• KEY_ACUTE = 180
• KEY_MU = 181
• KEY_PARAGRAPH = 182
• KEY_PERIODCENTERED = 183
• KEY_CEDILLA = 184
• KEY_ONESUPERIOR = 185
• KEY_MASCULINE = 186
• KEY_GUILLEMOTRIGHT = 187
• KEY_ONEQUARTER = 188
• KEY_ONEHALF = 189
• KEY_THREEQUARTERS = 190
• KEY_QUESTIONDOWN = 191
• KEY_AGRAVE = 192
• KEY_AACUTE = 193
• KEY_ACIRCUMFLEX = 194
• KEY_ATILDE = 195
• KEY_ADIAERESIS = 196
• KEY_ARING = 197
• KEY_AE = 198
• KEY_CCEDILLA = 199
• KEY_EGRAVE = 200
• KEY_EACUTE = 201
• KEY_ECIRCUMFLEX = 202
• KEY_EDIAERESIS = 203
• KEY_IGRAVE = 204
• KEY_IACUTE = 205
• KEY_ICIRCUMFLEX = 206
• KEY_IDIAERESIS = 207
• KEY_ETH = 208
• KEY_NTILDE = 209
• KEY_OGRAVE = 210
• KEY_OACUTE = 211
• KEY_OCIRCUMFLEX = 212
• KEY_OTILDE = 213
• KEY_ODIAERESIS = 214
• KEY_MULTIPLY = 215
• KEY_OOBLIQUE = 216
• KEY_UGRAVE = 217
• KEY_UACUTE = 218
• KEY_UCIRCUMFLEX = 219
• KEY_UDIAERESIS = 220
• KEY_YACUTE = 221
• KEY_THORN = 222
• KEY_SSHARP = 223
• KEY_DIVISION = 247
• KEY_YDIAERESIS = 255
• KEY_CODE_MASK = 33554431
• KEY_MODIFIER_MASK = -16777216
• KEY_MASK_SHIFT = 33554432
• KEY_MASK_ALT = 67108864
• KEY_MASK_META = 134217728
• KEY_MASK_CTRL = 268435456
• KEY_MASK_CMD = 268435456
• KEY_MASK_KPAD = 536870912
• KEY_MASK_GROUP_SWITCH = 1073741824
• BUTTON_LEFT = 1
• BUTTON_RIGHT = 2
• BUTTON_MIDDLE = 3
• BUTTON_WHEEL_UP = 4
• BUTTON_WHEEL_DOWN = 5
• BUTTON_WHEEL_LEFT = 6
• BUTTON_WHEEL_RIGHT = 7
• BUTTON_MASK_LEFT = 1
• BUTTON_MASK_RIGHT = 2
• BUTTON_MASK_MIDDLE = 4
• JOY_BUTTON_0 = 0 — Joystick Button 0
• JOY_BUTTON_1 = 1 — Joystick Button 1
• JOY_BUTTON_2 = 2 — Joystick Button 2
• JOY_BUTTON_3 = 3 — Joystick Button 3
• JOY_BUTTON_4 = 4 — Joystick Button 4
• JOY_BUTTON_5 = 5 — Joystick Button 5
• JOY_BUTTON_6 = 6 — Joystick Button 6
• JOY_BUTTON_7 = 7 — Joystick Button 7
• JOY_BUTTON_8 = 8 — Joystick Button 8
• JOY_BUTTON_9 = 9 — Joystick Button 9
• JOY_BUTTON_10 = 10 — Joystick Button 10
• JOY_BUTTON_11 = 11 — Joystick Button 11
• JOY_BUTTON_12 = 12 — Joystick Button 12
• JOY_BUTTON_13 = 13 — Joystick Button 13
• JOY_BUTTON_14 = 14 — Joystick Button 14
• JOY_BUTTON_15 = 15 — Joystick Button 15
• JOY_BUTTON_MAX = 16 — Joystick Button 16
• JOY_SNES_A = 1
• JOY_SNES_B = 0
• JOY_SNES_X = 3
• JOY_SNES_Y = 2
• JOY_SONY_CIRCLE = 1
• JOY_SONY_X = 0
• JOY_SONY_SQUARE = 2
• JOY_SONY_TRIANGLE = 3
• JOY_SEGA_B = 1
• JOY_SEGA_A = 0
• JOY_SEGA_X = 2
• JOY_SEGA_Y = 3
• JOY_XBOX_B = 1
• JOY_XBOX_A = 0
• JOY_XBOX_X = 2
• JOY_XBOX_Y = 3
• JOY_DS_A = 1
• JOY_DS_B = 0
• JOY_DS_X = 3
• JOY_DS_Y = 2
• JOY_SELECT = 10
• JOY_START = 11
• JOY_DPAD_UP = 12
• JOY_DPAD_DOWN = 13
• JOY_DPAD_LEFT = 14
• JOY_DPAD_RIGHT = 15
• JOY_L = 4
• JOY_L2 = 6
• JOY_L3 = 8
• JOY_R = 5
• JOY_R2 = 7
• JOY_R3 = 9
• JOY_AXIS_0 = 0
• JOY_AXIS_1 = 1
• JOY_AXIS_2 = 2
• JOY_AXIS_3 = 3
• JOY_AXIS_4 = 4
• JOY_AXIS_5 = 5
• JOY_AXIS_6 = 6
• JOY_AXIS_7 = 7
• JOY_AXIS_MAX = 8
• JOY_ANALOG_0_X = 0
• JOY_ANALOG_0_Y = 1
• JOY_ANALOG_1_X = 2
• JOY_ANALOG_1_Y = 3
• JOY_ANALOG_2_X = 4
• JOY_ANALOG_2_Y = 5
• JOY_ANALOG_L2 = 6
• JOY_ANALOG_R2 = 7
• OK = 0 — Functions that return Error return OK when everything went ok. Most functions don’t return error
anyway and/or just print errors to stdout.
• ERR_BUSY = 44
• ERR_HELP = 46
• ERR_BUG = 47
• ERR_WTF = 49
• PROPERTY_HINT_NONE = 0 — No hint for edited property.
• PROPERTY_HINT_RANGE = 1 — Hints that the string is a range, defined as “min,max” or “min,max,step”.
This is valid for integers and floats.
• PROPERTY_HINT_EXP_RANGE = 2 — Hints that the string is an exponential range, defined as “min,max”
or “min,max,step”. This is valid for integers and floats.
• PROPERTY_HINT_ENUM = 3 — Property hint for an enumerated value, like “Hello,Something,Else”. This
is valid for integer, float and string properties.
• PROPERTY_HINT_EXP_EASING = 4
• PROPERTY_HINT_LENGTH = 5
• PROPERTY_HINT_KEY_ACCEL = 7
• PROPERTY_HINT_FLAGS = 8 — Property hint for a bitmask description, for bits 0,1,2,3 and 5 the hint
would be like “Bit0,Bit1,Bit2,Bit3„Bit5”. Valid only for integers.
• PROPERTY_HINT_ALL_FLAGS = 9 — Property hint for a bitmask description that covers all 32 bits. Valid
only for integers.
• PROPERTY_HINT_FILE = 10 — String property is a file (so pop up a file dialog when edited). Hint string
can be a set of wildcards like “*.doc”.
• PROPERTY_HINT_DIR = 11 — String property is a directory (so pop up a file dialog when edited).
• PROPERTY_HINT_GLOBAL_FILE = 12
• PROPERTY_HINT_GLOBAL_DIR = 13
• PROPERTY_HINT_RESOURCE_TYPE = 14 — String property is a resource, so open the resource popup
menu when edited.
• PROPERTY_HINT_MULTILINE_TEXT = 15
• PROPERTY_HINT_COLOR_NO_ALPHA = 16
• PROPERTY_HINT_IMAGE_COMPRESS_LOSSY = 17
• PROPERTY_HINT_IMAGE_COMPRESS_LOSSLESS = 18
• PROPERTY_USAGE_STORAGE = 1 — Property will be used as storage (default).
• PROPERTY_USAGE_STORAGE = 1 — Property will be used as storage (default).
• PROPERTY_USAGE_EDITOR = 2 — Property will be visible in editor (default).
• PROPERTY_USAGE_NETWORK = 4
• PROPERTY_USAGE_DEFAULT = 7 — Default usage (storage and editor).
• METHOD_FLAG_NORMAL = 1
• METHOD_FLAG_EDITOR = 2
• METHOD_FLAG_NOSCRIPT = 4
• METHOD_FLAG_CONST = 8
• METHOD_FLAG_REVERSE = 16
• METHOD_FLAG_VIRTUAL = 32
• METHOD_FLAG_FROM_SCRIPT = 64
• METHOD_FLAGS_DEFAULT = 1
• TYPE_NIL = 0 — Variable is of type nil (only applied for null).
• TYPE_BOOL = 1 — Variable is of type bool.
• TYPE_INT = 2 — Variable is of type integer.
• TYPE_REAL = 3 — Variable is of type float/real.
• TYPE_STRING = 4 — Variable is of type String.
• TYPE_VECTOR2 = 5 — Variable is of type Vector2.
• TYPE_RECT2 = 6 — Variable is of type Rect2.
• TYPE_VECTOR3 = 7 — Variable is of type Vector3.
• TYPE_MATRIX32 = 8 — Variable is of type Matrix32.
• TYPE_PLANE = 9 — Variable is of type Plane.
• TYPE_QUAT = 10 — Variable is of type Quat.
• TYPE_AABB = 11 — Variable is of type AABB.
• TYPE_MATRIX3 = 12 — Variable is fo type Matrix3.
• TYPE_TRANSFORM = 13 — Variable is fo type Transform.
• TYPE_COLOR = 14 — Variable is fo type Color.
• TYPE_IMAGE = 15 — Variable is fo type Image.
• TYPE_NODE_PATH = 16 — Variable is fo type NodePath.
• TYPE_RID = 17 — Variable is fo type RID.
• TYPE_OBJECT = 18 — Variable is fo type Object.
• TYPE_INPUT_EVENT = 19 — Variable is fo type InputEvent.
• TYPE_DICTIONARY = 20 — Variable is fo type Dictionary.
• TYPE_ARRAY = 21 — Variable is fo type Array.
• TYPE_RAW_ARRAY = 22
• TYPE_INT_ARRAY = 23
• TYPE_REAL_ARRAY = 24
• TYPE_STRING_ARRAY = 25
• TYPE_VECTOR2_ARRAY = 26
• TYPE_VECTOR3_ARRAY = 27
• TYPE_COLOR_ARRAY = 28
• TYPE_MAX = 29
9.2.4 Description
Global scope constants and variables. This is all that resides in the globals, constants regarding error codes, scancodes,
property hints, etc. It’s not much.
Singletons are also documented here, since they can be accessed from anywhere.
9.3 AABB
• Vector3 pos
• Vector3 size
• Vector3 end
9.3.4 Description
AABB provides an 3D Axis-Aligned Bounding Box. It consists of a position, a size, and several utility functions. It is
typically used for simple (fast) overlap tests.
Return the intersection between two AABB. An empty AABB (size 0,0,0) is returned on failure.
• bool intersects ( AABB with )
Return true if the AABB overlaps with another.
• bool intersects_plane ( Plane plane )
Return true if the AABB is at both sides of a plane.
• bool intersects_segment ( Vector3 from, Vector3 to )
• AABB merge ( AABB with )
Combine this AABB with another, a larger one is returned that contains both.
• AABB AABB ( Vector3 pos, Vector3 size )
Optional constructor, accepts position and size.
9.4 AcceptDialog
Inherits: WindowDialog < Popup < Control < CanvasItem < Node < Object
Inherited By: ConfirmationDialog
Category: Core
Object get_ok ( )
Object get_label ( )
void set_hide_on_ok ( bool enabled )
bool get_hide_on_ok ( ) const
Button add_button ( String text, bool right=false, String action=”” )
Button add_cancel ( String name )
LineEdit register_text_enter ( Object line_edit )
void set_text ( String text )
String get_text ( ) const
9.4.3 Signals
• confirmed ( )
• custom_action ( String action )
9.4.4 Description
This dialog is useful for small notifications to the user about an event. It can only be accepted or closed, with the same
result.
• Object get_ok ( )
Return the OK Button.
• Object get_label ( )
Return the label used for built-in text.
• void set_hide_on_ok ( bool enabled )
Set whether the dialog is hidden when accepted (default true).
• bool get_hide_on_ok ( ) const
Return true if the dialog will be hidden when accepted (default true).
• Button add_button ( String text, bool right=false, String action=”” )
• Button add_cancel ( String name )
• LineEdit register_text_enter ( Object line_edit )
Register a LineEdit in the dialog. When the enter key is pressed, the dialog will be accepted.
• void set_text ( String text )
Set the built-in label text.
• String get_text ( ) const
Return the built-in label text.
9.5 AnimatedSprite
9.5.3 Signals
• frame_changed ( )
9.5.4 Description
9.6 AnimatedSprite3D
Inherits: SpriteBase3D < GeometryInstance < VisualInstance < Spatial < Node < Object
Category: Core
9.6.3 Signals
• frame_changed ( )
9.7 Animation
• TYPE_VALUE = 0 — Value tracks set values in node properties, but only those which can be Interpolated.
• TYPE_TRANSFORM = 1 — Transform tracks are used to change node local transforms or skeleton pose
bones. Transitions are Interpolated.
• TYPE_METHOD = 2 — Method tracks call functions with given arguments per key.
• INTERPOLATION_NEAREST = 0 — No interpolation (nearest value).
• INTERPOLATION_LINEAR = 1 — Linear interpolation.
• INTERPOLATION_CUBIC = 2 — Cubic interpolation.
9.7.4 Description
An Animation resource contains data used to animate everything in the engine. Animations are divided into tracks,
and each track must be linked to a node. The state of that node can be changed through time, by adding timed keys
(events) to the track.
Animations are just data containers, and must be added to odes such as an AnimationPlayer or AnimationTreePlayer
to be played back.
9.8 AnimationPlayer
9.8.3 Signals
9.8.5 Description
An animation player is used for general purpose playback of Animation resources. It contains a dictionary of anima-
tions (referenced by name) and custom blend times between their transitions. Additionally, animations can be played
and blended in different channels.
Add an animation resource to the player, which will be later referenced by the “name” argument.
• void remove_animation ( String name )
Remove an animation from the player (by supplying the same name used to add it).
• void rename_animation ( String name, String newname )
Rename an existing animation.
• bool has_animation ( String name ) const
Request whether an Animation name exist within the player.
• Animation get_animation ( String name ) const
Get an Animation resource by requesting a name.
• StringArray get_animation_list ( ) const
Get the list of names of the animations stored in the player.
• void set_blend_time ( String anim_from, String anim_to, float sec )
Specify a blend time (in seconds) between two animations, referenced by their names.
• float get_blend_time ( String anim_from, String anim_to ) const
Get the blend time between two animations, referenced by their names.
• void set_default_blend_time ( float sec )
Set the default blend time between animations.
• float get_default_blend_time ( ) const
Return the default blend time between animations.
• void play ( String name=””, float custom_blend=-1, float custom_speed=1, bool from_end=false )
Play a given animation by the animation name. Custom speed and blend times can be set. If custom speed is negative
(-1), ‘from_end’ being true can play the animation backwards.
• void play_backwards ( String name=””, float custom_blend=-1 )
• void stop ( bool reset=true )
• void stop_all ( )
Stop playback of animations (deprecated).
• bool is_playing ( ) const
Return whether an animation is playing.
• void set_current_animation ( String anim )
Set the current animation (even if no playback occurs). Using set_current_animation() and set_active() are similar to
calling play().
• String get_current_animation ( ) const
Return the name of the animation being played.
• void queue ( String name )
Queue an animation for playback once the current one is done.
• void clear_queue ( )
If animations are queued to play, clear them.
9.9 AnimationTreePlayer
• NODE_OUTPUT = 0
• NODE_ANIMATION = 1
• NODE_ONESHOT = 2
• NODE_MIX = 3
• NODE_BLEND2 = 4
• NODE_BLEND3 = 5
• NODE_BLEND4 = 6
• NODE_TIMESCALE = 7
• NODE_TIMESEEK = 8
• NODE_TRANSITION = 9
9.9.4 Description
Animation Player that uses a node graph for the blending. This kind of player is very useful when animating character
or other skeleton based rigs, because it can combine several animations to form a desired pose.
9.10 Area
9.10.3 Signals
9.11 Area2D
Inherits: CollisionObject2D < Node2D < CanvasItem < Node < Object
Category: Core
9.11.3 Signals
9.11.4 Description
General purpose area detection for 2D physics. Areas can be used for detection of objects that enter/exit them, as
well as overriding space parameters (changing gravity, damping, etc). For this, use any space override different from
AREA_SPACE_OVERRIDE_DISABLE and point gravity at the center of mass.
9.12 Array
9.12.3 Description
Generic array, contains several elements of any type, accessible by numerical index starting at 0. Arrays are always
passed by reference.
• bool empty ( )
Return true if the array is empty (size==0).
• void erase ( var value )
Remove the first occurrence of a value from the array.
• int find ( var value )
Searches the array for a value and returns its index or -1 if not found.
• int hash ( )
Return a hashed integer value representing the array contents.
• void insert ( int pos, var value )
Insert a new element at a given position in the array. The position must be valid, or at the end of the array (pos==size()).
• void invert ( )
Reverse the order of the elements in the array (so first element will now be the last).
• bool is_shared ( )
Get whether this is a shared array instance.
• void pop_back ( )
• void pop_front ( )
• void push_back ( var value )
Append an element at the end of the array.
• void push_front ( var value )
• void remove ( int pos )
Remove an element from the array by index.
• void resize ( int pos )
Resize the array to contain a different number of elements. If the array size is smaller, elements are cleared, if bigger,
new elements are Null.
• int size ( )
Return the amount of elements in the array.
• void sort ( )
Sort the array using natural order.
• void sort_custom ( Object obj, String func )
Sort the array using a custom method. The arguments are an object that holds the method and the name of such
method. The custom method receives two arguments (a pair of elements from the array) and must return true if the
first argument is less than the second, and return false otherwise.
• Array Array ( RawArray from )
Construct an array from a RawArray.
• Array Array ( IntArray from )
Construct an array from a RawArray.
9.13 AtlasTexture
9.14 AudioServer
Inherits: Object
Inherited By: AudioServerSW
Category: Core
9.14.4 Description
AudioServer is a low level server interface for audio access. It is in charge of creating sample data (playable audio) as
well as its playback via a voice interface.
9.15 AudioServerSW
9.15.2 Description
This is a software audio server. It does not use any kind of hardware acceleration.
This class does not expose any new method.
9.16 AudioStream
9.16.2 Description
Base class for audio streams. Audio streams are used for music playback, or other types of streamed sounds that don’t
fit or require more flexibility than a Sample.
9.17 AudioStreamMPC
9.17.2 Description
9.18 AudioStreamOGGVorbis
9.18.2 Description
9.19 AudioStreamOpus
9.19.2 Description
9.20 AudioStreamPlayback
9.21 AudioStreamSpeex
9.21.2 Description
Speex audio stream driver. Speex is very useful for compressed speech. It allows loading a very large amount of
speech in memory at little IO/latency cost.
9.22 BackBufferCopy
• COPY_MODE_DISABLED = 0
• COPY_MODE_RECT = 1
• COPY_MODE_VIEWPORT = 2
9.23 BakedLight
• MODE_OCTREE = 0
• MODE_LIGHTMAPS = 1
• BAKE_DIFFUSE = 0
• BAKE_SPECULAR = 1
• BAKE_TRANSLUCENT = 2
• BAKE_CONSERVE_ENERGY = 3
• BAKE_MAX = 5
• void clear_lightmaps ( )
• void set_cell_subdivision ( int cell_subdivision )
• int get_cell_subdivision ( ) const
• void set_initial_lattice_subdiv ( int cell_subdivision )
• int get_initial_lattice_subdiv ( ) const
• void set_plot_size ( float plot_size )
• float get_plot_size ( ) const
• void set_bounces ( int bounces )
• int get_bounces ( ) const
• void set_cell_extra_margin ( float cell_extra_margin )
• float get_cell_extra_margin ( ) const
• void set_edge_damp ( float edge_damp )
• float get_edge_damp ( ) const
• void set_normal_damp ( float normal_damp )
• float get_normal_damp ( ) const
• void set_tint ( float tint )
• float get_tint ( ) const
• void set_saturation ( float saturation )
• float get_saturation ( ) const
• void set_ao_radius ( float ao_radius )
• float get_ao_radius ( ) const
• void set_ao_strength ( float ao_strength )
• float get_ao_strength ( ) const
• void set_format ( int format )
• int get_format ( ) const
• void set_transfer_lightmaps_only_to_uv2 ( bool enable )
• bool get_transfer_lightmaps_only_to_uv2 ( ) const
• void set_energy_multiplier ( float energy_multiplier )
• float get_energy_multiplier ( ) const
• void set_gamma_adjust ( float gamma_adjust )
• float get_gamma_adjust ( ) const
• void set_bake_flag ( int flag, bool enabled )
• bool get_bake_flag ( int flag ) const
9.24 BakedLightInstance
9.24.3 Signals
• baked_light_changed ( )
9.25 BakedLightSampler
• PARAM_RADIUS = 0
• PARAM_STRENGTH = 1
• PARAM_ATTENUATION = 2
• PARAM_DETAIL_RATIO = 3
• PARAM_MAX = 4
9.26 BaseButton
9.26.3 Signals
• released ( )
• toggled ( bool pressed )
• pressed ( )
• DRAW_NORMAL = 0
• DRAW_PRESSED = 1
• DRAW_HOVER = 2
• DRAW_DISABLED = 3
9.26.5 Description
BaseButton is the abstract base class for buttons, so it shouldn’t be used directly (It doesn’t display anything). Other
types of buttons inherit from it.
9.27 BitMap
9.28 BoneAttachment
9.29 bool
9.29.3 Description
9.30 BoxContainer
Inherits: Container < Control < CanvasItem < Node < Object
Inherited By: VBoxContainer, ButtonGroup, HBoxContainer, ColorPicker
Category: Core
• ALIGN_BEGIN = 0
• ALIGN_CENTER = 1
• ALIGN_END = 2
9.30.4 Description
Base class for Box containers. It arranges children controls vertically or horizontally, and rearranges them automati-
cally when their minimum size changes.
9.31 BoxShape
9.31.3 Description
9.32 Button
Inherits: BaseButton < Control < CanvasItem < Node < Object
Inherited By: OptionButton, ColorPickerButton, CheckButton, MenuButton, ToolButton, CheckBox
Category: Core
9.32.3 Description
Button is just the standard themed button: image src=”images/button_example.png”/ It can contain text and an icon,
and will display them according to the current Theme.
9.33 ButtonArray
Category: Core
Array of Buttons.
9.33.3 Signals
9.33.5 Description
Array of Buttons. A Button array is useful to have an array of buttons laid out vertically or horizontally. Only one can
be selected. This is useful for joy pad based interfaces and option menus.
9.34 ButtonGroup
Inherits: BoxContainer < Container < Control < CanvasItem < Node < Object
Category: Core
Group of Buttons.
9.34.3 Description
Group of Button. All direct and indirect children buttons become radios. Only one allows being pressed.
9.35 Camera
9.35.4 Description
Camera is a special node that displays what is visible from its current location. Cameras register themselves in the
nearest Viewport node (when ascending the tree). Only one camera can be active per viewport. If no viewport is
available ascending the tree, the Camera will register in the global viewport. In other words, a Camera just provides
3D display capabilities to a Viewport, and, without one, a Scene registered in that Viewport (or higher viewports) can’t
be displayed.
9.36 Camera2D
• ANCHOR_MODE_DRAG_CENTER = 1
• ANCHOR_MODE_FIXED_TOP_LEFT = 0
9.36.4 Description
Camera node for 2D scenes. It forces the screen (current layer) to scroll following this node. This makes it easier (and
faster) to program scrollable scenes than manually changing the position of CanvasItem based nodes.
This node is intended to be a simple helper get get things going quickly and it may happen often that more functionality
is desired to change how the camera works. To make your own custom camera node, simply inherit from Node2D and
change the transform of the canvas by calling get_viewport().set_canvas_transform(m) in Viewport.
9.37 CanvasItem
9.37.3 Signals
• item_rect_changed ( )
• draw ( )
• visibility_changed ( )
• hide ( )
• BLEND_MODE_MIX = 0 — Mix blending mode. Colors are assumed to be independent of the alpha (opacity)
value.
• BLEND_MODE_ADD = 1 — Additive blending mode.
• BLEND_MODE_SUB = 2 — Substractive blending mode.
• BLEND_MODE_MUL = 3 — Multiplicative blending mode.
• BLEND_MODE_PREMULT_ALPHA = 4 — Mix blending mode. Colors are assumed to be premultiplied
by the alpha (opacity) value.
• NOTIFICATION_DRAW = 30 — CanvasItem is requested to draw.
• NOTIFICATION_VISIBILITY_CHANGED = 31 — Canvas item visibility has changed.
• NOTIFICATION_ENTER_CANVAS = 32 — Canvas item has entered the canvas.
• NOTIFICATION_EXIT_CANVAS = 33 — Canvas item has exited the canvas.
• NOTIFICATION_TRANSFORM_CHANGED = 29 — Canvas item transform has changed. Only received
if requested.
9.37.5 Description
Base class of anything 2D. Canvas items are laid out in a tree and children inherit and extend the transform of their
parent. CanvasItem is extended by Control, for anything GUI related, and by Node2D for anything 2D engine related.
Any CanvasItem can draw. For this, the “update” function must be called, then NOTIFICATION_DRAW will be
received on idle time to request redraw. Because of this, canvas items don’t need to be redraw on every frame,
improving the performance significan’tly. Several functions for drawing on the CanvasItem are provided (see draw_*
functions). They can only be used inside the notification, signal or _draw() overrides function, though.
Canvas items are draw in tree order. By default, children are on top of their parents so a root CanvasItem will be drawn
behind everything (this can be changed per item though).
Canvas items can also be hidden (hiding also their subtree). They provide many means for changing standard param-
eters such as opacity (for it and the subtree) and self opacity, blend mode.
Ultimately, a transform notification can be requested, which will notify the node that its global position changed in
case the parent tree changed.
9.38 CanvasItemMaterial
• SHADING_NORMAL = 0
• SHADING_UNSHADED = 1
• SHADING_ONLY_LIGHT = 2
9.39 CanvasItemShader
9.40 CanvasItemShaderGraph
Inherits: ShaderGraph < Shader < Resource < Reference < Object
Category: Core
9.41 CanvasLayer
9.41.3 Description
Canvas Item layer. CanvasItem nodes that are direct or indirect children of a CanvasLayer will be drawn in that layer.
The layer is a numeric index that defines the draw order. The default 2D scene renders with index 0, so a CanvasLayer
with index -1 will be drawn below, and one with index 1 will be drawn above. This is very useful for HUDs (in layer
1+ or above), or backgrounds (in layer -1 or below).
9.42 CanvasModulate
9.43 CapsuleShape
9.43.3 Description
9.44 CapsuleShape2D
9.44.3 Description
Capsule 2D shape resource for physics. A capsule (or sometimes called “pill”) is like a line grown in all directions. It
has a radius and a height, and is often useful for modeling biped characters.
9.45 CenterContainer
Inherits: Container < Control < CanvasItem < Node < Object
Category: Core
9.45.3 Description
CenterContainer Keeps children controls centered. This container keeps all children to their minimum size, in the
center.
9.46 CheckBox
Inherits: Button < BaseButton < Control < CanvasItem < Node < Object
Category: Core
9.47 CheckButton
Inherits: Button < BaseButton < Control < CanvasItem < Node < Object
Category: Core
Checkable button.
9.47.2 Description
9.48 CircleShape2D
9.48.3 Description
Circular Shape for 2D Physics. This shape is useful for modeling balls or small characters and it’s collision detection
with everything else is very fast.
9.49 CollisionObject
void _input_event ( Object camera, InputEvent event, Vector3 click_pos, Vector3 click_normal, int
shape_idx ) virtual
void add_shape ( Shape shape, Transform transform=Transform() )
int get_shape_count ( ) const
void set_shape ( int shape_idx, Shape shape )
void set_shape_transform ( int shape_idx, Transform transform )
void set_shape_as_trigger ( int shape_idx, bool enable )
bool is_shape_set_as_trigger ( int shape_idx ) const
Shape get_shape ( int shape_idx ) const
Trans- get_shape_transform ( int shape_idx ) const
form
void remove_shape ( int shape_idx )
void clear_shapes ( )
void set_ray_pickable ( bool ray_pickable )
bool is_ray_pickable ( ) const
void set_capture_input_on_drag ( bool enable )
bool get_capture_input_on_drag ( ) const
RID get_rid ( ) const
9.49.3 Signals
• mouse_enter ( )
• input_event ( Object camera, InputEvent event, Vector3 click_pos, Vector3 click_normal, int shape_idx )
• mouse_exit ( )
• void _input_event ( Object camera, InputEvent event, Vector3 click_pos, Vector3 click_normal, int shape_idx )
virtual
• void add_shape ( Shape shape, Transform transform=Transform() )
• int get_shape_count ( ) const
• void set_shape ( int shape_idx, Shape shape )
• void set_shape_transform ( int shape_idx, Transform transform )
• void set_shape_as_trigger ( int shape_idx, bool enable )
• bool is_shape_set_as_trigger ( int shape_idx ) const
• Shape get_shape ( int shape_idx ) const
• Transform get_shape_transform ( int shape_idx ) const
• void remove_shape ( int shape_idx )
• void clear_shapes ( )
• void set_ray_pickable ( bool ray_pickable )
9.50 CollisionObject2D
9.50.3 Signals
• mouse_enter ( )
• input_event ( Object viewport, InputEvent event, int shape_idx )
• mouse_exit ( )
9.50.4 Description
CollisionObject2D is the base class for 2D physics collisionables. They can hold any number of 2D collision shapes.
Usually, they are edited by placing CollisionShape2D and/or CollisionPolygon2D nodes as children. Such nodes are
for reference and not present outside the editor, so code should use the regular shape API.
9.51 CollisionPolygon
9.52 CollisionPolygon2D
9.52.3 Description
Editor-only class. This is not present when running the game. It’s used in the editor to properly edit and position
collision shapes in CollisionObject2D. This is not accessible from regular code. This class is for editing custom shape
polygons.
9.53 CollisionShape
9.54 CollisionShape2D
9.54.3 Description
Editor-only class. This is not present when running the game. It’s used in the editor to properly edit and position
collision shapes in CollisionObject2D. This is not accessible from regular code.
9.55 Color
• float r
• float g
• float b
• float a
• float h
• float s
• float v
• int r8
• int g8
• int b8
• int a8
9.55.4 Description
A color is represented as red, green and blue (r,g,b) components. Additionally, “a” represents the alpha component,
often used for transparency. Values are in floating point and usually range from 0 to 1. Some methods (such as
set_modulate() ) may accept values > 1.
9.56 ColorArray
Array of Colors
9.56.3 Description
Array of Color, can only contains colors. Optimized for memory usage, can’t fragment the memory.
9.57 ColorPicker
Inherits: BoxContainer < Container < Control < CanvasItem < Node < Object
Category: Core
9.57.3 Signals
9.57.4 Description
This is a simple color picker Control. It’s useful for selecting a color from an RGB/RGBA colorspace.
9.58 ColorPickerButton
Inherits: Button < BaseButton < Control < CanvasItem < Node < Object
Category: Core
9.58.3 Signals
9.59 ColorRamp
9.60 ConcavePolygonShape
9.60.3 Description
Concave polygon shape resource, which can be set into a PhysicsBody or area. This shape is created by feeding a list
of triangles.
9.61 ConcavePolygonShape2D
9.61.3 Description
Concave polygon 2D shape resource for physics. It is made out of segments and is very optimal for complex polygonal
concave collisions. It is really not advised to use for RigidBody nodes. A CollisionPolygon2D in convex decomposi-
tion mode (solids) or several convex objects are advised for that instead. Otherwise, a concave polygon 2D shape is
better for static collisions.
The main difference between a ConvexPolygonShape2D and a ConcavePolygonShape2D is that a concave polygon
assumes it is concave and uses a more complex method of collision detection, and a convex one forces itself to be
convex in order to speed up collision detection.
9.62 ConeTwistJoint
• PARAM_SWING_SPAN = 0
• PARAM_TWIST_SPAN = 1
• PARAM_BIAS = 2
• PARAM_SOFTNESS = 3
• PARAM_RELAXATION = 4
• PARAM_MAX = 5
9.63 ConfigFile
9.64 ConfirmationDialog
Inherits: AcceptDialog < WindowDialog < Popup < Control < CanvasItem < Node < Object
Inherited By: EditorFileDialog, FileDialog
Category: Core
Button get_cancel ( )
9.64.3 Description
Dialog for confirmation of actions. This dialog inherits from AcceptDialog, but has by default an OK and Cancel
button (in host OS order).
• Button get_cancel ( )
Return the cancel button.
9.65 Container
void queue_sort ( )
void fit_child_in_rect ( Control child, Rect2 rect )
9.65.3 Signals
• sort_children ( )
9.65.5 Description
Base node for containers. A Container contains other controls and automatically arranges them in a certain way.
A Control can inherit this to create custom container classes.
• void queue_sort ( )
Queue resort of the contained children. This is called automatically anyway, but can be called upon request.
• void fit_child_in_rect ( Control child, Rect2 rect )
Fit a child control in a given rect. This is mainly a helper for creating custom container classes.
9.66 Control
9.66.3 Signals
• focus_enter ( )
• mouse_enter ( )
• resized ( )
• minimum_size_changed ( )
• size_flags_changed ( )
• focus_exit ( )
• input_event ( InputEvent ev )
• modal_close ( )
• mouse_exit ( )
• CURSOR_HSPLIT = 15
• CURSOR_HELP = 16
• SIZE_EXPAND = 1
• SIZE_FILL = 2
• SIZE_EXPAND_FILL = 3
9.66.5 Description
Control is the base class Node for all the GUI components. Every GUI component inherits from it, directly or indi-
rectly. In this way, sections of the scene tree made of contiguous control nodes, become user interfaces.
Controls are relative to the parent position and size by using anchors and margins. This ensures that they can adapt
easily in most situation to changing dialog and screen sizes. When more flexibility is desired, Container derived nodes
can be used.
Anchors work by defining which margin do they follow, and a value relative to it. Allowed anchoring modes are
ANCHOR_BEGIN, where the margin is relative to the top or left margins of the parent (in pixels), ANCHOR_END
for the right and bottom margins of the parent and ANCHOR_RATIO, which is a ratio from 0 to 1 in the parent range.
Input device events (InputEvent) are first sent to the root controls via the Node._input, which distribute it through the
tree, then delivers them to the adequate one (under cursor or keyboard focus based) by calling Node._input_event.
There is no need to enable input processing on controls to receive such events. To ensure that no one else will receive
the event (not even Node._unhandled_input), the control can accept it by calling accept_event.
Only one control can hold the keyboard focus (receiving keyboard events), for that the control must define the focus
mode with set_focus_mode. Focus is lost when another control gains it, or the current focus owner is hidden.
It is sometimes desired for a control to ignore mouse/pointer events. This is often the case when placing other controls
on top of a button, in such cases. Calling set_ignore_mouse enables this function.
Finally, controls are skinned according to a Theme. Setting a Theme on a control will propagate all the skinning down
the tree. Optionally, skinning can be overrided per each control by calling the add_*_override functions, or from the
editor.
Return the minimum size this Control can shrink to. A control will never be displayed or resized smaller than its
minimum size.
• Vector2 get_combined_minimum_size ( ) const
• void set_anchor ( int margin, int anchor_mode )
Change the anchor (ANCHOR_BEGIN, ANCHOR_END, ANCHOR_RATIO) type for a margin (MARGIN_LEFT,
MARGIN_TOP, MARGIN_RIGHT, MARGIN_BOTTOM). Changing the anchor mode converts the current margin
offset from the previous anchor mode to the new one, so margin offsets (set_margin) must be done after setting anchors,
or at the same time (set_anchor_and_margin).
• int get_anchor ( int margin ) const
Return the anchor type (ANCHOR_BEGIN, ANCHOR_END, ANCHOR_RATIO) for a given margin (MAR-
GIN_LEFT, MARGIN_TOP, MARGIN_RIGHT, MARGIN_BOTTOM).
• void set_margin ( int margin, float offset )
Set a margin offset. Margin can be one of (MARGIN_LEFT, MARGIN_TOP, MARGIN_RIGHT, MAR-
GIN_BOTTOM). Offset value being set depends on the anchor mode.
• void set_anchor_and_margin ( int margin, int anchor_mode, float offset )
Change the anchor (ANCHOR_BEGIN, ANCHOR_END, ANCHOR_RATIO) type for a margin (MARGIN_LEFT,
MARGIN_TOP, MARGIN_RIGHT, MARGIN_BOTTOM), and also set its offset. This is a helper (see set_anchor
and set_margin).
• void set_begin ( Vector2 pos )
Sets MARGIN_LEFT and MARGIN_TOP at the same time. This is a helper (see set_margin).
• void set_end ( Vector2 pos )
Sets MARGIN_RIGHT and MARGIN_BOTTOM at the same time. This is a helper (see set_margin).
• void set_pos ( Vector2 pos )
Move the Control to a new position, relative to the top-left corner of the parent Control, changing all margins if needed
and without changing current anchor mode. This is a helper (see set_margin).
• void set_size ( Vector2 size )
Changes MARGIN_RIGHT and MARGIN_BOTTOM to fit a given size. This is a helper (see set_margin).
• void set_custom_minimum_size ( Vector2 size )
• void set_global_pos ( Vector2 pos )
Move the Control to a new position, relative to the top-left corner of the window Control, and without changing current
anchor mode. (see set_margin).
• void set_rotation ( float rotation )
• void set_scale ( Vector2 scale )
• float get_margin ( int margin ) const
Return a margin offset. Margin can be one of (MARGIN_LEFT, MARGIN_TOP, MARGIN_RIGHT, MAR-
GIN_BOTTOM). Offset value being returned depends on the anchor mode.
• Vector2 get_begin ( ) const
• Vector2 get_end ( ) const
Returns MARGIN_LEFT and MARGIN_TOP at the same time. This is a helper (see set_margin).
• Vector2 get_pos ( ) const
Returns the Control position, relative to the top-left corner of the parent Control and independent of the anchor mode.
• Vector2 get_size ( ) const
Returns the size of the Control, computed from all margins, however the size returned will never be smaller than the
minimum size reported by :ref:‘get_minimum_size<class_Control_get_minimum_size>‘. This means that even
if end position of the Control rectangle is smaller than the begin position, the Control will still display and interact
correctly. (see description, get_minimum_size, set_margin, set_anchor).
• float get_rotation ( ) const
• Vector2 get_scale ( ) const
• Vector2 get_custom_minimum_size ( ) const
• Vector2 get_parent_area_size ( ) const
• Vector2 get_global_pos ( ) const
Returns the Control position, relative to the top-left corner of the parent Control and independent of the anchor mode.
• Rect2 get_rect ( ) const
Return position and size of the Control, relative to the top-left corner of the parent Control. This is a helper (see
get_pos, get_size).
• Rect2 get_global_rect ( ) const
Return position and size of the Control, relative to the top-left corner of the window Control. This is a helper (see
get_global_pos, get_size).
• void set_area_as_parent_rect ( int margin=0 )
Change all margins and anchors, so this Control always takes up the same area as the parent Control. This is a helper
(see set_anchor, set_margin).
• void show_modal ( bool exclusive=false )
Display a Control as modal. Control must be a subwindow (see set_as_subwindow). Modal controls capture the input
signals until closed or the area outside them is accessed. When a modal control loses focus, or the ESC key is pressed,
they automatically hide. Modal controls are used extensively for popup dialogs and menus.
• void set_focus_mode ( int mode )
Set the focus access mode for the control (FOCUS_NONE, FOCUS_CLICK, FOCUS_ALL). Only one Control can
be focused at the same time, and it will receive keyboard signals.
• bool has_focus ( ) const
Return whether the Control is the current focused control (see set_focus_mode).
• void grab_focus ( )
Steal the focus from another control and become the focused control (see set_focus_mode).
• void release_focus ( )
Give up the focus, no other control will be able to receive keyboard input.
• Control get_focus_owner ( ) const
Return which control is owning the keyboard focus, or null if no one.
• void set_h_size_flags ( int flags )
Hint for containers, set horizontal positioning flags.
• int get_h_size_flags ( ) const
Set the default cursor shape for this control. See enum CURSOR_* for the list of shapes.
• int get_default_cursor_shape ( ) const
Return the default cursor shape for this control. See enum CURSOR_* for the list of shapes.
• int get_cursor_shape ( Vector2 pos=Vector2(0,0) ) const
Return the cursor shape at a certain position in the control.
• void set_focus_neighbour ( int margin, NodePath neighbour )
Force a neighbour for moving the input focus to. When pressing TAB or directional/joypad directions focus is moved
to the next control in that direction. However, the neighbour to move to can be forced with this function.
• NodePath get_focus_neighbour ( int margin ) const
Return the forced neighbour for moving the input focus to. When pressing TAB or directional/joypad directions focus
is moved to the next control in that direction. However, the neighbour to move to can be forced with this function.
• void set_ignore_mouse ( bool ignore )
Ignore mouse events on this control (even touchpad events send mouse events).
• bool is_ignoring_mouse ( ) const
Return if the control is ignoring mouse events (even touchpad events send mouse events).
• void force_drag ( var data, Object preview )
• void set_stop_mouse ( bool stop )
• bool is_stopping_mouse ( ) const
• void grab_click_focus ( )
• void set_drag_preview ( Control control )
• void warp_mouse ( Vector2 to_pos )
9.67 ConvexPolygonShape
9.67.3 Description
Convex polygon shape resource, which can be set into a PhysicsBody or area.
9.68 ConvexPolygonShape2D
9.68.3 Description
Convex Polygon Shape for 2D physics. A convex polygon, whatever its shape, is internally decomposed into as many
convex polygons as needed to ensure all collision checks against it are always done on convex polygons (which are
faster to check).
The main difference between a ConvexPolygonShape2D and a ConcavePolygonShape2D is that a concave polygon
assumes it is concave and uses a more complex method of collision detection, and a convex one forces itself to be
convex in order to speed up collision detection.
9.69 CubeMap
• STORAGE_RAW = 0
• STORAGE_COMPRESS_LOSSY = 1
• STORAGE_COMPRESS_LOSSLESS = 2
• SIDE_LEFT = 0
• SIDE_RIGHT = 1
• SIDE_BOTTOM = 2
• SIDE_TOP = 3
• SIDE_FRONT = 4
• SIDE_BACK = 5
• FLAG_MIPMAPS = 1
• FLAG_REPEAT = 2
• FLAG_FILTER = 4
• FLAGS_DEFAULT = 7
9.70 Curve2D
9.70.3 Description
This class describes a Bezier curve in 2D space. It is mainly used to give a shape to a Path2D, but can be manually
sampled for other purposes.
It keeps a cache of precalculated points along the curve, to speed further calculations up.
Adds a point to a curve, at position “pos”, with control points “in” and “out”.
If “atpos” is given, the point is inserted before the point number “atpos”, moving that point (and every point after)
after the inserted point. If “atpos” is not given, or is an illegal value (atpos <0 or atpos >= get_point_count), the point
will be appended at the end of the point list.
• void set_point_pos ( int idx, Vector2 pos )
Sets the position for the vertex “idx”. If the index is out of bounds, the function sends an error to the console.
• Vector2 get_point_pos ( int idx ) const
Returns the position of the vertex “idx”. If the index is out of bounds, the function sends an error to the console, and
returns (0, 0).
• void set_point_in ( int idx, Vector2 pos )
Sets the position of the control point leading to the vertex “idx”. If the index is out of bounds, the function sends an
error to the console.
• Vector2 get_point_in ( int idx ) const
Returns the position of the control point leading to the vertex “idx”. If the index is out of bounds, the function sends
an error to the console, and returns (0, 0).
• void set_point_out ( int idx, Vector2 pos )
Sets the position of the control point leading out of the vertex “idx”. If the index is out of bounds, the function sends
an error to the console.
• Vector2 get_point_out ( int idx ) const
Returns the position of the control point leading out of the vertex “idx”. If the index is out of bounds, the function
sends an error to the console, and returns (0, 0).
• void remove_point ( int idx )
Deletes the point “idx” from the curve. Sends an error to the console if “idx” is out of bounds.
• Vector2 interpolate ( int idx, float t ) const
Returns the position between the vertex “idx” and the vertex “idx”+1, where “t” controls if the point is the first vertex
(t = 0.0), the last vertex (t = 1.0), or in between. Values of “t” outside the range (0.0 >= t <=1) give strange, but
predictable results.
If “idx” is out of bounds it is truncated to the first or last vertex, and “t” is ignored. If the curve has no points, the
function sends an error to the console, and returns (0, 0).
• Vector2 interpolatef ( float fofs ) const
Returns the position at the vertex “fofs”. It calls interpolate using the integer part of fofs as “idx”, and its fractional
part as “t”.
• void set_bake_interval ( float distance )
Sets the distance in pixels between two adjacent cached points. Changing it forces the cache to be recomputed the
next time a xxx_baked_xxx function is called. The less distance, the more points the cache will have, and the more
memory it will consume, so use with care.
• float get_bake_interval ( ) const
Returns the distance between two adjacent cached points.
• float get_baked_length ( ) const
Returns the total length of the curve, based on the cached points. Given enough density (see set_bake_interval), it
should be approximate enough.
9.71 Curve3D
9.71.3 Description
This class describes a Bezier curve in 3D space. It is mainly used to give a shape to a Path, but can be manually
sampled for other purposes.
It keeps a cache of precalculated points along the curve, to speed further calculations up.
Sets the tilt angle in radians for the point “idx”. If the index is out of bounds, the function sends an error to the console.
The tilt controls the rotation along the look-at axis an object traveling the path would have. In the case of a curve
controlling a PathFollow, this tilt is an offset over the natural tilt the PathFollow calculates.
• float get_point_tilt ( int idx ) const
Returns the tilt angle in radians for the point “idx”. If the index is out of bounds, the function sends an error to the
console, and returns 0.
• void set_point_in ( int idx, Vector3 pos )
Sets the position of the control point leading to the vertex “idx”. If the index is out of bounds, the function sends an
error to the console.
• Vector3 get_point_in ( int idx ) const
Returns the position of the control point leading to the vertex “idx”. If the index is out of bounds, the function sends
an error to the console, and returns (0, 0, 0).
• void set_point_out ( int idx, Vector3 pos )
Sets the position of the control point leading out of the vertex “idx”. If the index is out of bounds, the function sends
an error to the console.
• Vector3 get_point_out ( int idx ) const
Returns the position of the control point leading out of the vertex “idx”. If the index is out of bounds, the function
sends an error to the console, and returns (0, 0, 0).
• void remove_point ( int idx )
Deletes the point “idx” from the curve. Sends an error to the console if “idx” is out of bounds.
• Vector3 interpolate ( int idx, float t ) const
Returns the position between the vertex “idx” and the vertex “idx”+1, where “t” controls if the point is the first vertex
(t = 0.0), the last vertex (t = 1.0), or in between. Values of “t” outside the range (0.0 >= t <=1) give strange, but
predictable results.
If “idx” is out of bounds it is truncated to the first or last vertex, and “t” is ignored. If the curve has no points, the
function sends an error to the console, and returns (0, 0, 0).
• Vector3 interpolatef ( float fofs ) const
Returns the position at the vertex “fofs”. It calls interpolate using the integer part of fofs as “idx”, and its fractional
part as “t”.
• void set_bake_interval ( float distance )
Sets the distance in 3D units between two adjacent cached points. Changing it forces the cache to be recomputed the
next time a xxx_baked_xxx function is called. The less distance, the more points the cache will have, and the more
memory it will consume, so use with care.
• float get_bake_interval ( ) const
Returns the distance between two adjacent cached points.
• float get_baked_length ( ) const
Returns the total length of the curve, based on the cached points. Given enough density (see set_bake_interval), it
should be approximate enough.
• Vector3 interpolate_baked ( float offset, bool cubic=false ) const
Returns a point within the curve at position “offset”, where “offset” is measured as a distance in 3D units along the
curve.
To do that, it finds the two cached points where the “offset” lies between, then interpolates the values. This interpola-
tion is cubic if “cubic” is set to true, or linear if set to false.
Cubic interpolation tends to follow the curves better, but linear is faster (and often, precise enough).
• Vector3Array get_baked_points ( ) const
Returns the cache of points as a Vector3Array.
• RealArray get_baked_tilts ( ) const
Returns the cache of tilts as a RealArray.
• Vector3Array tesselate ( int max_stages=5, float tolerance_degrees=4 ) const
Returns a list of points along the curve, with a curvature controlled point density. That is, the curvier parts will have
more points than the straighter parts.
This approximation makes straight segments between each point, then subdivides those segments until the resulting
shape is similar enough.
“max_stages” controls how many subdivisions a curve segment may face before it is considered approximate enough.
Each subdivision splits the segment in half, so the default 5 stages may mean up to 32 subdivisions per curve segment.
Increase with care!
“tolerance_degrees” controls how many degrees the midpoint of a segment may deviate from the real curve, before
the segment has to be subdivided.
9.72 DampedSpringJoint2D
Inherits: Joint2D < Node2D < CanvasItem < Node < Object
Category: Core
9.72.3 Description
Damped spring constraint for 2D physics. This resembles a spring joint that always want to go back to a given length.
9.73 Dictionary
Dictionary type.
void clear ( )
bool empty ( )
void erase ( var value )
bool has ( var value )
int hash ( )
Array keys ( )
int parse_json ( String json )
int size ( )
String to_json ( )
9.73.3 Description
Dictionary type. Associative container which contains values referenced by unique keys. Dictionaries are always
passed by reference.
• void clear ( )
Clear the dictionary, removing all key/value pairs.
• bool empty ( )
Return true if the dictionary is empty.
• void erase ( var value )
Erase a dictionary key/value pair by key.
• bool has ( var value )
Return true if the dictionary has a given key.
• int hash ( )
Return a hashed integer value representing the dictionary contents.
• Array keys ( )
Return the list of keys in the dictionary.
• int parse_json ( String json )
• int size ( )
Return the size of the dictionary (in pairs).
• String to_json ( )
9.74 DirectionalLight
Inherits: Light < VisualInstance < Spatial < Node < Object
Category: Core
• SHADOW_ORTHOGONAL = 0
• SHADOW_PERSPECTIVE = 1
• SHADOW_PARALLEL_2_SPLITS = 2
• SHADOW_PARALLEL_4_SPLITS = 3
• SHADOW_PARAM_MAX_DISTANCE = 0
• SHADOW_PARAM_PSSM_SPLIT_WEIGHT = 1
• SHADOW_PARAM_PSSM_ZOFFSET_SCALE = 2
9.74.4 Description
A DirectionalLight is a type of Light node that emits light constantly in one direction (the negative z axis of the node).
It is used lights with strong intensity that are located far away from the scene to model sunlight or moonlight. The
worldpace location of the DirectionalLight transform (origin) is ignored, only the basis is used do determine light
direction.
9.75 Directory
9.76 EditorFileDialog
Inherits: ConfirmationDialog < AcceptDialog < WindowDialog < Popup < Control < CanvasItem < Node < Object
Category: Core
void clear_filters ( )
void add_filter ( String filter )
String get_current_dir ( ) const
String get_current_file ( ) const
String get_current_path ( ) const
void set_current_dir ( String dir )
void set_current_file ( String file )
void set_current_path ( String path )
void set_mode ( int mode )
int get_mode ( ) const
VBoxContainer get_vbox ( )
void set_access ( int access )
int get_access ( ) const
void set_show_hidden_files ( bool show )
bool is_showing_hidden_files ( ) const
void set_display_mode ( int mode )
int get_display_mode ( ) const
void invalidate ( )
9.76.3 Signals
• MODE_OPEN_FILE = 0
• MODE_OPEN_FILES = 1
• MODE_OPEN_DIR = 2
• MODE_SAVE_FILE = 3
• ACCESS_RESOURCES = 0
• ACCESS_USERDATA = 1
• ACCESS_FILESYSTEM = 2
• void clear_filters ( )
• void add_filter ( String filter )
• String get_current_dir ( ) const
• String get_current_file ( ) const
• String get_current_path ( ) const
• void set_current_dir ( String dir )
• void set_current_file ( String file )
• void set_current_path ( String path )
• void set_mode ( int mode )
• int get_mode ( ) const
• VBoxContainer get_vbox ( )
• void set_access ( int access )
• int get_access ( ) const
• void set_show_hidden_files ( bool show )
• bool is_showing_hidden_files ( ) const
• void set_display_mode ( int mode )
• int get_display_mode ( ) const
• void invalidate ( )
9.77 EditorImportPlugin
9.78 EditorPlugin
• CONTAINER_TOOLBAR = 0
• CONTAINER_SPATIAL_EDITOR_MENU = 1
• CONTAINER_SPATIAL_EDITOR_SIDE = 2
• CONTAINER_SPATIAL_EDITOR_BOTTOM = 3
• CONTAINER_CANVAS_EDITOR_MENU = 4
• CONTAINER_CANVAS_EDITOR_SIDE = 5
9.79 EditorScenePostImport
9.80 EditorScript
9.81 Environment
• BG_KEEP = 0
• BG_DEFAULT_COLOR = 1
• BG_COLOR = 2
• BG_TEXTURE = 3
• BG_CUBEMAP = 4
• BG_CANVAS = 5
• BG_MAX = 6
• BG_PARAM_CANVAS_MAX_LAYER = 0
• BG_PARAM_COLOR = 1
• BG_PARAM_TEXTURE = 2
• BG_PARAM_CUBEMAP = 3
• BG_PARAM_ENERGY = 4
• BG_PARAM_GLOW = 6
• BG_PARAM_MAX = 7
• FX_AMBIENT_LIGHT = 0
• FX_FXAA = 1
• FX_GLOW = 2
• FX_DOF_BLUR = 3
• FX_HDR = 4
• FX_FOG = 5
• FX_BCS = 6
• FX_SRGB = 7
• FX_MAX = 8
• FX_BLUR_BLEND_MODE_ADDITIVE = 0
• FX_BLUR_BLEND_MODE_SCREEN = 1
• FX_BLUR_BLEND_MODE_SOFTLIGHT = 2
• FX_HDR_TONE_MAPPER_LINEAR = 0
• FX_HDR_TONE_MAPPER_LOG = 1
• FX_HDR_TONE_MAPPER_REINHARDT = 2
• FX_HDR_TONE_MAPPER_REINHARDT_AUTOWHITE = 3
• FX_PARAM_AMBIENT_LIGHT_COLOR = 0
• FX_PARAM_AMBIENT_LIGHT_ENERGY = 1
• FX_PARAM_GLOW_BLUR_PASSES = 2
• FX_PARAM_GLOW_BLUR_SCALE = 3
• FX_PARAM_GLOW_BLUR_STRENGTH = 4
• FX_PARAM_GLOW_BLUR_BLEND_MODE = 5
• FX_PARAM_GLOW_BLOOM = 6
• FX_PARAM_GLOW_BLOOM_TRESHOLD = 7
• FX_PARAM_DOF_BLUR_PASSES = 8
• FX_PARAM_DOF_BLUR_BEGIN = 9
• FX_PARAM_DOF_BLUR_RANGE = 10
• FX_PARAM_HDR_TONEMAPPER = 11
• FX_PARAM_HDR_EXPOSURE = 12
• FX_PARAM_HDR_WHITE = 13
• FX_PARAM_HDR_GLOW_TRESHOLD = 14
• FX_PARAM_HDR_GLOW_SCALE = 15
• FX_PARAM_HDR_MIN_LUMINANCE = 16
• FX_PARAM_HDR_MAX_LUMINANCE = 17
• FX_PARAM_HDR_EXPOSURE_ADJUST_SPEED = 18
• FX_PARAM_FOG_BEGIN = 19
• FX_PARAM_FOG_ATTENUATION = 22
• FX_PARAM_FOG_BEGIN_COLOR = 20
• FX_PARAM_FOG_END_COLOR = 21
• FX_PARAM_FOG_BG = 23
• FX_PARAM_BCS_BRIGHTNESS = 24
• FX_PARAM_BCS_CONTRAST = 25
• FX_PARAM_BCS_SATURATION = 26
• FX_PARAM_MAX = 27
9.82 EventPlayer
9.82.3 Description
Class for event stream playback. Event streams are music expressed as a series of events (note on, note off, instrument
change...), as opposed to audio streams, which are just audio data. Examples of event-based streams are MIDI files, or
MOD music.
Currently, only MOD, S3M, IT, and XM music is supported.
Return the playback position. May be in seconds, but depends on the stream type.
• void seek_pos ( float time )
Set the playback position. May be in seconds, but depends on the stream type.
• float get_length ( ) const
Return the song length. May be in seconds, but depends on the stream type.
• void set_autoplay ( bool enabled )
Set whether this player will start playing as soon as it enters the scene tree.
• bool has_autoplay ( ) const
Return whether this player will start playing as soon as it enters the scene tree.
• void set_channel_volume ( int channel, float channel_volume )
Set the volume scale for an individual channel of the stream, with the same value range as set_volume. The channel
number depends on the stream format. For example, MIDIs range from 0 to 15, and MODs from 0 to 63.
Many stream formats are multichannel, so this allows to affect only a part of the music.
• float get_channel_volume ( int channel ) const
Return the volume scale for an individual channel of the stream.
• float get_channel_last_note_time ( int channel ) const
Return the time at which the last note of a given channel in the stream plays.
9.83 EventStream
9.83.2 Description
Base class for all event-based stream drivers. Event streams are music expressed as a series of events (note on, note
off, instrument change...), as opposed to audio streams, which are just audio data. Examples of event-based streams
are MIDI files, of MOD music.
This class exposes no methods.
9.84 EventStreamChibi
9.84.2 Description
This driver plays MOD music. MOD music, as all event-based streams, is a music format defined by note events
occurring at defined moments, instead of a stream of audio samples.
Currently, this driver supports the MOD, S3M, IT, and XM formats.
This class exposes no methods.
This class can return its playback position in seconds, but does not allow to set it, failing with only a console warning.
This class can not return its song length, returning 1.0 when queried.
This class does not limit its volume settings, allowing for overflow/distortion and wave inversion.
9.85 File
• READ = 1
• WRITE = 2
• READ_WRITE = 3
• WRITE_READ = 7
9.86 FileDialog
Inherits: ConfirmationDialog < AcceptDialog < WindowDialog < Popup < Control < CanvasItem < Node < Object
Category: Core
void clear_filters ( )
void add_filter ( String filter )
String get_current_dir ( ) const
String get_current_file ( ) const
String get_current_path ( ) const
void set_current_dir ( String dir )
void set_current_file ( String file )
void set_current_path ( String path )
void set_mode ( int mode )
int get_mode ( ) const
VBoxContainer get_vbox ( )
void set_access ( int access )
int get_access ( ) const
void set_show_hidden_files ( bool show )
bool is_showing_hidden_files ( ) const
void invalidate ( )
9.86.3 Signals
• MODE_OPEN_FILE = 0 — The dialog allows the selection of one, and only one file.
• MODE_OPEN_FILES = 1 — The dialog allows the selection of multiple files.
• MODE_OPEN_DIR = 2 — The dialog functions as a folder selector, disallowing the selection of any file.
• MODE_SAVE_FILE = 3 — The dialog will warn when a file exists.
• ACCESS_RESOURCES = 0
• ACCESS_USERDATA = 1
• ACCESS_FILESYSTEM = 2
9.86.5 Description
FileDialog is a preset dialog used to choose files and directories in the filesystem. It supports filter masks.
• void clear_filters ( )
Clear all the added filters in the dialog.
• void add_filter ( String filter )
Add a custom filter. Filter format is: “mask ; description”, example (C++): dialog->add_filter(“*.png ; PNG Images”);
9.87 FixedMaterial
9.87.4 Description
FixedMaterial is a simple type of material Resource, which contains a fixed amount of parameters. It is the only type
of material supported in fixed-pipeline devices and APIs. It is also an often a better alternative to ShaderMaterial for
most simple use cases.
9.88 float
9.89 Font
9.89.3 Description
Font contains an unicode compatible character set, as well as the ability to draw it with variable width, ascent, descent
and kerning. For creating fonts from TTF files (or other font formats), see the editor support for fonts. TODO check
wikipedia for graph of ascent/baseline/descent/height/etc.
• float draw_char ( RID canvas_item, Vector2 pos, int char, int next=-1, Color modulate=Color(1,1,1,1) ) const
Draw character “char” into a canvas item using the font at a given “pos” position, with “modulate” color, and optionally
kerning if “next” is apassed. clipping the width. “pos” specifies the baseline, not the top. To draw from the top, ascent
must be added to the Y axis. The width used by the character is returned, making this function useful for drawing
strings character by character.
• void set_fallback ( Object fallback )
• Object get_fallback ( ) const
9.90 FuncRef
void call_func ( var arg0=NULL, var arg1=NULL, var arg2=NULL, var arg3=NULL, var arg4=NULL, var
arg5=NULL, var arg6=NULL, var arg7=NULL, var arg8=NULL, var arg9=NULL )
void set_instance ( Object instance )
void set_function ( String name )
• void call_func ( var arg0=NULL, var arg1=NULL, var arg2=NULL, var arg3=NULL, var arg4=NULL, var
arg5=NULL, var arg6=NULL, var arg7=NULL, var arg8=NULL, var arg9=NULL )
• void set_instance ( Object instance )
• void set_function ( String name )
9.91 GDFunctionState
9.92 GDNativeClass
void new ( )
• void new ( )
9.93 GDScript
void new ( )
RawArray get_as_byte_code ( ) const
• void new ( )
• RawArray get_as_byte_code ( ) const
9.94 Generic6DOFJoint
• PARAM_LINEAR_LOWER_LIMIT = 0
• PARAM_LINEAR_UPPER_LIMIT = 1
• PARAM_LINEAR_LIMIT_SOFTNESS = 2
• PARAM_LINEAR_RESTITUTION = 3
• PARAM_LINEAR_DAMPING = 4
• PARAM_ANGULAR_LOWER_LIMIT = 5
• PARAM_ANGULAR_UPPER_LIMIT = 6
• PARAM_ANGULAR_LIMIT_SOFTNESS = 7
• PARAM_ANGULAR_DAMPING = 8
• PARAM_ANGULAR_RESTITUTION = 9
• PARAM_ANGULAR_FORCE_LIMIT = 10
• PARAM_ANGULAR_ERP = 11
• PARAM_ANGULAR_MOTOR_TARGET_VELOCITY = 12
• PARAM_ANGULAR_MOTOR_FORCE_LIMIT = 13
• PARAM_MAX = 14
• FLAG_ENABLE_LINEAR_LIMIT = 0
• FLAG_ENABLE_ANGULAR_LIMIT = 1
• FLAG_ENABLE_MOTOR = 2
• FLAG_MAX = 3
9.95 Geometry
Inherits: Object
Category: Core
9.96 GeometryInstance
• FLAG_VISIBLE = 0
• FLAG_CAST_SHADOW = 3
• FLAG_RECEIVE_SHADOWS = 4
• FLAG_BILLBOARD = 1
• FLAG_BILLBOARD_FIX_Y = 2
• FLAG_DEPH_SCALE = 5
• FLAG_VISIBLE_IN_ALL_ROOMS = 6
• FLAG_MAX = 8
9.96.4 Description
Base node for geometry based visual instances. Shares some common functionality like visibility and custom materi-
als.
9.97 Globals
Inherits: Object
Category: Core
9.97.3 Description
Contains global variables accessible from everywhere. Use the normal Object API, such as “Globals.get(variable)”,
“Globals.set(variable,value)” or “Globals.has(variable)” to access them. Variables stored in engine.cfg are also loaded
into globals, making this object very useful for reading custom game configuration options.
9.98 GraphEdit
GraphEdit is an area capable of showing various GraphNodes. It manages connection events between them.
Error connect_node ( String from, int from_port, String to, int to_port )
bool is_node_connected ( String from, int from_port, String to, int to_port )
void disconnect_node ( String from, int from_port, String to, int to_port )
Array get_connection_list ( ) const
Vector2 get_scroll_ofs ( ) const
void set_zoom ( float p_zoom )
float get_zoom ( ) const
void set_right_disconnects ( bool enable )
bool is_right_disconnects_enabled ( ) const
9.98.3 Signals
• delete_nodes_request ( )
• duplicate_nodes_request ( )
• popup_request ( Vector2 p_position )
• _begin_node_move ( )
• disconnection_request ( String from, int from_slot, String to, int to_slot )
• connection_request ( String from, int from_slot, String to, int to_slot )
• _end_node_move ( )
9.98.4 Description
GraphEdit manages the showing of GraphNodes it contains, as well as connections an disconnections between them.
Signals are sent for each of these two events. Disconnection between GraphNodes slots is disabled by default.
It is greatly advised to enable low processor usage mode (see OS.set_low_processor_usage_mode) when using
GraphEdits.
• Error connect_node ( String from, int from_port, String to, int to_port )
Create a connection between ‘from_port’ slot of ‘from’ GraphNode and ‘to_port’ slot of ‘to’ GraphNode. If the
connection already exists, no connection is created.
• bool is_node_connected ( String from, int from_port, String to, int to_port )
Return true if the ‘from_port’ slot of ‘from’ GraphNode is connected to the ‘to_port’ slot of ‘to’ GraphNode.
• void disconnect_node ( String from, int from_port, String to, int to_port )
Remove the connection between ‘from_port’ slot of ‘from’ GraphNode and ‘to_port’ slot of ‘to’ GraphNode, if con-
nection exists.
• Array get_connection_list ( ) const
Return an Array containing the list of connections. A connection consists in a structure of the form {from_slot: 0,
from: “GraphNode name 0”, to_slot: 1, to: “GraphNode name 1” }
• Vector2 get_scroll_ofs ( ) const
• void set_zoom ( float p_zoom )
• float get_zoom ( ) const
• void set_right_disconnects ( bool enable )
Enable the disconnection of existing connections in the visual GraphEdit by left-clicking a connection and releasing
into the void.
• bool is_right_disconnects_enabled ( ) const
Return true is the disconnection of connections is enable in the visual GraphEdit. False otherwise.
9.99 GraphNode
Inherits: Container < Control < CanvasItem < Node < Object
Category: Core
A GraphNode is a container with several input and output slots allowing connections between GraphNodes. Slots can
have different, incompatible types.
9.99.3 Signals
• raise_request ( )
• close_request ( )
• dragged ( Vector2 from, Vector2 to )
• offset_changed ( )
9.99.4 Description
A GraphNode is a container defined by a title. It can have 1 or more input and output slots, which can be enabled
(shown) or disabled (not shown) and have different (incompatible) types. Colors can also be assigned to slots. A tuple
of input and output slots is defined for each GUI element included in the GraphNode. Input and output connections
are left and right slots, but only enabled slots are counted as connections.
9.100 GridContainer
Inherits: Container < Control < CanvasItem < Node < Object
Category: Core
9.101 GridMap
• INVALID_CELL_ITEM = -1
9.102 GrooveJoint2D
Inherits: Joint2D < Node2D < CanvasItem < Node < Object
Category: Core
9.102.3 Description
Groove constraint for 2D physics. This is useful for making a body “slide” through a segment placed in another.
9.103 HBoxContainer
Inherits: BoxContainer < Container < Control < CanvasItem < Node < Object
Category: Core
9.103.2 Description
9.104 HButtonArray
Inherits: ButtonArray < Control < CanvasItem < Node < Object
Category: Core
9.104.2 Description
9.105 HingeJoint
• PARAM_BIAS = 0
• PARAM_LIMIT_UPPER = 1
• PARAM_LIMIT_LOWER = 2
• PARAM_LIMIT_BIAS = 3
• PARAM_LIMIT_SOFTNESS = 4
• PARAM_LIMIT_RELAXATION = 5
• PARAM_MOTOR_TARGET_VELOCITY = 6
• PARAM_MOTOR_MAX_IMPULSE = 7
• PARAM_MAX = 8
• FLAG_USE_LIMIT = 0
• FLAG_ENABLE_MOTOR = 1
• FLAG_MAX = 2
9.106 HScrollBar
Inherits: ScrollBar < Range < Control < CanvasItem < Node < Object
Category: Core
9.106.2 Description
Horizontal scroll bar. See ScrollBar. This one goes from left (min) to right (max).
9.107 HSeparator
Inherits: Separator < Control < CanvasItem < Node < Object
Category: Core
Horizontal separator.
9.107.2 Description
Horizontal separator. See Separator. It is used to separate objects vertiacally, though (but it looks horizontal!).
9.108 HSlider
Inherits: Slider < Range < Control < CanvasItem < Node < Object
Category: Core
Horizontal slider.
9.108.2 Description
Horizontal slider. See Slider. This one goes from left (min) to right (max).
9.109 HSplitContainer
Inherits: SplitContainer < Container < Control < CanvasItem < Node < Object
Category: Core
9.109.2 Description
Horizontal split container. See SplitContainer. This goes from left to right.
9.110 HTTPClient
Error connect ( String host, int port, bool use_ssl=false, bool verify_host=true )
void set_connection ( StreamPeer connection )
int request ( int method, String url, StringArray headers, String body=”” )
int send_body_text ( String body )
int send_body_data ( RawArray body )
void close ( )
bool has_response ( ) const
bool is_response_chunked ( ) const
int get_response_code ( ) const
StringArray get_response_headers ( )
Dictionary get_response_headers_as_dictionary ( )
int get_response_body_length ( ) const
RawArray read_response_body_chunk ( )
void set_read_chunk_size ( int bytes )
void set_blocking_mode ( bool enabled )
bool is_blocking_mode_enabled ( ) const
int get_status ( ) const
Error poll ( )
String query_string_from_dict ( Dictionary fields )
• METHOD_GET = 0
• METHOD_HEAD = 1
• METHOD_POST = 2
• METHOD_PUT = 3
• METHOD_DELETE = 4
• METHOD_OPTIONS = 5
• METHOD_TRACE = 6
• METHOD_CONNECT = 7
• METHOD_MAX = 8
• STATUS_DISCONNECTED = 0
• STATUS_RESOLVING = 1
• STATUS_CANT_RESOLVE = 2
• STATUS_CONNECTING = 3
• STATUS_CANT_CONNECT = 4
• STATUS_CONNECTED = 5
• STATUS_REQUESTING = 6
• STATUS_BODY = 7
• STATUS_CONNECTION_ERROR = 8
• STATUS_SSL_HANDSHAKE_ERROR = 9
• RESPONSE_CONTINUE = 100
• RESPONSE_SWITCHING_PROTOCOLS = 101
• RESPONSE_PROCESSING = 102
• RESPONSE_OK = 200
• RESPONSE_CREATED = 201
• RESPONSE_ACCEPTED = 202
• RESPONSE_NON_AUTHORITATIVE_INFORMATION = 203
• RESPONSE_NO_CONTENT = 204
• RESPONSE_RESET_CONTENT = 205
• RESPONSE_PARTIAL_CONTENT = 206
• RESPONSE_MULTI_STATUS = 207
• RESPONSE_IM_USED = 226
• RESPONSE_MULTIPLE_CHOICES = 300
• RESPONSE_MOVED_PERMANENTLY = 301
• RESPONSE_FOUND = 302
• RESPONSE_SEE_OTHER = 303
• RESPONSE_NOT_MODIFIED = 304
• RESPONSE_USE_PROXY = 305
• RESPONSE_TEMPORARY_REDIRECT = 307
• RESPONSE_BAD_REQUEST = 400
• RESPONSE_UNAUTHORIZED = 401
• RESPONSE_PAYMENT_REQUIRED = 402
• RESPONSE_FORBIDDEN = 403
• RESPONSE_NOT_FOUND = 404
• RESPONSE_METHOD_NOT_ALLOWED = 405
• RESPONSE_NOT_ACCEPTABLE = 406
• RESPONSE_PROXY_AUTHENTICATION_REQUIRED = 407
• RESPONSE_REQUEST_TIMEOUT = 408
• RESPONSE_CONFLICT = 409
• RESPONSE_GONE = 410
• RESPONSE_LENGTH_REQUIRED = 411
• RESPONSE_PRECONDITION_FAILED = 412
• RESPONSE_REQUEST_ENTITY_TOO_LARGE = 413
• RESPONSE_REQUEST_URI_TOO_LONG = 414
• RESPONSE_UNSUPPORTED_MEDIA_TYPE = 415
• RESPONSE_REQUESTED_RANGE_NOT_SATISFIABLE = 416
• RESPONSE_EXPECTATION_FAILED = 417
• RESPONSE_UNPROCESSABLE_ENTITY = 422
• RESPONSE_LOCKED = 423
• RESPONSE_FAILED_DEPENDENCY = 424
• RESPONSE_UPGRADE_REQUIRED = 426
• RESPONSE_INTERNAL_SERVER_ERROR = 500
• RESPONSE_NOT_IMPLEMENTED = 501
• RESPONSE_BAD_GATEWAY = 502
• RESPONSE_SERVICE_UNAVAILABLE = 503
• RESPONSE_GATEWAY_TIMEOUT = 504
• RESPONSE_HTTP_VERSION_NOT_SUPPORTED = 505
• RESPONSE_INSUFFICIENT_STORAGE = 507
• RESPONSE_NOT_EXTENDED = 510
• Error connect ( String host, int port, bool use_ssl=false, bool verify_host=true )
Connect to a host. This needs to be done before any requests are sent.
The host should not have http:// prepended but will strip the protocol identifier if provided.
verify_host will check the SSL identity of the host if set to true.
• void set_connection ( StreamPeer connection )
• int request ( int method, String url, StringArray headers, String body=”” )
Sends a request to the connected host. The url is what is normally behind the hostname, i.e. in
http://somehost.com/index.php, url would be “index.php”.
Headers are HTTP request headers.
To create a POST request with query strings to push to the server, do:
var fields = {"username" : "user", "password" : "pass"}
This needs to be called in order to have any request processed. Check results with get_status
• String query_string_from_dict ( Dictionary fields )
Generates a GET/POST application/x-www-form-urlencoded style query string from a provided dictionary, e.g.:
var fields = {"username": "user", "password": "pass"}
returns:= "username=user&password=pass"
9.111 Image
Image datatype.
• COMPRESS_BC = 0
• COMPRESS_PVRTC2 = 1
• COMPRESS_PVRTC4 = 2
• COMPRESS_ETC = 3
• FORMAT_GRAYSCALE = 0
• FORMAT_INTENSITY = 1
• FORMAT_GRAYSCALE_ALPHA = 2
• FORMAT_RGB = 3
• FORMAT_RGBA = 4
• FORMAT_INDEXED = 5
• FORMAT_INDEXED_ALPHA = 6
• FORMAT_YUV_422 = 7
• FORMAT_YUV_444 = 8
• FORMAT_BC1 = 9
• FORMAT_BC2 = 10
• FORMAT_BC3 = 11
• FORMAT_BC4 = 12
• FORMAT_BC5 = 13
• FORMAT_PVRTC2 = 14
• FORMAT_PVRTC2_ALPHA = 15
• FORMAT_PVRTC4 = 16
• FORMAT_PVRTC4_ALPHA = 17
• FORMAT_ETC = 18
• FORMAT_ATC = 19
• FORMAT_ATC_ALPHA_EXPLICIT = 20
• FORMAT_ATC_ALPHA_INTERPOLATED = 21
• FORMAT_CUSTOM = 22
9.111.4 Description
Built in native image datatype. Contains image data, which can be converted to a texture, and several functions to
interact with it.
• RawArray get_data ( )
• int get_format ( )
• int get_height ( )
• Color get_pixel ( int x, int y, int mipmap_level=0 )
• Image get_rect ( Rect2 area=0 )
• Rect2 get_used_rect ( )
• int get_width ( )
• int load ( String path=0 )
• void put_pixel ( int x, int y, Color color, int mipmap_level=0 )
• Image resized ( int x, int y, int interpolation=1 )
• int save_png ( String path=0 )
• Image Image ( int width, int height, bool mipmaps, int format )
Create an empty image of a specific size and format.
9.112 ImageTexture
void create ( int width, int height, int format, int flags=7 )
void create_from_image ( Image image, int flags=7 )
int get_format ( ) const
void load ( String path )
void set_data ( Image image )
Image get_data ( ) const
void set_storage ( int mode )
int get_storage ( ) const
void set_lossy_storage_quality ( float quality )
float get_lossy_storage_quality ( ) const
void fix_alpha_edges ( )
void premultiply_alpha ( )
void normal_to_xy ( )
void shrink_x2_and_keep_size ( )
void set_size_override ( Vector2 size )
• STORAGE_RAW = 0
• STORAGE_COMPRESS_LOSSY = 1
• STORAGE_COMPRESS_LOSSLESS = 2
• void create ( int width, int height, int format, int flags=7 )
• void create_from_image ( Image image, int flags=7 )
• int get_format ( ) const
• void load ( String path )
• void set_data ( Image image )
• Image get_data ( ) const
• void set_storage ( int mode )
• int get_storage ( ) const
• void set_lossy_storage_quality ( float quality )
• float get_lossy_storage_quality ( ) const
• void fix_alpha_edges ( )
• void premultiply_alpha ( )
• void normal_to_xy ( )
• void shrink_x2_and_keep_size ( )
• void set_size_override ( Vector2 size )
9.113 ImmediateGeometry
Inherits: GeometryInstance < VisualInstance < Spatial < Node < Object
Category: Core
9.114 Input
Inherits: Object
Inherited By: InputDefault
Category: Core
9.114.3 Signals
• MOUSE_MODE_VISIBLE = 0
• MOUSE_MODE_HIDDEN = 1
• MOUSE_MODE_CAPTURED = 2
9.115 InputDefault
9.116 InputEvent
• int type
• int device
• int ID
• SCREEN_TOUCH = 6
• SCREEN_DRAG = 7
• ACTION = 8
9.116.5 Description
Built-in input event data. InputEvent is a built-in engine datatype, given that it’s passed around and used so much.
Depending on it’s type, the members contained can be different, so read the documentation well!. Input events can
also represent actions (editable from the project settings).
9.117 InputEventAction
• int type
• int device
• int ID
• NONE = 0
• KEY = 1
• MOUSE_MOTION = 2
• MOUSE_BUTTON = 3
• JOYSTICK_MOTION = 4
• JOYSTICK_BUTTON = 5
• SCREEN_TOUCH = 6
• SCREEN_DRAG = 7
• ACTION = 8
9.118 InputEventJoyButton
• int type
• int device
• int ID
• int button_index
• bool pressed
• float pressure
• NONE = 0
• KEY = 1
• MOUSE_MOTION = 2
• MOUSE_BUTTON = 3
• JOYSTICK_MOTION = 4
• JOYSTICK_BUTTON = 5
• SCREEN_TOUCH = 6
• SCREEN_DRAG = 7
• ACTION = 8
9.119 InputEventJoyMotion
• int type
• int device
• int ID
• int axis
• float value
• NONE = 0
• KEY = 1
• MOUSE_MOTION = 2
• MOUSE_BUTTON = 3
• JOYSTICK_MOTION = 4
• JOYSTICK_BUTTON = 5
• SCREEN_TOUCH = 6
• SCREEN_DRAG = 7
• ACTION = 8
9.120 InputEventKey
• int type
• int device
• int ID
• bool shift
• bool alt
• bool control
• bool meta
• bool pressed
• bool echo
• int scancode
• int unicode
• NONE = 0
• KEY = 1
• MOUSE_MOTION = 2
• MOUSE_BUTTON = 3
• JOYSTICK_MOTION = 4
• JOYSTICK_BUTTON = 5
• SCREEN_TOUCH = 6
• SCREEN_DRAG = 7
• ACTION = 8
9.121 InputEventMouseButton
• int type
• int device
• int ID
• bool shift
• bool alt
• bool control
• bool meta
• int button_mask
• int x
• int y
• Vector2 pos
• int global_x
• int global_y
• Vector2 global_pos
• int button_index
• bool pressed
• bool doubleclick
• NONE = 0
• KEY = 1
• MOUSE_MOTION = 2
• MOUSE_BUTTON = 3
• JOYSTICK_MOTION = 4
• JOYSTICK_BUTTON = 5
• SCREEN_TOUCH = 6
• SCREEN_DRAG = 7
• ACTION = 8
9.122 InputEventMouseMotion
• int type
• int device
• int ID
• bool shift
• bool alt
• bool control
• bool meta
• int button_mask
• int x
• int y
• Vector2 pos
• int global_x
• int global_y
• Vector2 global_pos
• int relative_x
• int relative_y
• Vector2 relative_pos
• float speed_x
• float speed_y
• Vector2 speed
• NONE = 0
• KEY = 1
• MOUSE_MOTION = 2
• MOUSE_BUTTON = 3
• JOYSTICK_MOTION = 4
• JOYSTICK_BUTTON = 5
• SCREEN_TOUCH = 6
• SCREEN_DRAG = 7
• ACTION = 8
9.123 InputEventScreenDrag
• int type
• int device
• int ID
• int index
• float x
• float y
• Vector2 pos
• float relative_x
• float relative_y
• Vector2 relative_pos
• float speed_x
• float speed_y
• Vector2 speed
• NONE = 0
• KEY = 1
• MOUSE_MOTION = 2
• MOUSE_BUTTON = 3
• JOYSTICK_MOTION = 4
• JOYSTICK_BUTTON = 5
• SCREEN_TOUCH = 6
• SCREEN_DRAG = 7
• ACTION = 8
9.124 InputEventScreenTouch
• int type
• int device
• int ID
• int index
• float x
• float y
• Vector2 pos
• bool pressed
• NONE = 0
• KEY = 1
• MOUSE_MOTION = 2
• MOUSE_BUTTON = 3
• JOYSTICK_MOTION = 4
• JOYSTICK_BUTTON = 5
• SCREEN_TOUCH = 6
• SCREEN_DRAG = 7
• ACTION = 8
9.125 InputMap
Inherits: Object
Category: Core
9.125.3 Description
Singleton that manages actions. InputMap has a list of the actions used in InputEvent, which can be modified.
9.126 InstancePlaceholder
9.127 int
9.127.3 Description
9.128 IntArray
Integer Array.
9.128.3 Description
Integer Array. Array of integers. Can only contain integers. Optimized for memory usage, can’t fragment the memory.
9.129 InterpolatedCamera
9.130 IP
Inherits: Object
Inherited By: IP_Unix
Category: Core
• RESOLVER_STATUS_NONE = 0
• RESOLVER_STATUS_WAITING = 1
• RESOLVER_STATUS_DONE = 2
• RESOLVER_STATUS_ERROR = 3
• RESOLVER_MAX_QUERIES = 32
• RESOLVER_INVALID_ID = -1
9.130.4 Description
IP contains some support functions for the IPv4 protocol. TCP/IP support is in different classes (see TCP_Client,
TCP_Server). IP provides hostname resolution support, both blocking and threaded.
9.130. IP 487
Godot Engine Documentation, Release latest
9.131 IP_Unix
9.132 ItemList
9.132.3 Signals
• ICON_MODE_TOP = 0
• ICON_MODE_LEFT = 1
• SELECT_SINGLE = 0
• SELECT_MULTI = 1
9.133 Joint
9.134 Joint2D
9.134.3 Description
Base node for all joint constraints in 2D phyisics. Joints take 2 bodies and apply a custom constraint.
9.135 KinematicBody
Inherits: PhysicsBody < CollisionObject < Spatial < Node < Object
Category: Core
9.136 KinematicBody2D
Inherits: PhysicsBody2D < CollisionObject2D < Node2D < CanvasItem < Node < Object
Category: Core
9.136.3 Description
Kinematic bodies are special types of bodies that are meant to be user-controlled. They are not affected by physics at
all (to other types of bodies, such a character or a rigid body, these are the same as a static body). They have however,
two main uses:
Simulated Motion: When these bodies are moved manually, either from code or from an AnimationPlayer (with
process mode set to fixed), the physics will automatically compute an estimate of their linear and angular velocity.
This makes them very useful for moving platforms or other AnimationPlayer-controlled objects (like a door, a bridge
that opens, etc).
Kinematic Characters: KinematicBody2D also has an api for moving objects (the move method) while performing
collision tests. This makes them really useful to implement characters that collide against a world, but that don’t
require advanced physics.
9.137 Label
9.137.4 Description
Label is a control that displays formatted text, optionally autowrapping it to the Control area. It inherits from range to
be able to scroll wrapped text vertically.
9.138 LargeTexture
9.139 Light
• PARAM_RADIUS = 2
• PARAM_ENERGY = 3
• PARAM_ATTENUATION = 4
• PARAM_SPOT_ANGLE = 1
• PARAM_SPOT_ATTENUATION = 0
• PARAM_SHADOW_DARKENING = 5
• PARAM_SHADOW_Z_OFFSET = 6
• COLOR_DIFFUSE = 0
• COLOR_SPECULAR = 1
• BAKE_MODE_DISABLED = 0
• BAKE_MODE_INDIRECT = 1
• BAKE_MODE_INDIRECT_AND_SHADOWS = 2
• BAKE_MODE_FULL = 3
9.139.4 Description
Light is the abstract base class for light nodes, so it shouldn’t be used directly (It can’t be instanced). Other types of
light nodes inherit from it. Light contains the common variables and parameters used for lighting.
9.140 Light2D
• MODE_ADD = 0
• MODE_SUB = 1
• MODE_MIX = 2
• MODE_MASK = 3
9.141 LightOccluder2D
9.142 LineEdit
9.142.3 Signals
• ALIGN_LEFT = 0
• ALIGN_CENTER = 1
• ALIGN_RIGHT = 2
• ALIGN_FILL = 3
9.142.5 Description
LineEdit provides a single line string editor, used for text fields.
Set the text in the LineEdit, clearing the existing one and the selection.
• String get_text ( ) const
Return the text in the LineEdit.
• void set_cursor_pos ( int pos )
Set the cursor position inside the LineEdit, causing it to scroll if needed.
• int get_cursor_pos ( ) const
Return the cursor position inside the LineEdit.
• void set_max_length ( int chars )
Set the maximum amount of characters the LineEdit can edit, and cropping existing text in case it exceeds that limit.
Setting 0 removes the limit.
• int get_max_length ( ) const
Return the maximum amount of characters the LineEdit can edit. If 0 is returned, no limit exists.
• void append_at_cursor ( String text )
Append text at cursor, scrolling the LineEdit when needed.
• void set_editable ( bool enabled )
Set the editable status of the LineEdit. When disabled, existing text can’t be modified and new text can’t be added.
• bool is_editable ( ) const
Return the editable status of the LineEdit (see set_editable).
• void set_secret ( bool enabled )
Set the secret status of the LineEdit. When enabled, every character is displayed as “*”.
• bool is_secret ( ) const
Return the secret status of the LineEdit (see set_secret).
• void select ( int from=0, int to=-1 )
9.143 LineShape2D
9.143.3 Description
Line shape for 2D collision objects. It works like a 2D plane and will not allow any body to go to the negative side.
Not recommended for rigid bodies, and usually not recommended for static bodies either because it forces checks
against it on every frame.
9.144 MainLoop
Inherits: Object
Inherited By: SceneTree
Category: Core
• NOTIFICATION_WM_MOUSE_ENTER = 3
• NOTIFICATION_WM_MOUSE_EXIT = 4
• NOTIFICATION_WM_FOCUS_IN = 5
• NOTIFICATION_WM_FOCUS_OUT = 6
• NOTIFICATION_WM_QUIT_REQUEST = 7
• NOTIFICATION_WM_UNFOCUS_REQUEST = 8
• NOTIFICATION_OS_MEMORY_WARNING = 9
9.144.4 Description
Main loop is the abstract main loop base class. All other main loop classes are derived from it. Upon application start, a
MainLoop has to be provided to OS, else the application will exit. This happens automatically (and a SceneMainLoop
is created), unless a main Script is supplied, which may or not create and return a MainLoop.
9.145 MarginContainer
Inherits: Container < Control < CanvasItem < Node < Object
Category: Core
9.145.2 Description
9.146 Marshalls
9.147 Material
9.147.4 Description
Material is a base Resource used for coloring and shading geometry. All materials inherit from it and almost all
VisualInstance derived nodes carry a Material. A few flags and parameters are shared between all material types and
are configured here.
9.148 MaterialShader
9.149 MaterialShaderGraph
Inherits: ShaderGraph < Shader < Resource < Reference < Object
Category: Core
9.150 Matrix3
float determinant ( )
Vector3 get_euler ( )
int get_orthogonal_index ( )
Vector3 get_scale ( )
Matrix3 inverse ( )
Matrix3 orthonormalized ( )
Matrix3 rotated ( Vector3 axis, float phi )
Matrix3 scaled ( Vector3 scale )
float tdotx ( Vector3 with )
float tdoty ( Vector3 with )
float tdotz ( Vector3 with )
Matrix3 transposed ( )
Vector3 xform ( Vector3 v )
Vector3 xform_inv ( Vector3 v )
Matrix3 Matrix3 ( Vector3 x_axis, Vector3 y_axis, Vector3 z_axis )
Matrix3 Matrix3 ( Vector3 axis, float phi )
Matrix3 Matrix3 ( Quat from )
• Vector3 x
• Vector3 y
• Vector3 z
9.150.4 Description
3x3 matrix used for 3D rotation and scale. Contains 3 vector fields x,y and z. Can also be accessed as array of 3D
vectors. Almost always used as orthogonal basis for a Transform.
• float determinant ( )
Return the determinant of the matrix.
• Vector3 get_euler ( )
Return euler angles from the matrix.
• int get_orthogonal_index ( )
• Vector3 get_scale ( )
• Matrix3 inverse ( )
Return the affine inverse of the matrix.
• Matrix3 orthonormalized ( )
Return the orthonormalized version of the matrix (useful to call from time to time to avoid rounding error).
• Matrix3 rotated ( Vector3 axis, float phi )
Return the rotated version of the matrix, by a given axis and angle.
• Matrix3 scaled ( Vector3 scale )
Return the scaled version of the matrix, by a 3D scale.
• float tdotx ( Vector3 with )
Transposed dot product with the x axis of the matrix.
• float tdoty ( Vector3 with )
Transposed dot product with the y axis of the matrix.
• float tdotz ( Vector3 with )
Transposed dot product with the z axis of the matrix.
• Matrix3 transposed ( )
Return the transposed version of the matrix.
• Vector3 xform ( Vector3 v )
Return a vector transformed by the matrix and return it.
• Vector3 xform_inv ( Vector3 v )
Return a vector transformed by the transposed matrix and return it.
• Matrix3 Matrix3 ( Vector3 x_axis, Vector3 y_axis, Vector3 z_axis )
Create a matrix from 3 axis vectors.
• Matrix3 Matrix3 ( Vector3 axis, float phi )
Create a matrix from an axis vector and an angle.
• Matrix3 Matrix3 ( Quat from )
Create a matrix from a quaternion.
9.151 Matrix32
Matrix32 affine_inverse ( )
Matrix32 basis_xform ( var v )
Matrix32 basis_xform_inv ( var v )
Vector2 get_origin ( )
float get_rotation ( )
Vector2 get_scale ( )
Matrix32 interpolate_with ( Matrix32 m, float c )
Matrix32 inverse ( )
Matrix32 orthonormalized ( )
Matrix32 rotated ( float phi )
Matrix32 scaled ( Vector2 scale )
Matrix32 translated ( Vector2 offset )
Matrix32 xform ( var v )
Matrix32 xform_inv ( var v )
Matrix32 Matrix32 ( float rot, Vector2 pos )
Matrix32 Matrix32 ( Vector2 x_axis, Vector2 y_axis, Vector2 origin )
Matrix32 Matrix32 ( Transform from )
• Vector2 x
• Vector2 y
• Vector2 o
9.151.4 Description
• Matrix32 affine_inverse ( )
• Matrix32 basis_xform ( var v )
• Matrix32 basis_xform_inv ( var v )
• Vector2 get_origin ( )
• float get_rotation ( )
• Vector2 get_scale ( )
• Matrix32 interpolate_with ( Matrix32 m, float c )
• Matrix32 inverse ( )
• Matrix32 orthonormalized ( )
• Matrix32 rotated ( float phi )
• Matrix32 scaled ( Vector2 scale )
• Matrix32 translated ( Vector2 offset )
9.152 MenuButton
Inherits: Button < BaseButton < Control < CanvasItem < Node < Object
Category: Core
PopupMenu get_popup ( )
9.152.3 Signals
• about_to_show ( )
9.152.4 Description
Special button that brings up a PopupMenu when clicked. That’s pretty much all it does, as it’s just a helper class when
building GUIs.
• PopupMenu get_popup ( )
Return the PopupMenu contained in this button.
9.153 Mesh
• NO_INDEX_ARRAY = -1 — Default value used for index_array_len when no indices are present.
• ARRAY_WEIGHTS_SIZE = 4 — Amount of weights/bone indices per vertex (always 4).
• ARRAY_VERTEX = 0 — Vertex array (array of Vector3 vertices).
• ARRAY_NORMAL = 1 — Normal array (array of Vector3 normals).
• ARRAY_TANGENT = 2 — Tangent array, array of groups of 4 floats. first 3 floats determine the tangent, and
the last the binormal direction as -1 or 1.
• ARRAY_COLOR = 3 — Vertex array (array of Color colors).
• ARRAY_TEX_UV = 4 — UV array (array of Vector3 UVs or float array of groups of 2 floats (u,v)).
• ARRAY_TEX_UV2 = 5 — Second UV array (array of Vector3 UVs or float array of groups of 2 floats (u,v)).
• ARRAY_BONES = 6 — Array of bone indices, as a float array. Each element in groups of 4 floats.
• ARRAY_WEIGHTS = 7 — Array of bone weights, as a float array. Each element in groups of 4 floats.
• ARRAY_INDEX = 8 — Array of integers, used as indices referencing vertices. No index can be beyond the
vertex array size.
• ARRAY_FORMAT_VERTEX = 1 — Array format will include vertices (mandatory).
• ARRAY_FORMAT_NORMAL = 2 — Array format will include normals
• ARRAY_FORMAT_TANGENT = 4 — Array format will include tangents
• ARRAY_FORMAT_COLOR = 8 — Array format will include a color array.
• ARRAY_FORMAT_TEX_UV = 16 — Array format will include UVs.
9.153.4 Description
Mesh is a type of Resource that contains vertex-array based geometry, divided in surfaces. Each surface contains a
completely separate array and a material used to draw it. Design wise, a mesh with multiple surfaces is preferred to a
single surface, because objects created in 3D editing software commonly contain multiple materials.
9.154 MeshDataTool
void clear ( )
int create_from_surface ( Object mesh, int surface )
int commit_to_surface ( Object mesh )
int get_format ( ) const
int get_vertex_count ( ) const
int get_edge_count ( ) const
int get_face_count ( ) const
void set_vertex ( int idx, Vector3 vertex )
Vector3 get_vertex ( int idx ) const
void set_vertex_normal ( int idx, Vector3 normal )
Vector3 get_vertex_normal ( int idx ) const
void set_vertex_tangent ( int idx, Plane tangent )
Continued on next page
• void clear ( )
• int create_from_surface ( Object mesh, int surface )
• int commit_to_surface ( Object mesh )
• int get_format ( ) const
• int get_vertex_count ( ) const
• int get_edge_count ( ) const
• int get_face_count ( ) const
• void set_vertex ( int idx, Vector3 vertex )
• Vector3 get_vertex ( int idx ) const
• void set_vertex_normal ( int idx, Vector3 normal )
• Vector3 get_vertex_normal ( int idx ) const
• void set_vertex_tangent ( int idx, Plane tangent )
• Plane get_vertex_tangent ( int idx ) const
• void set_vertex_uv ( int idx, Vector2 uv )
• Vector2 get_vertex_uv ( int idx ) const
9.155 MeshInstance
Inherits: GeometryInstance < VisualInstance < Spatial < Node < Object
Category: Core
9.155.3 Description
MeshInstance is a Node that takes a Mesh resource and adds it to the current Scenario by creating an instance of it.
This is the class most often used to get 3D geometry rendered and can be used to instance a single Mesh in many
places. This allows to reuse geometry and save on resources. When a Mesh has to be instanced more than thousands
of times at close proximity, consider using a MultiMesh in a MultiMeshInstance instead.
9.156 MeshLibrary
Library of meshes.
9.156.3 Description
Library of meshes. Contains a list of Mesh resources, each with name and ID. Useful for GridMap or painting Terrain.
9.157 MultiMesh
9.157.3 Description
MultiMesh provides low level mesh instancing. If the amount of Mesh instances needed goes from hundreds to
thousands (and most need to be visible at close proximity) creating such a large amount of MeshInstance nodes may
affect performance by using too much CPU or video memory.
For this case a MultiMesh becomes very useful, as it can draw thousands of instances with little API overhead.
As a drawback, if the instances are too far away of each other, performance may be reduced as every
single instance will always rendered (they are spatially indexed as one, for the whole object).
Since instances may have any behavior, the AABB used for visibility must be provided by the user, or generated with
generate_aabb.
9.158 MultiMeshInstance
Inherits: GeometryInstance < VisualInstance < Spatial < Node < Object
Category: Core
9.158.3 Description
MultiMeshInstance is a Node that takes a MultiMesh resource and adds it to the current Scenario by creating an
instance of it (yes, this is an instance of instances).
9.159 Mutex
void lock ( )
Error try_lock ( )
void unlock ( )
• void lock ( )
• Error try_lock ( )
• void unlock ( )
9.160 Navigation
9.161 Navigation2D
9.162 NavigationMesh
9.163 NavigationMeshInstance
9.164 NavigationPolygon
9.165 NavigationPolygonInstance
9.166 Nil
9.167 Node
Inherits: Object
Inherited By: Viewport, Timer, CanvasLayer, EventPlayer, SoundRoomParams, Spatial, AnimationPlayer, Editor-
Plugin, ResourcePreloader, AnimationTreePlayer, SamplePlayer, InstancePlaceholder, StreamPlayer, CanvasItem,
Tween
Category: Core
9.167.3 Signals
• renamed ( )
• enter_tree ( )
• exit_tree ( )
• NOTIFICATION_ENTER_TREE = 10
• NOTIFICATION_EXIT_TREE = 11
• NOTIFICATION_MOVED_IN_PARENT = 12
• NOTIFICATION_READY = 13
• NOTIFICATION_FIXED_PROCESS = 16
• NOTIFICATION_PROCESS = 17 — Notification received every frame when the process flag is set (see
set_process).
• NOTIFICATION_PARENTED = 18 — Notification received when a node is set as a child of another node.
Note that this doesn’t mean that a node entered the Scene Tree.
• NOTIFICATION_UNPARENTED = 19 — Notification received when a node is unparented (parent removed
it from the list of children).
• NOTIFICATION_PAUSED = 14
• NOTIFICATION_UNPAUSED = 15
• NOTIFICATION_INSTANCED = 20
• PAUSE_MODE_INHERIT = 0
• PAUSE_MODE_STOP = 1
• PAUSE_MODE_PROCESS = 2
9.167.5 Description
Nodes can be set as children of other nodes, resulting in a tree arrangement. Any tree of nodes is called a “Scene”.
Scenes can be saved to disk, and then instanced into other scenes. This allows for very high flexibility in the architec-
ture and data model of the projects.
SceneMainLoop contains the “active” tree of nodes, and a node becomes active (receiving NOTIFICA-
TION_ENTER_SCENE) when added to that tree.
A node can contain any number of nodes as a children (but there is only one tree root) with the requirement that no
two children with the same name can exist.
Nodes can, optionally, be added to groups. This makes it easy to reach a number of nodes from the code (for example
an “enemies” group).
Nodes can be set to “process” state, so they constantly receive a callback requesting them to process (do anything).
Normal processing (_process) happens as fast as possible and is dependent on the frame rate, so the processing time
delta is variable. Fixed processing (_fixed_process) happens a fixed amount of times per second (by default 60) and is
useful to link itself to the physics.
Nodes can also process input events. When set, the _input function will be called with every input that the program
receives. Since this is usually too overkill (unless used for simple projects), an _unhandled_input function is called
when the input was not handled by anyone else (usually, GUI Control nodes).
To keep track of the scene hierarchy (specially when instancing scenes into scenes) an “owner” can be set to a node.
This keeps track of who instanced what. This is mostly useful when writing editors and tools, though.
Finally, when a node is freed, it will free all its children nodes too.
The optional boolean argument enforces creating child node with human-readable names, based on the name of node
being instanced instead of its type only.
• void remove_child ( Node node )
Remove a child Node. Node is NOT deleted and will have to be deleted manually.
• int get_child_count ( ) const
Return the amount of children nodes.
• Array get_children ( ) const
• Node get_child ( int idx ) const
Return a children node by it’s index (see get_child_count). This method is often used for iterating all children of a
node.
• bool has_node ( NodePath path ) const
• Node get_node ( NodePath path ) const
Fetch a node. NodePath must be valid (or else error will occur) and can be either the path to child node, a relative path
(from the current node to another node), or an absolute path to a node.
Note: fetching absolute paths only works when the node is inside the scene tree (see is_inside_scene). Examples.
Assume your current node is Character and following tree:
root/
root/Character
root/Character/Sword
root/Character/Backpack/Dagger
root/MyGame
root/Swamp/Alligator
root/Swamp/Mosquito
root/Swamp/Goblin
Possible paths are:
• get_node(“Sword”)
• get_node(“Backpack/Dagger”)
• get_node(”../Swamp/Alligator”)
• get_node(“/root/MyGame”)
• Parent get_parent ( ) const
Return the parent Node of the current Node, or an empty Object if the node lacks a parent.
• Node find_node ( String mask, bool recursive=true, bool owned=true ) const
• bool has_node_and_resource ( NodePath path ) const
• Array get_node_and_resource ( NodePath path )
• bool is_inside_tree ( ) const
• bool is_a_parent_of ( Node node ) const
Return true if the “node” argument is a direct or indirect child of the current node, otherwise return false.
Return a filename that may be containedA node can contained by the node. When a scene is instanced from a file, it
topmost node contains the filename from where it was loaded (see set_filename).
• void propagate_notification ( int what )
Notify the current node and all its children recursively by calling notification() in all of them.
• void set_fixed_process ( bool enable )
Enables or disables node fixed framerate processing. When a node is being processed, it will receive a NOTIFI-
CATION_PROCESS at a fixed (usually 60 fps, check OS to change that) interval (and the _fixed_process callback
will be called if exists). It is common to check how much time was elapsed since the previous frame by calling
get_fixed_process_time.
• float get_fixed_process_delta_time ( ) const
Return the time elapsed since the last fixed frame. This is always the same in fixed processing unless the frames per
second is changed in OS.
• bool is_fixed_processing ( ) const
Return true if fixed processing is enabled (see set_fixed_process).
• void set_process ( bool enable )
Enables or disables node processing. When a node is being processed, it will receive a NOTIFICATION_PROCESS
on every drawn frame (and the _process callback will be called if exists). It is common to check how much time was
elapsed since the previous frame by calling get_process_time.
• float get_process_delta_time ( ) const
Return the time elapsed (in seconds) since the last process callback. This is almost always different each time.
• bool is_processing ( ) const
Return whether processing is enabled in the current node (see set_process).
• void set_process_input ( bool enable )
Enable input processing for node. This is not required for GUI controls! It hooks up the node to receive all input (see
_input).
• bool is_processing_input ( ) const
Return true if the node is processing input (see set_process_input).
• void set_process_unhandled_input ( bool enable )
Enable unhandled input processing for node. This is not required for GUI controls! It hooks up the node to receive all
input that was not previously handled before (usually by a Control). (see _unhandled_input).
• bool is_processing_unhandled_input ( ) const
Return true if the node is processing unhandled input (see set_process_unhandled_input).
• void set_process_unhandled_key_input ( bool enable )
• bool is_processing_unhandled_key_input ( ) const
• void set_pause_mode ( int mode )
• int get_pause_mode ( ) const
• bool can_process ( ) const
Return true if the node can process.
• void print_stray_nodes ( )
9.168 Node2D
9.168.3 Description
Base node for 2D system. Node2D contains a position, rotation and scale, which is used to position and animate. It
can alternatively be used with a custom 2D transform (Matrix32). A tree of Node2Ds allows complex hierarchies for
animation and positioning.
9.169 NodePath
9.169.3 Description
Built-in type optimized for path traversing. A Node path is an optimized compiled path used for traversing the scene
tree. It references nodes and can reference properties in that node, or even reference properties inside the resources of
the node.
9.170 Object
Inherited By: Reference, Physics2DServer, Input, SpatialSound2DServer, Node, Geometry, TreeItem, Physics-
DirectSpaceState, Physics2DDirectSpaceState, MainLoop, InputMap, UndoRedo, PhysicsServer, Resource-
Saver, Performance, PathRemap, ResourceLoader, AudioServer, SpatialSoundServer, IP, VisualServer, OS,
Physics2DDirectBodyState, Globals, PhysicsDirectBodyState, TranslationServer
Category: Core
9.170.3 Signals
• script_changed ( )
• CONNECT_DEFERRED = 1 — Connect a signal in deferred mode. This way, signal emissions are stored in
a queue, then set on idle time.
• CONNECT_PERSIST = 2 — Persisting connections are saved when the object is serialized to file.
• CONNECT_ONESHOT = 4 — One short connections disconnect themselves after emission.
9.170.5 Description
Base class for all non built-in types. Everything not a built-in type starts the inheritance chain from this class.
Objects do not manage memory, if inheriting from one the object will most likely have to be deleted manually (call
the free function from the script or delete from C++).
Some derivates add memory management, such as Reference (which keeps a reference count and deletes itself auto-
matically when no longer referenced) and Node, which deletes the children tree when deleted.
Objects export properties, which are mainly useful for storage and editing, but not really so much in programming.
Properties are exported in _get_property_list and handled in _get and _set. However, scripting languages and C++
have simper means to export them.
Objects also receive notifications (_notification). Notifications are a simple way to notify the object about simple
events, so they can all be handled together.
9.171 OccluderPolygon2D
• CULL_DISABLED = 0
• CULL_CLOCKWISE = 1
• CULL_COUNTER_CLOCKWISE = 2
9.172 OmniLight
Inherits: Light < VisualInstance < Spatial < Node < Object
Category: Core
9.172.2 Description
An OmniDirectional light is a type of Light node that emits lights in all directions. The light is attenuated through the
distance and this attenuation can be configured by changing the energy, radius and attenuation parameters of Light.
TODO: Image of an omnilight.
9.173 OptionButton
Inherits: Button < BaseButton < Control < CanvasItem < Node < Object
Category: Core
9.173.3 Signals
• item_selected ( int ID )
9.173.4 Description
OptionButton is a type button that provides a selectable list of items when pressed. The item selected becomes the
“current” item and is displayed as the button text.
Add an item, with a “texture” icon, text “label” and (optionally) id. If no “id” is passed, “id” becomes the item index.
New items are appended at the end.
• void set_item_text ( int idx, String text )
Set the text of an item at index “idx”.
• void set_item_icon ( int idx, Texture texture )
Set the icon of an item at index “idx”.
• void set_item_disabled ( int idx, bool disabled )
• void set_item_ID ( int idx, int id )
Set the ID of an item at index “idx”.
• void set_item_metadata ( int idx, var metadata )
• String get_item_text ( int idx ) const
Return the text of the item at index “idx”.
• Texture get_item_icon ( int idx ) const
Return the icon of the item at index “idx”.
• int get_item_ID ( int idx ) const
Return the ID of the item at index “idx”.
• void get_item_metadata ( int idx ) const
• bool is_item_disabled ( int idx ) const
• int get_item_count ( ) const
Return the amount of items in the OptionButton.
• void add_separator ( )
Add a separator to the list of items. Separators help to group items. Separator also takes up an index and is appended
at the end.
• void clear ( )
Clear all the items in the OptionButton.
• void select ( int idx )
Select an item by index and make it the current item.
• int get_selected ( ) const
Return the current item index
• int get_selected_ID ( ) const
• void get_selected_metadata ( ) const
• void remove_item ( int idx )
9.174 OS
Inherits: Object
Category: Core
9.174. OS 547
Godot Engine Documentation, Release latest
• DAY_SUNDAY = 0
9.174. OS 549
Godot Engine Documentation, Release latest
• DAY_MONDAY = 1
• DAY_TUESDAY = 2
• DAY_WEDNESDAY = 3
• DAY_THURSDAY = 4
• DAY_FRIDAY = 5
• DAY_SATURDAY = 6
• MONTH_JANUARY = 0
• MONTH_FEBRUARY = 1
• MONTH_MARCH = 2
• MONTH_APRIL = 3
• MONTH_MAY = 4
• MONTH_JUNE = 5
• MONTH_JULY = 6
• MONTH_AUGUST = 7
• MONTH_SEPTEMBER = 8
• MONTH_OCTOBER = 9
• MONTH_NOVEMBER = 10
• MONTH_DECEMBER = 11
• SCREEN_ORIENTATION_LANDSCAPE = 0
• SCREEN_ORIENTATION_PORTRAIT = 1
• SCREEN_ORIENTATION_REVERSE_LANDSCAPE = 2
• SCREEN_ORIENTATION_REVERSE_PORTRAIT = 3
• SCREEN_ORIENTATION_SENSOR_LANDSCAPE = 4
• SCREEN_ORIENTATION_SENSOR_PORTRAIT = 5
• SCREEN_ORIENTATION_SENSOR = 6
• SYSTEM_DIR_DESKTOP = 0
• SYSTEM_DIR_DCIM = 1
• SYSTEM_DIR_DOCUMENTS = 2
• SYSTEM_DIR_DOWNLOADS = 3
• SYSTEM_DIR_MOVIES = 4
• SYSTEM_DIR_MUSIC = 5
• SYSTEM_DIR_PICTURES = 6
• SYSTEM_DIR_RINGTONES = 7
9.174.4 Description
Operating System functions. OS Wraps the most common functionality to communicate with the host Operating
System, such as:
-Mouse Grabbing
-Mouse Cursors
-Clipboard
-Video Mode
-Date ” Time
-Timers
-Environment Variables
-Execution of Binaries
-Command Line
9.174. OS 551
Godot Engine Documentation, Release latest
9.174. OS 553
Godot Engine Documentation, Release latest
9.175 PackedDataContainer
9.176 PackedDataContainerRef
9.177 PackedScene
9.177.3 Description
TODO: explain ownership, and that node does not need to own itself
9.178 PacketPeer
9.178.3 Description
PacketPeer is an abstraction and base class for packet-based protocols (such as UDP). It provides an API for sending
and receiving packets both as raw data or variables. This makes it easy to transfer data over a protocol, without having
to encode data as low level bytes or having to worry about network ordering.
9.179 PacketPeerStream
9.179.3 Description
PacketStreamPeer provides a wrapper for working using packets over a stream. This allows for using packet based
code with StreamPeers. PacketPeerStream implements a custom protocol over the StreamPeer, so the user should not
read or write to the wrapped StreamPeer directly.
9.180 PacketPeerUDP
9.181 Panel
9.181.2 Description
Panel is a Control that displays an opaque background. It’s commonly used as a parent and container for other types
of Control nodes. image<class_image>‘images/panel_example.png:ref:/image<class_/image>‘
9.182 PanelContainer
Inherits: Container < Control < CanvasItem < Node < Object
Category: Core
9.182.2 Description
Panel container type. This container fits controls inside of the delimited area of a stylebox. It’s useful for giving
controls an outline.
9.183 ParallaxBackground
9.184 ParallaxLayer
9.185 ParticleAttractor2D
9.186 Particles
Inherits: GeometryInstance < VisualInstance < Spatial < Node < Object
Category: Core
• VAR_LIFETIME = 0
• VAR_SPREAD = 1
• VAR_GRAVITY = 2
• VAR_LINEAR_VELOCITY = 3
• VAR_ANGULAR_VELOCITY = 4
• VAR_LINEAR_ACCELERATION = 5
• VAR_DRAG = 6
• VAR_TANGENTIAL_ACCELERATION = 7
• VAR_INITIAL_SIZE = 9
• VAR_FINAL_SIZE = 10
• VAR_INITIAL_ANGLE = 11
• VAR_HEIGHT = 12
• VAR_HEIGHT_SPEED_SCALE = 13
• VAR_MAX = 14
9.186.4 Description
Particles is a particle system 3D Node that is used to simulate several types of particle effects, such as explosions, rain,
snow, fireflies, or other magical-like shinny sparkles. Particles are drawn using impostors, and given their dynamic
behavior, the user must provide a visibility AABB (although helpers to create one automatically exist).
9.187 Particles2D
• PARAM_DIRECTION = 0
• PARAM_SPREAD = 1
• PARAM_LINEAR_VELOCITY = 2
• PARAM_SPIN_VELOCITY = 3
• PARAM_ORBIT_VELOCITY = 4
• PARAM_GRAVITY_DIRECTION = 5
• PARAM_GRAVITY_STRENGTH = 6
• PARAM_RADIAL_ACCEL = 7
• PARAM_TANGENTIAL_ACCEL = 8
• PARAM_DAMPING = 9
• PARAM_INITIAL_ANGLE = 10
• PARAM_INITIAL_SIZE = 11
• PARAM_FINAL_SIZE = 12
• PARAM_HUE_VARIATION = 13
• PARAM_ANIM_SPEED_SCALE = 14
• PARAM_ANIM_INITIAL_POS = 15
• PARAM_MAX = 16
• MAX_COLOR_PHASES = 4
9.188 Patch9Frame
9.189 Path
9.189.3 Description
This class is a container/Node-ification of a Curve3D, so it can have Spatial properties and Node info.
9.190 Path2D
9.190.3 Description
This class is a container/Node-ification of a Curve2D, so it can have Node2D properties and Node info.
9.191 PathFollow
9.191.4 Description
This node takes its parent Path, and returns the coordinates of a point within it, given a distance from the first vertex.
It is useful for making other nodes follow a path, without coding the movement pattern. For that, the nodes must be
descendants of this node. Then, when setting an offset in this node, the descendant nodes will move accordingly.
9.192 PathFollow2D
9.192.3 Description
This node takes its parent Path2D, and returns the coordinates of a point within it, given a distance from the first vertex.
It is useful for making other nodes follow a path, without coding the movement pattern. For that, the nodes must be
descendants of this node. Then, when setting an offset in this node, the descendant nodes will move accordingly.
9.193 PathRemap
Inherits: Object
Category: Core
9.193.3 Description
When exporting, the types of some resources may change internally so they are converted to more optimized versions.
While it’s not usually necessary to access to this directly (path remapping happens automatically when opening a file),
it’s exported just for information.
• void clear_remaps ( )
Clear all remaps.
9.194 PCKPacker
9.195 Performance
Inherits: Object
Category: Core
• TIME_FPS = 0
• TIME_PROCESS = 1
• TIME_FIXED_PROCESS = 2
• MEMORY_STATIC = 3
• MEMORY_DYNAMIC = 4
• MEMORY_STATIC_MAX = 5
• MEMORY_DYNAMIC_MAX = 6
• MEMORY_MESSAGE_BUFFER_MAX = 7
• OBJECT_COUNT = 8
• OBJECT_RESOURCE_COUNT = 9
• OBJECT_NODE_COUNT = 10
• RENDER_OBJECTS_IN_FRAME = 11
• RENDER_VERTICES_IN_FRAME = 12
• RENDER_MATERIAL_CHANGES_IN_FRAME = 13
• RENDER_SHADER_CHANGES_IN_FRAME = 14
• RENDER_SURFACE_CHANGES_IN_FRAME = 15
• RENDER_DRAW_CALLS_IN_FRAME = 16
• RENDER_USAGE_VIDEO_MEM_TOTAL = 20
• RENDER_VIDEO_MEM_USED = 17
• RENDER_TEXTURE_MEM_USED = 18
• RENDER_VERTEX_MEM_USED = 19
• PHYSICS_2D_ACTIVE_OBJECTS = 21
• PHYSICS_2D_COLLISION_PAIRS = 22
• PHYSICS_2D_ISLAND_COUNT = 23
• PHYSICS_3D_ACTIVE_OBJECTS = 24
• PHYSICS_3D_COLLISION_PAIRS = 25
• PHYSICS_3D_ISLAND_COUNT = 26
• MONITOR_MAX = 27
9.196 PHashTranslation
Optimized translation.
9.196.3 Description
Optimized translation. Uses real-time compressed translations, which results in very small dictionaries.
9.197 Physics2DDirectBodyState
Inherits: Object
Inherited By: Physics2DDirectBodyStateSW
Category: Core
9.197.3 Description
Direct access object to a physics body in the Physics2DServer. This object is passed via the direct state callback of
rigid/character bodies, and is intended for changing the direct state of that body.
9.198 Physics2DDirectBodyStateSW
9.199 Physics2DDirectSpaceState
Inherits: Object
Category: Core
• TYPE_MASK_STATIC_BODY = 1
• TYPE_MASK_KINEMATIC_BODY = 2
• TYPE_MASK_RIGID_BODY = 4
• TYPE_MASK_CHARACTER_BODY = 8
• TYPE_MASK_AREA = 16
• TYPE_MASK_COLLISION = 15
9.199.4 Description
Direct access object to a space in the Physics2DServer. It’s used mainly to do queries against objects and areas residing
in a given space.
Intersect a given shape (RID or Shape2D) against the space, the intersected shapes are returned in a special result
object.
• Array cast_motion ( Physics2DShapeQueryParameters shape )
• Array collide_shape ( Physics2DShapeQueryParameters shape, int max_results=32 )
• Dictionary get_rest_info ( Physics2DShapeQueryParameters shape )
9.200 Physics2DServer
Inherits: Object
Inherited By: Physics2DServerSW
Category: Core
Physics 2D Server.
• SHAPE_LINE = 0
• SHAPE_SEGMENT = 2
• SHAPE_CIRCLE = 3
• SHAPE_RECTANGLE = 4
• SHAPE_CAPSULE = 5
• SHAPE_CONVEX_POLYGON = 6
• SHAPE_CONCAVE_POLYGON = 7
• SHAPE_CUSTOM = 8
• AREA_PARAM_GRAVITY = 0
• AREA_PARAM_GRAVITY_VECTOR = 1
• AREA_PARAM_GRAVITY_IS_POINT = 2
• AREA_PARAM_GRAVITY_DISTANCE_SCALE = 3
• AREA_PARAM_GRAVITY_POINT_ATTENUATION = 4
• AREA_PARAM_LINEAR_DAMP = 5
• AREA_PARAM_ANGULAR_DAMP = 6
• AREA_PARAM_PRIORITY = 7
• AREA_SPACE_OVERRIDE_DISABLED = 0 — This area does not affect gravity/damp. These are generally
areas that exist only to detect collisions, and objects entering or exiting them.
• AREA_SPACE_OVERRIDE_COMBINE = 1 — This area adds its gravity/damp values to whatever has been
calculated so far. This way, many overlapping areas can combine their physics to make interesting effects.
• AREA_SPACE_OVERRIDE_COMBINE_REPLACE = 2 — This area adds its gravity/damp values to what-
ever has been calculated so far. Then stops taking into account the rest of the areas, even the default one.
• AREA_SPACE_OVERRIDE_REPLACE = 3 — This area replaces any gravity/damp, even the default one,
and stops taking into account the rest of the areas.
• AREA_SPACE_OVERRIDE_REPLACE_COMBINE = 4 — This area replaces any gravity/damp calculated
so far, but keeps calculating the rest of the areas, down to the default one.
• BODY_MODE_STATIC = 0
• BODY_MODE_KINEMATIC = 1
• BODY_MODE_RIGID = 2
• BODY_MODE_CHARACTER = 3
• BODY_PARAM_BOUNCE = 0
• BODY_PARAM_FRICTION = 1
• BODY_PARAM_MASS = 2
• BODY_PARAM_GRAVITY_SCALE = 3
• BODY_PARAM_LINEAR_DAMP = 4
• BODY_PARAM_ANGULAR_DAMP = 5
• BODY_PARAM_MAX = 6
• BODY_STATE_TRANSFORM = 0
• BODY_STATE_LINEAR_VELOCITY = 1
• BODY_STATE_ANGULAR_VELOCITY = 2
• BODY_STATE_SLEEPING = 3
• BODY_STATE_CAN_SLEEP = 4
• JOINT_PIN = 0
• JOINT_GROOVE = 1
• JOINT_DAMPED_SPRING = 2
• DAMPED_STRING_REST_LENGTH = 0
• DAMPED_STRING_STIFFNESS = 1
• DAMPED_STRING_DAMPING = 2
• CCD_MODE_DISABLED = 0
• CCD_MODE_CAST_RAY = 1
• CCD_MODE_CAST_SHAPE = 2
• AREA_BODY_ADDED = 0
• AREA_BODY_REMOVED = 1
• INFO_ACTIVE_OBJECTS = 0
• INFO_COLLISION_PAIRS = 1
• INFO_ISLAND_COUNT = 2
9.200.4 Description
• bool body_test_motion ( RID body, Vector2 motion, float margin=0.08, Physics2DTestMotionResult re-
sult=NULL )
• void joint_set_param ( RID joint, int param, float value )
• float joint_get_param ( RID joint, int param ) const
• RID pin_joint_create ( Vector2 anchor, RID body_a, RID body_b=RID() )
• RID groove_joint_create ( Vector2 groove1_a, Vector2 groove2_a, Vector2 anchor_b, RID body_a=RID(), RID
body_b=RID() )
• RID damped_spring_joint_create ( Vector2 anchor_a, Vector2 anchor_b, RID body_a, RID body_b=RID() )
• void damped_string_joint_set_param ( RID joint, int param, float value=RID() )
• float damped_string_joint_get_param ( RID joint, int param ) const
• int joint_get_type ( RID joint ) const
• void free_rid ( RID rid )
• void set_active ( bool active )
• int get_process_info ( int process_info )
9.201 Physics2DServerSW
9.202 Physics2DShapeQueryParameters
9.203 Physics2DShapeQueryResult
9.204 Physics2DTestMotionResult
9.205 PhysicsBody
9.205.3 Description
PhysicsBody is an abstract base class for implementing a physics body. All PhysicsBody types inherit from it.
9.206 PhysicsBody2D
Inherits: CollisionObject2D < Node2D < CanvasItem < Node < Object
Inherited By: RigidBody2D, StaticBody2D, KinematicBody2D
Category: Core
9.206.3 Description
PhysicsBody2D is an abstract base class for implementing a physics body. All *Body2D types inherit from it.
Set/clear individual bits on the layer mask. This makes getting a body in/out of only one layer easier.
• bool get_layer_mask_bit ( int bit ) const
Return an individual bit on the collision mask.
• void set_one_way_collision_direction ( Vector2 dir )
Set a direction in which bodies can go through this one. If this value is different from (0,0), any movement within 90
degrees of this vector is considered a valid movement. Set this direction to (0,0) to disable one-way collisions.
• Vector2 get_one_way_collision_direction ( ) const
Return the direction used for one-way collision detection.
• void set_one_way_collision_max_depth ( float depth )
Set how far a body can go through this one, when it allows one-way collisions (see set_one_way_collision_detection).
9.207 PhysicsDirectBodyState
Inherits: Object
Inherited By: PhysicsDirectBodyStateSW
Category: Core
9.208 PhysicsDirectBodyStateSW
9.209 PhysicsDirectSpaceState
Inherits: Object
Category: Core
Dictio- intersect_ray ( Vector3 from, Vector3 to, Array exclude=Array(), int layer_mask=2147483647, int
nary type_mask=15 )
Array intersect_shape ( PhysicsShapeQueryParameters shape, int max_results=32 )
Array cast_motion ( PhysicsShapeQueryParameters shape, Vector3 motion )
Array collide_shape ( PhysicsShapeQueryParameters shape, int max_results=32 )
Dictio- get_rest_info ( PhysicsShapeQueryParameters shape )
nary
• TYPE_MASK_STATIC_BODY = 1
• TYPE_MASK_KINEMATIC_BODY = 2
• TYPE_MASK_RIGID_BODY = 4
• TYPE_MASK_CHARACTER_BODY = 8
• TYPE_MASK_AREA = 16
• TYPE_MASK_COLLISION = 15
• Dictionary intersect_ray ( Vector3 from, Vector3 to, Array exclude=Array(), int layer_mask=2147483647, int
type_mask=15 )
• Array intersect_shape ( PhysicsShapeQueryParameters shape, int max_results=32 )
• Array cast_motion ( PhysicsShapeQueryParameters shape, Vector3 motion )
• Array collide_shape ( PhysicsShapeQueryParameters shape, int max_results=32 )
• Dictionary get_rest_info ( PhysicsShapeQueryParameters shape )
9.210 PhysicsServer
Inherits: Object
Inherited By: PhysicsServerSW
Category: Core
• JOINT_PIN = 0
• JOINT_HINGE = 1
• JOINT_SLIDER = 2
• JOINT_CONE_TWIST = 3
• JOINT_6DOF = 4
• PIN_JOINT_BIAS = 0
• PIN_JOINT_DAMPING = 1
• PIN_JOINT_IMPULSE_CLAMP = 2
• HINGE_JOINT_BIAS = 0
• HINGE_JOINT_LIMIT_UPPER = 1
• HINGE_JOINT_LIMIT_LOWER = 2
• HINGE_JOINT_LIMIT_BIAS = 3
• HINGE_JOINT_LIMIT_SOFTNESS = 4
• HINGE_JOINT_LIMIT_RELAXATION = 5
• HINGE_JOINT_MOTOR_TARGET_VELOCITY = 6
• HINGE_JOINT_MOTOR_MAX_IMPULSE = 7
• HINGE_JOINT_FLAG_USE_LIMIT = 0
• HINGE_JOINT_FLAG_ENABLE_MOTOR = 1
• SLIDER_JOINT_LINEAR_LIMIT_UPPER = 0
• SLIDER_JOINT_LINEAR_LIMIT_LOWER = 1
• SLIDER_JOINT_LINEAR_LIMIT_SOFTNESS = 2
• SLIDER_JOINT_LINEAR_LIMIT_RESTITUTION = 3
• SLIDER_JOINT_LINEAR_LIMIT_DAMPING = 4
• SLIDER_JOINT_LINEAR_MOTION_SOFTNESS = 5
• SLIDER_JOINT_LINEAR_MOTION_RESTITUTION = 6
• SLIDER_JOINT_LINEAR_MOTION_DAMPING = 7
• SLIDER_JOINT_LINEAR_ORTHOGONAL_SOFTNESS = 8
• SLIDER_JOINT_LINEAR_ORTHOGONAL_RESTITUTION = 9
• SLIDER_JOINT_LINEAR_ORTHOGONAL_DAMPING = 10
• SLIDER_JOINT_ANGULAR_LIMIT_UPPER = 11
• SLIDER_JOINT_ANGULAR_LIMIT_LOWER = 12
• SLIDER_JOINT_ANGULAR_LIMIT_SOFTNESS = 13
• SLIDER_JOINT_ANGULAR_LIMIT_RESTITUTION = 14
• SLIDER_JOINT_ANGULAR_LIMIT_DAMPING = 15
• SLIDER_JOINT_ANGULAR_MOTION_SOFTNESS = 16
• SLIDER_JOINT_ANGULAR_MOTION_RESTITUTION = 17
• SLIDER_JOINT_ANGULAR_MOTION_DAMPING = 18
• SLIDER_JOINT_ANGULAR_ORTHOGONAL_SOFTNESS = 19
• SLIDER_JOINT_ANGULAR_ORTHOGONAL_RESTITUTION = 20
• SLIDER_JOINT_ANGULAR_ORTHOGONAL_DAMPING = 21
• SLIDER_JOINT_MAX = 22
• CONE_TWIST_JOINT_SWING_SPAN = 0
• CONE_TWIST_JOINT_TWIST_SPAN = 1
• CONE_TWIST_JOINT_BIAS = 2
• CONE_TWIST_JOINT_SOFTNESS = 3
• CONE_TWIST_JOINT_RELAXATION = 4
• G6DOF_JOINT_LINEAR_LOWER_LIMIT = 0
• G6DOF_JOINT_LINEAR_UPPER_LIMIT = 1
• G6DOF_JOINT_LINEAR_LIMIT_SOFTNESS = 2
• G6DOF_JOINT_LINEAR_RESTITUTION = 3
• G6DOF_JOINT_LINEAR_DAMPING = 4
• G6DOF_JOINT_ANGULAR_LOWER_LIMIT = 5
• G6DOF_JOINT_ANGULAR_UPPER_LIMIT = 6
• G6DOF_JOINT_ANGULAR_LIMIT_SOFTNESS = 7
• G6DOF_JOINT_ANGULAR_DAMPING = 8
• G6DOF_JOINT_ANGULAR_RESTITUTION = 9
• G6DOF_JOINT_ANGULAR_FORCE_LIMIT = 10
• G6DOF_JOINT_ANGULAR_ERP = 11
• G6DOF_JOINT_ANGULAR_MOTOR_TARGET_VELOCITY = 12
• G6DOF_JOINT_ANGULAR_MOTOR_FORCE_LIMIT = 13
• G6DOF_JOINT_FLAG_ENABLE_LINEAR_LIMIT = 0
• G6DOF_JOINT_FLAG_ENABLE_ANGULAR_LIMIT = 1
• G6DOF_JOINT_FLAG_ENABLE_MOTOR = 2
• SHAPE_PLANE = 0
• SHAPE_RAY = 1
• SHAPE_SPHERE = 2
• SHAPE_BOX = 3
• SHAPE_CAPSULE = 4
• SHAPE_CONVEX_POLYGON = 5
• SHAPE_CONCAVE_POLYGON = 6
• SHAPE_HEIGHTMAP = 7
• SHAPE_CUSTOM = 8
• AREA_PARAM_GRAVITY = 0
• AREA_PARAM_GRAVITY_VECTOR = 1
• AREA_PARAM_GRAVITY_IS_POINT = 2
• AREA_PARAM_GRAVITY_DISTANCE_SCALE = 3
• AREA_PARAM_GRAVITY_POINT_ATTENUATION = 4
• AREA_PARAM_LINEAR_DAMP = 5
• AREA_PARAM_ANGULAR_DAMP = 6
• AREA_PARAM_PRIORITY = 7
• AREA_SPACE_OVERRIDE_DISABLED = 0 — This area does not affect gravity/damp. These are generally
areas that exist only to detect collisions, and objects entering or exiting them.
• AREA_SPACE_OVERRIDE_COMBINE = 1 — This area adds its gravity/damp values to whatever has been
calculated so far. This way, many overlapping areas can combine their physics to make interesting effects.
• AREA_SPACE_OVERRIDE_COMBINE_REPLACE = 2 — This area adds its gravity/damp values to what-
ever has been calculated so far. Then stops taking into account the rest of the areas, even the default one.
• AREA_SPACE_OVERRIDE_REPLACE = 3 — This area replaces any gravity/damp, even the default one,
and stops taking into account the rest of the areas.
• AREA_SPACE_OVERRIDE_REPLACE_COMBINE = 4 — This area replaces any gravity/damp calculated
so far, but keeps calculating the rest of the areas, down to the default one.
• BODY_MODE_STATIC = 0
• BODY_MODE_KINEMATIC = 1
• BODY_MODE_RIGID = 2
• BODY_MODE_CHARACTER = 3
• BODY_PARAM_BOUNCE = 0
• BODY_PARAM_FRICTION = 1
• BODY_PARAM_MASS = 2
• BODY_PARAM_GRAVITY_SCALE = 3
• BODY_PARAM_ANGULAR_DAMP = 5
• BODY_PARAM_LINEAR_DAMP = 4
• BODY_PARAM_MAX = 6
• BODY_STATE_TRANSFORM = 0
• BODY_STATE_LINEAR_VELOCITY = 1
• BODY_STATE_ANGULAR_VELOCITY = 2
• BODY_STATE_SLEEPING = 3
• BODY_STATE_CAN_SLEEP = 4
• AREA_BODY_ADDED = 0
• AREA_BODY_REMOVED = 1
• INFO_ACTIVE_OBJECTS = 0
• INFO_COLLISION_PAIRS = 1
• INFO_ISLAND_COUNT = 2
9.211 PhysicsServerSW
9.212 PhysicsShapeQueryParameters
9.213 PhysicsShapeQueryResult
9.214 PinJoint
• PARAM_BIAS = 0
• PARAM_DAMPING = 1
• PARAM_IMPULSE_CLAMP = 2
9.215 PinJoint2D
Inherits: Joint2D < Node2D < CanvasItem < Node < Object
Category: Core
9.215.3 Description
Pin Joint for 2D Rigid Bodies. It pins 2 bodies (rigid or static) together, or a single body to a fixed position in space.
9.216 Plane
Vector3 center ( )
float distance_to ( Vector3 point )
Vector3 get_any_point ( )
bool has_point ( Vector3 point, float epsilon=0.00001 )
Vector3 intersect_3 ( Plane b, Plane c )
Vector3 intersects_ray ( Vector3 from, Vector3 dir )
Vector3 intersects_segment ( Vector3 begin, Vector3 end )
bool is_point_over ( Vector3 point )
Plane normalized ( )
Vector3 project ( Vector3 point )
Plane Plane ( float a, float b, float c, float d )
Plane Plane ( Vector3 v1, Vector3 v2, Vector3 v3 )
Plane Plane ( Vector3 normal, float d )
• Vector3 normal
• float x
• float y
• float z
• float d
9.216.4 Description
Plane represents a normalized plane equation. Basically, “normal” is the normal of the plane (a,b,c normalized),
and “d” is the distance from the origin to the plane (in the direction of “normal”). “Over” or “Above” the plane is
considered the side of the plane towards where the normal is pointing.
• Vector3 center ( )
Returns the center of the plane.
• float distance_to ( Vector3 point )
Returns the shortest distance from the plane to the position “point”.
• Vector3 get_any_point ( )
Returns a point on the plane.
• bool has_point ( Vector3 point, float epsilon=0.00001 )
Returns true if “point” is inside the plane (by a very minimum threshold).
• Vector3 intersect_3 ( Plane b, Plane c )
Returns the intersection point of the three planes “b”, “c” and this plane. If no intersection is found null is returned.
• Vector3 intersects_ray ( Vector3 from, Vector3 dir )
Returns the intersection point of a ray consisting of the position “from” and the direction normal “dir” with this plane.
If no intersection is found null is returned.
• Vector3 intersects_segment ( Vector3 begin, Vector3 end )
Returns the intersection point of a segment from position “begin” to position “end” with this plane. If no intersection
is found null is returned.
• bool is_point_over ( Vector3 point )
Returns true if “point” is located above the plane.
• Plane normalized ( )
Returns a copy of the plane, normalized.
• Vector3 project ( Vector3 point )
Returns the orthogonal projection of point “p” into a point in the plane.
• Plane Plane ( float a, float b, float c, float d )
Creates a plane from the three parameters “a”, “b”, “c” and “d”.
• Plane Plane ( Vector3 v1, Vector3 v2, Vector3 v3 )
Creates a plane from three points.
• Plane Plane ( Vector3 normal, float d )
Creates a plane from the normal and the plane’s distance to the origin.
9.217 PlaneShape
9.218 Polygon2D
9.219 PolygonPathFinder
9.220 Popup
9.220.3 Signals
• popup_hide ( )
• about_to_show ( )
• NOTIFICATION_POST_POPUP = 80
• NOTIFICATION_POPUP_HIDE = 81
9.220.5 Description
Popup is a base Control used to show dialogs and popups. It’s a subwindow and modal by default (see Control) and
has helpers for custom popup behavior.
9.221 PopupDialog
Inherits: Popup < Control < CanvasItem < Node < Object
Category: Core
9.222 PopupMenu
Inherits: Popup < Control < CanvasItem < Node < Object
Category: Core
void add_icon_item ( Object texture, String label, int id=-1, int accel=0 )
void add_item ( String label, int id=-1, int accel=0 )
void add_icon_check_item ( Object texture, String label, int id=-1, int accel=0 )
void add_check_item ( String label, int id=-1, int accel=0 )
void add_submenu_item ( String label, String submenu, int id=-1 )
void set_item_text ( int idx, String text )
void set_item_icon ( int idx, Object icon )
void set_item_accelerator ( int idx, int accel )
void set_item_metadata ( int idx, var metadata )
void set_item_checked ( int idx, bool checked )
void set_item_disabled ( int idx, bool disabled )
void set_item_submenu ( int idx, String submenu )
void set_item_as_separator ( int idx, bool enable )
void set_item_as_checkable ( int idx, bool enable )
void set_item_ID ( int idx, int id )
String get_item_text ( int idx ) const
Object get_item_icon ( int idx ) const
void get_item_metadata ( int idx ) const
int get_item_accelerator ( int idx ) const
String get_item_submenu ( int idx ) const
bool is_item_separator ( int idx ) const
bool is_item_checkable ( int idx ) const
bool is_item_checked ( int idx ) const
bool is_item_disabled ( int idx ) const
int get_item_ID ( int idx ) const
int get_item_index ( int id ) const
int get_item_count ( ) const
void add_separator ( )
void remove_item ( int idx )
void clear ( )
9.222.3 Signals
• item_pressed ( int ID )
9.222.4 Description
PopupMenu is the typical Control that displays a list of options. They are popular in toolbars or context menus.
• void add_icon_item ( Object texture, String label, int id=-1, int accel=0 )
Add a new item with text “label” and icon “texture”. An id can optionally be provided, as well as an accelerator. If no
id is provided, one will be created from the index.
• void add_item ( String label, int id=-1, int accel=0 )
Add a new item with text “label”. An id can optionally be provided, as well as an accelerator. If no id is provided, one
will be created from the index.
• void add_icon_check_item ( Object texture, String label, int id=-1, int accel=0 )
Add a new check able item with text “label” and icon “texture”. An id can optionally be provided, as well as an
accelerator. If no id is provided, one will be created from the index. Note that checkable items just display a checkmark,
but don’t have any built-in checking behavior and must be checked/unchecked manually.
• void add_check_item ( String label, int id=-1, int accel=0 )
Add a new checkable item with text “label”. An id can optionally be provided, as well as an accelerator. If no id is
provided, one will be created from the index. Note that checkable items just display a checkmark, but don’t have any
built-in checking behavior and must be checked/unchecked manually.
• void add_submenu_item ( String label, String submenu, int id=-1 )
• void set_item_text ( int idx, String text )
Set the text of the item at index “idx”.
• void set_item_icon ( int idx, Object icon )
Set the icon of the item at index “idx”.
• void set_item_accelerator ( int idx, int accel )
Set the accelerator of the item at index “idx”. Accelerators are special combinations of keys that activate the item, no
matter which control is focused.
• void set_item_metadata ( int idx, var metadata )
• void set_item_checked ( int idx, bool checked )
Set the checkstate status of the item at index “idx”.
• void set_item_disabled ( int idx, bool disabled )
• void set_item_submenu ( int idx, String submenu )
• void set_item_as_separator ( int idx, bool enable )
• void set_item_as_checkable ( int idx, bool enable )
• void set_item_ID ( int idx, int id )
Set the id of the item at index “idx”.
9.223 PopupPanel
Inherits: Popup < Control < CanvasItem < Node < Object
Category: Core
9.224 Portal
9.224.3 Description
Portals provide virtual openings to RoomInstance nodes, so cameras can look at them from the outside. Note that
portals are a visibility optimization technique, and are in no way related to the game of the same name (as in, they
are not used for teleportation). For more information on how rooms and portals work, see RoomInstance. Portals are
represented as 2D convex polygon shapes (in the X,Y local plane), and are placed on the surface of the areas occupied
by a RoomInstance, to indicate that the room can be accessed or looked-at through them. If two rooms are next to
each other, and two similar portals in each of them share the same world position (and are parallel and opposed to each
other), they will automatically “connect” and form “doors” (for example, the portals that connect a kitchen to a living
room are placed in the door they share). Portals must always have a RoomInstance node as a parent, grandparent or
far parent, or else they will not be active.
Return whether the portal is active. When disabled it causes the parent RoomInstance to not be visible any longer
when looking through the portal.
• void set_disable_distance ( float distance )
Set the distance threshold for disabling the portal. Every time that the portal goes beyond “distance”, it disables itself,
becoming the opaque color (see set_disabled_color).
• float get_disable_distance ( ) const
Return the distance threshold for disabling the portal. Every time that the portal goes beyond “distance”, it disables
itself, becoming the opaque color (see set_disabled_color).
• void set_disabled_color ( Color color )
When the portal goes beyond the disable distance (see set_disable_distance), it becomes opaque and displayed with
color “color”.
• Color get_disabled_color ( ) const
Return the color for when the portal goes beyond the disable distance (see set_disable_distance) and becomes disabled.
9.225 Position2D
9.225.2 Description
Generic 2D Position hint for editing. It’s just like a plain Node2D but displays as a cross in the 2D-Editor at all times.
9.226 Position3D
9.226.2 Description
Generic 3D Position hint for editing. It’s just like a plain Spatial but displays as a cross in the 3D-Editor at all times.
9.227 ProgressBar
Inherits: Range < Control < CanvasItem < Node < Object
Category: Core
9.227.3 Description
General purpose progress bar. Shows fill percentage from right to left.
9.228 ProximityGroup
9.228.3 Signals
9.228.4 Description
9.229 Quad
Inherits: GeometryInstance < VisualInstance < Spatial < Node < Object
Category: Core
9.230 Quat
Quaternion.
• float x
• float y
• float z
• float w
9.230.4 Description
Quaternion is a 4 dimensional vector that is used to represent a rotation. It mainly exists to perform SLERP (spherical-
linear interpolation) between to rotations obtained by a Matrix3 cheaply. Adding quaternions also cheaply adds the
rotations, however quaternions need to be often normalized, or else they suffer from precision issues.
Returns the inverse of the quaternion (applies to the inverse rotation too).
• float length ( )
Returns the length of the quaternion.
• float length_squared ( )
Returns the length of the quaternion, squared.
• Quat normalized ( )
Returns a copy of the quaternion, normalized to unit length.
• Quat slerp ( Quat b, float t )
Perform a spherical-linear interpolation with another quaternion.
• Quat slerpni ( Quat b, float t )
• Vector3 xform ( Vector3 v )
• Quat Quat ( float x, float y, float z, float w )
• Quat Quat ( Vector3 axis, float angle )
• Quat Quat ( Matrix3 from )
9.231 Range
9.231.3 Signals
9.231.4 Description
Range is a base class for Control nodes that change a floating point value between a minimum and a maximum, using
step and page, for example a ScrollBar.
9.232 RawArray
String get_string_from_ascii ( )
String get_string_from_utf8 ( )
void push_back ( int byte )
void resize ( int idx )
void set ( int idx, int byte )
int size ( )
RawArray RawArray ( Array from )
9.232.3 Description
Raw byte array. Contains bytes. Optimized for memory usage, can’t fragment the memory.
• String get_string_from_ascii ( )
Returns a copy of the array’s contents formatted as String. Fast alternative to get_string_from_utf8(), assuming the
content is ASCII-only (unlike the UTF-8 function, this function maps every byte to a character in the string, so any
multibyte sequence will be torn apart).
• String get_string_from_utf8 ( )
Returns a copy of the array’s contents formatted as String, assuming the array is formatted as UTF-8. Slower than
get_string_from_ascii(), but works for UTF-8. Usually you should prefer this function over get_string_from_ascii()
to support international input.
• void push_back ( int byte )
• void resize ( int idx )
• void set ( int idx, int byte )
• int size ( )
• RawArray RawArray ( Array from )
9.233 RayCast
9.234 RayCast2D
9.235 RayShape
9.236 RayShape2D
9.236.3 Description
Ray 2D shape resource for physics. A ray is not really a collision body, instead it tries to separate itself from whatever
is touching its far endpoint. It’s often useful for characters.
9.237 RealArray
Real Array .
9.237.3 Description
Real Array. Array of floating point values. Can only contain floats. Optimized for memory usage, can’t fragment the
memory.
9.238 Rect2
• Vector2 pos
• Vector2 size
• Vector2 end
9.238.4 Description
Rect2 provides an 2D Axis-Aligned Bounding Box. It consists of a position, a size, and several utility functions. It is
typically used for fast overlap tests.
9.239 RectangleShape2D
9.239.3 Description
Rectangle Shape for 2D Physics. This shape is useful for modeling box-like 2D objects.
9.240 Reference
Inherits: Object
Inherited By: RegEx, SurfaceTool, EditorScenePostImport, PhysicsShapeQueryResult, Physics2DTestMotionResult,
FuncRef , File, TCP_Server, Physics2DShapeQueryResult, ConfigFile, StreamPeer, HTTPClient, AudioStream-
Playback, MeshDataTool, GDFunctionState, Physics2DShapeQueryParameters, EditorScript, Mutex, PacketPeer,
Semaphore, XMLParser, EditorImportPlugin, Directory, Marshalls, WeakRef , SceneState, GDNativeClass, PCK-
Packer, Resource, Thread, PackedDataContainerRef , ResourceInteractiveLoader, ResourceImportMetadata, Physic-
sShapeQueryParameters
Category: Core
bool init_ref ( )
void reference ( )
bool unreference ( )
9.240.3 Description
Base class for anything that keeps a reference count. Resource and many other helper objects inherit this. References
keep an internal reference counter so they are only released when no longer in use.
• bool init_ref ( )
• void reference ( )
Increase the internal reference counter. Use this only if you really know what you are doing.
• bool unreference ( )
Decrease the internal reference counter. Use this only if you really know what you are doing.
9.241 ReferenceFrame
9.241.2 Description
Reference frame for GUI. It’s just like an empty control, except a red box is displayed while editing around its size at
all times.
9.242 RegEx
9.242.3 Description
Class for finding text patterns in a string using regular expressions. Regular expressions are a way to define patterns
of text to be searched.
This class only finds patterns in a string. It can not perform replacements.
Usage of regular expressions is too long to be explained here, but Internet is full of tutorials and detailed explanations.
Currently supported features:
Capturing () and non-capturing (?:) groups
Any character .
Shorthand character classes \w \W \s \S \d \D
User-defined character classes such as :ref:‘A-Za-z<class_a-za-z>‘
Simple quantifiers ?, \* and +
Range quantifiers {x,y}
Lazy (non-greedy) quantifiers \*?
Beginning ^ and end $ anchors
Alternation |
Backreferences \1 and \g{1}
POSIX character classes :ref:‘[:alnum:<class_[:alnum:>‘]
Lookahead (?=), (?!) and lookbehind (?<=), (?<!)
ASCII \xFF and Unicode \uFFFF code points (in a style similar to Python)
Word boundaries \b, \B
This method resets the state of the object, as it was freshly created. Namely, it unassigns the regular expression of this
object, and forgets all captures made by the last find.
• bool is_valid ( ) const
Returns whether this object has a valid regular expression assigned.
• int get_capture_count ( ) const
Returns the number of capturing groups. A captured group is the part of a string that matches a part of the pattern
delimited by parentheses (unless they are non-capturing parentheses (?:)).
• String get_capture ( int capture ) const
Returns a captured group. A captured group is the part of a string that matches a part of the pattern delimited by
parentheses (unless they are non-capturing parentheses (?:)).
• StringArray get_captures ( ) const
Return a list of all the captures made by the regular expression.
9.243 RemoteTransform2D
9.244 RenderTargetTexture
9.245 Resource
Inherited By: Theme, AudioStream, EventStream, CubeMap, Translation, Curve2D, Shape, Shape2D, BakedLight,
ColorRamp, StyleBox, Environment, Material, VideoStream, RoomBounds, PackedScene, Texture, Script, Occluder-
Polygon2D, Mesh, TileSet, BitMap, Animation, Sample, PolygonPathFinder, Shader, World, SampleLibrary, World2D,
Font, SpriteFrames, MeshLibrary, Curve3D, NavigationPolygon, MultiMesh, CanvasItemMaterial, PackedDataCon-
tainer, NavigationMesh
Category: Core
9.245.3 Signals
• changed ( )
9.245.4 Description
Resource is the base class for all resource types. Resources are primarily data containers. They are reference counted
and freed when no longer in use. They are also loaded only once from disk, and further attempts to load the resource
will return the same reference (all this in contrast to a Node, which is not reference counted and can be instanced from
disk as many times as desired). Resources can be saved externally on disk or bundled into another object, such as a
Node or another resource.
9.246 ResourceImportMetadata
9.247 ResourceInteractiveLoader
Object get_resource ( )
int poll ( )
int wait ( )
int get_stage ( ) const
int get_stage_count ( ) const
9.247.3 Description
Interactive Resource Loader. This object is returned by ResourceLoader when performing an interactive load. It allows
to load with high granularity, so this is mainly useful for displaying load bars/percentages.
• Object get_resource ( )
Return the loaded resource (only if loaded). Otherwise, returns null.
• int poll ( )
Poll the load. If OK is returned, this means poll will have to be called again. If ERR_EOF is returned, them the load
has finished and the resource can be obtained by calling get_resource.
• int wait ( )
• int get_stage ( ) const
Return the load stage. The total amount of stages can be queried with get_stage_count
• int get_stage_count ( ) const
Return the total amount of stages (calls to poll) needed to completely load this resource.
9.248 ResourceLoader
Inherits: Object
Category: Core
Resource Loader.
9.248.3 Description
Resource Loader. This is a static object accessible as ResourceLoader. GDScript has a simplified load() function,
though.
9.249 ResourcePreloader
9.249.3 Description
Resource Preloader Node. This node is used to preload sub-resources inside a scene, so when the scene is loaded all
the resources are ready to use and be retrieved from here.
9.250 ResourceSaver
Inherits: Object
Category: Core
• FLAG_RELATIVE_PATHS = 1
• FLAG_BUNDLE_RESOURCES = 2
• FLAG_CHANGE_PATH = 4
• FLAG_OMIT_EDITOR_PROPERTIES = 8
• FLAG_SAVE_BIG_ENDIAN = 16
• FLAG_COMPRESS = 32
9.250.4 Description
Resource Saving Interface. This interface is used for saving resources to disk.
9.251 RichTextLabel
9.251.3 Signals
• ALIGN_LEFT = 0
• ALIGN_CENTER = 1
• ALIGN_RIGHT = 2
• ALIGN_FILL = 3
• LIST_NUMBERS = 0
• LIST_LETTERS = 1
• LIST_DOTS = 2
• ITEM_FRAME = 0
• ITEM_TEXT = 1
• ITEM_IMAGE = 2
• ITEM_NEWLINE = 3
• ITEM_FONT = 4
• ITEM_COLOR = 5
• ITEM_UNDERLINE = 6
• ITEM_ALIGN = 7
• ITEM_INDENT = 8
• ITEM_LIST = 9
• ITEM_META = 11
9.251.5 Description
Label that displays rich text. Rich text can contain custom text, fonts, images and some basic formatting. It also adapts
itself to given width/heights.
9.252 RID
int get_id ( )
RID RID ( Object from )
• int get_id ( )
• RID RID ( Object from )
9.253 RigidBody
Inherits: PhysicsBody < CollisionObject < Spatial < Node < Object
Category: Core
9.253.3 Signals
• MODE_STATIC = 1
• MODE_KINEMATIC = 3
• MODE_RIGID = 0
• MODE_CHARACTER = 2
9.254 RigidBody2D
Inherits: PhysicsBody2D < CollisionObject2D < Node2D < CanvasItem < Node < Object
Category: Core
9.254.3 Signals
• MODE_STATIC = 1 — Static mode. The body behaves like a StaticBody2D, and can only move by user code.
• MODE_KINEMATIC = 3 — Kinematic body. The body behaves like a KinematicBody2D, and can only move
by user code.
• MODE_RIGID = 0 — Rigid body. This is the “natural” state of a rigid body. It is affected by forces, and can
move, rotate, and be affected by user code.
• MODE_CHARACTER = 2 — Character body. This behaves like a rigid body, but can not rotate.
• CCD_MODE_DISABLED = 0 — Disables continuous collision detection. This is the fastest way to detect
body collisions, but can miss small, fast-moving objects.
• CCD_MODE_CAST_RAY = 1 — Enables continuous collision detection by raycasting. It is faster than
shapecasting, but less precise.
• CCD_MODE_CAST_SHAPE = 2 — Enables continuous collision detection by shapecasting. It is the slowest
CCD method, and the most precise.
9.254.5 Description
Rigid body 2D node. This node is used for placing rigid bodies in the scene. It can contain a number of shapes, and
also shift state between regular Rigid body, Kinematic, Character or Static.
Character mode forbids the node from being rotated. This node can have a custom force integrator function, for writing
complex physics motion behavior per node.
As a warning, don’t change this node position every frame or very often. Sporadic changes work fine, but physics runs
at a different granularity (fixed hz) than usual rendering (process callback) and maybe even in a separate thread, so
changing this from a process loop will yield strange behavior.
Set an axis velocity. The velocity in the given vector axis will be set as the given vector length. This is useful for
jumping behavior.
• void apply_impulse ( Vector2 pos, Vector2 impulse )
Apply a positioned impulse (which will be affected by the body mass and shape). This is the equivalent of hitting a
billiard ball with a cue: a force that is applied once, and only once.
• void set_applied_force ( Vector2 force )
Set the applied force vector. This is the equivalent of pushing a box over the ground: the force applied is applied
constantly.
• Vector2 get_applied_force ( ) const
Return the applied force vector.
• void set_sleeping ( bool sleeping )
Set whether a body is sleeping or not. Sleeping bodies are not affected by forces until a collision or an apply_impulse
/ set_applied_force wakes them up. Until then, they behave like a static body.
• bool is_sleeping ( ) const
Return whether the body is sleeping.
• void set_can_sleep ( bool able_to_sleep )
Set the body ability to fall asleep when not moving. This saves an enormous amount of processor time when there are
plenty of rigid bodies (non static) in a scene.
Sleeping bodies are not affected by forces until a collision or an apply_impulse / set_applied_force wakes them up.
Until then, they behave like a static body.
• bool is_able_to_sleep ( ) const
Return true if the body has the ability to fall asleep when not moving. See set_can_sleep.
• bool test_motion ( Vector2 motion, float margin=0.08, Physics2DTestMotionResult result=NULL )
Return whether the body would collide, if it tried to move in the given vector. This method allows two extra parameters:
A margin, which increases slightly the size of the shapes involved in the collision detection, and an object of type
Physics2DTestMotionResult, which will store additional information about the collision (should there be one).
• Array get_colliding_bodies ( ) const
Return a list of the bodies colliding with this one.
9.255 Room
9.255.3 Description
Room contains the data to define the bounds of a scene (using a BSP Tree). It is instanced by a RoomInstance node to
create rooms. See that class documentation for more information about rooms.
9.256 RoomBounds
• void regenerate_bsp ( )
• void regenerate_bsp_cubic ( )
9.257 Sample
9.257.4 Description
Sample provides an audio sample class, containing audio data, together with some information for playback, such as
format, mix rate and loop. It is used by sound playback routines.
9.258 SampleLibrary
9.258.3 Description
Library that contains a collection of Sample, each identified by a text ID. This is used as a data container for the
majority of the SamplePlayer classes and derivatives.
9.259 SamplePlayer
9.259.4 Description
SamplePlayer is a Node meant for simple sample playback. A library of samples is loaded and played back “as is”,
without positioning or anything.
9.260 SamplePlayer2D
Inherits: SoundPlayer2D < Node2D < CanvasItem < Node < Object
Category: Core
9.260.4 Description
Sample player for positional 2D Sound. Plays sound samples positionally, left and right depending on the dis-
tance/place on the screen.
9.261 SceneState
9.262 SceneTree
9.262.3 Signals
• screen_resized ( )
• GROUP_CALL_DEFAULT = 0
• GROUP_CALL_REVERSE = 1
• GROUP_CALL_REALTIME = 2
• GROUP_CALL_UNIQUE = 4
• STRETCH_MODE_DISABLED = 0
• STRETCH_MODE_2D = 1
• STRETCH_MODE_VIEWPORT = 2
• STRETCH_ASPECT_IGNORE = 0
• STRETCH_ASPECT_KEEP = 1
• STRETCH_ASPECT_KEEP_WIDTH = 2
• STRETCH_ASPECT_KEEP_HEIGHT = 3
9.263 Script
9.263.3 Description
Base class for scripts. Any script that is loaded becomes one of these resources, which can then create instances.
9.264 ScrollBar
Inherits: Range < Control < CanvasItem < Node < Object
Inherited By: HScrollBar, VScrollBar
Category: Core
9.264.3 Description
Scrollbars are a Range based Control, that display a draggable area (the size of the page). Horizontal (HScrollBar)
and Vertical (VScrollBar) versions are available.
9.265 ScrollContainer
Inherits: Container < Control < CanvasItem < Node < Object
Category: Core
9.265.3 Description
A ScrollContainer node with a Control child and scrollbar child (HScrollbar, VScrollBar, or both) will only draw the
Control within the ScrollContainer area. Scrollbars will automatically be drawn at the right (for vertical) or bottom
(for horizontal) and will enable dragging to move the viewable Control (and its children) within the ScrollContainer.
Scrollbars will also automatically resize the grabber based on the minimum_size of the Control relative to the Scroll-
Container. Works great with a Panel control.
9.266 SegmentShape2D
9.266.3 Description
Segment Shape for 2D Collision Detection, consists of two points, ‘a’ and ‘b’.
9.267 Semaphore
Error wait ( )
Error post ( )
• Error wait ( )
• Error post ( )
9.268 Separator
9.268.2 Description
Separator is a Control used for separating other controls. It’s purely a visual decoration. Horizontal (HSeparator) and
Vertical (VSeparator) versions are available.
9.269 Shader
To be changed, ignore.
• MODE_MATERIAL = 0
• MODE_CANVAS_ITEM = 1
• MODE_POST_PROCESS = 2
9.269.4 Description
To be changed, ignore.
9.270 ShaderGraph
9.270.3 Signals
• updated ( )
• NODE_INPUT = 0
• NODE_SCALAR_CONST = 1
• NODE_VEC_CONST = 2
• NODE_RGB_CONST = 3
• NODE_XFORM_CONST = 4
• NODE_TIME = 5
• NODE_SCREEN_TEX = 6
• NODE_SCALAR_OP = 7
• NODE_VEC_OP = 8
• NODE_VEC_SCALAR_OP = 9
• NODE_RGB_OP = 10
• NODE_XFORM_MULT = 11
• NODE_XFORM_VEC_MULT = 12
• NODE_XFORM_VEC_INV_MULT = 13
• NODE_SCALAR_FUNC = 14
• NODE_VEC_FUNC = 15
• NODE_VEC_LEN = 16
• NODE_DOT_PROD = 17
• NODE_VEC_TO_SCALAR = 18
• NODE_SCALAR_TO_VEC = 19
• NODE_VEC_TO_XFORM = 21
• NODE_XFORM_TO_VEC = 20
• NODE_SCALAR_INTERP = 22
• NODE_VEC_INTERP = 23
• NODE_COLOR_RAMP = 24
• NODE_CURVE_MAP = 25
• NODE_SCALAR_INPUT = 26
• NODE_VEC_INPUT = 27
• NODE_RGB_INPUT = 28
• NODE_XFORM_INPUT = 29
• NODE_TEXTURE_INPUT = 30
• NODE_CUBEMAP_INPUT = 31
• NODE_DEFAULT_TEXTURE = 32
• NODE_OUTPUT = 33
• NODE_COMMENT = 34
• NODE_TYPE_MAX = 35
• SLOT_TYPE_SCALAR = 0
• SLOT_TYPE_VEC = 1
• SLOT_TYPE_XFORM = 2
• SLOT_TYPE_TEXTURE = 3
• SLOT_MAX = 4
• SHADER_TYPE_VERTEX = 0
• SHADER_TYPE_FRAGMENT = 1
• SHADER_TYPE_LIGHT = 2
• SHADER_TYPE_MAX = 3
• SLOT_IN = 0
• SLOT_OUT = 1
• GRAPH_OK = 0
• GRAPH_ERROR_CYCLIC = 1
• GRAPH_ERROR_MISSING_CONNECTIONS = 2
• SCALAR_OP_ADD = 0
• SCALAR_OP_SUB = 1
• SCALAR_OP_MUL = 2
• SCALAR_OP_DIV = 3
• SCALAR_OP_MOD = 4
• SCALAR_OP_POW = 5
• SCALAR_OP_MAX = 6
• SCALAR_OP_MIN = 7
• SCALAR_OP_ATAN2 = 8
• SCALAR_MAX_OP = 9
• VEC_OP_ADD = 0
• VEC_OP_SUB = 1
• VEC_OP_MUL = 2
• VEC_OP_DIV = 3
• VEC_OP_MOD = 4
• VEC_OP_POW = 5
• VEC_OP_MAX = 6
• VEC_OP_MIN = 7
• VEC_OP_CROSS = 8
• VEC_MAX_OP = 9
• VEC_SCALAR_OP_MUL = 0
• VEC_SCALAR_OP_DIV = 1
• VEC_SCALAR_OP_POW = 2
• VEC_SCALAR_MAX_OP = 3
• RGB_OP_SCREEN = 0
• RGB_OP_DIFFERENCE = 1
• RGB_OP_DARKEN = 2
• RGB_OP_LIGHTEN = 3
• RGB_OP_OVERLAY = 4
• RGB_OP_DODGE = 5
• RGB_OP_BURN = 6
• RGB_OP_SOFT_LIGHT = 7
• RGB_OP_HARD_LIGHT = 8
• RGB_MAX_OP = 9
• SCALAR_FUNC_SIN = 0
• SCALAR_FUNC_COS = 1
• SCALAR_FUNC_TAN = 2
• SCALAR_FUNC_ASIN = 3
• SCALAR_FUNC_ACOS = 4
• SCALAR_FUNC_ATAN = 5
• SCALAR_FUNC_SINH = 6
• SCALAR_FUNC_COSH = 7
• SCALAR_FUNC_TANH = 8
• SCALAR_FUNC_LOG = 9
• SCALAR_FUNC_EXP = 10
• SCALAR_FUNC_SQRT = 11
• SCALAR_FUNC_ABS = 12
• SCALAR_FUNC_SIGN = 13
• SCALAR_FUNC_FLOOR = 14
• SCALAR_FUNC_ROUND = 15
• SCALAR_FUNC_CEIL = 16
• SCALAR_FUNC_FRAC = 17
• SCALAR_FUNC_SATURATE = 18
• SCALAR_FUNC_NEGATE = 19
• SCALAR_MAX_FUNC = 20
• VEC_FUNC_NORMALIZE = 0
• VEC_FUNC_SATURATE = 1
• VEC_FUNC_NEGATE = 2
• VEC_FUNC_RECIPROCAL = 3
• VEC_FUNC_RGB2HSV = 4
• VEC_FUNC_HSV2RGB = 5
• VEC_MAX_FUNC = 6
9.271 ShaderMaterial
9.272 Shape
9.273 Shape2D
9.273.3 Description
Base class for all 2D Shapes. All 2D shape types inherit from this.
This method needs the transformation matrix for this shape (local_xform), the movement to test on this shape
(local_motion), the shape to check collisions with (with_shape), the transformation matrix of that shape
(shape_xform), and the movement to test onto the other object (shape_motion).
9.274 Skeleton
• NOTIFICATION_UPDATE_SKELETON = 50
9.274.4 Description
Skeleton provides a hierarchical interface for managing bones, including pose, rest and animation (see Animation).
Skeleton will support rag doll dynamics in the future.
9.275 Slider
Inherits: Range < Control < CanvasItem < Node < Object
Inherited By: HSlider, VSlider
Category: Core
9.275.3 Description
9.276 SliderJoint
• PARAM_LINEAR_LIMIT_UPPER = 0
• PARAM_LINEAR_LIMIT_LOWER = 1
• PARAM_LINEAR_LIMIT_SOFTNESS = 2
• PARAM_LINEAR_LIMIT_RESTITUTION = 3
• PARAM_LINEAR_LIMIT_DAMPING = 4
• PARAM_LINEAR_MOTION_SOFTNESS = 5
• PARAM_LINEAR_MOTION_RESTITUTION = 6
• PARAM_LINEAR_MOTION_DAMPING = 7
• PARAM_LINEAR_ORTHOGONAL_SOFTNESS = 8
• PARAM_LINEAR_ORTHOGONAL_RESTITUTION = 9
• PARAM_LINEAR_ORTHOGONAL_DAMPING = 10
• PARAM_ANGULAR_LIMIT_UPPER = 11
• PARAM_ANGULAR_LIMIT_LOWER = 12
• PARAM_ANGULAR_LIMIT_SOFTNESS = 13
• PARAM_ANGULAR_LIMIT_RESTITUTION = 14
• PARAM_ANGULAR_LIMIT_DAMPING = 15
• PARAM_ANGULAR_MOTION_SOFTNESS = 16
• PARAM_ANGULAR_MOTION_RESTITUTION = 17
• PARAM_ANGULAR_MOTION_DAMPING = 18
• PARAM_ANGULAR_ORTHOGONAL_SOFTNESS = 19
• PARAM_ANGULAR_ORTHOGONAL_RESTITUTION = 20
• PARAM_ANGULAR_ORTHOGONAL_DAMPING = 21
• PARAM_MAX = 22
9.277 SoundPlayer2D
• PARAM_VOLUME_DB = 0
• PARAM_PITCH_SCALE = 1
• PARAM_ATTENUATION_MIN_DISTANCE = 2
• PARAM_ATTENUATION_MAX_DISTANCE = 3
• PARAM_ATTENUATION_DISTANCE_EXP = 4
• PARAM_MAX = 5
9.277.4 Description
9.278 SoundRoomParams
9.279 Spatial
9.279.3 Signals
• visibility_changed ( )
9.279.5 Description
Spatial is the base for every type of 3D Node. It contains a 3D Transform which can be set or get as local or global. If
a Spatial Node has Spatial children, their transforms will be relative to the parent.
• void orthonormalize ( )
• void set_identity ( )
• void look_at ( Vector3 target, Vector3 up )
• void look_at_from_pos ( Vector3 pos, Vector3 target, Vector3 up )
9.280 SpatialPlayer
• PARAM_VOLUME_DB = 0
• PARAM_PITCH_SCALE = 1
• PARAM_ATTENUATION_MIN_DISTANCE = 2
• PARAM_ATTENUATION_MAX_DISTANCE = 3
• PARAM_ATTENUATION_DISTANCE_EXP = 4
• PARAM_EMISSION_CONE_DEGREES = 5
• PARAM_EMISSION_CONE_ATTENUATION_DB = 6
• PARAM_MAX = 7
9.281 SpatialSamplePlayer
• INVALID_VOICE = -1
• NEXT_VOICE = -2
9.282 SpatialSound2DServer
Inherits: Object
Inherited By: SpatialSound2DServerSW
Category: Core
9.282.2 Description
9.283 SpatialSound2DServerSW
9.284 SpatialSoundServer
Inherits: Object
Inherited By: SpatialSoundServerSW
Category: Core
9.285 SpatialSoundServerSW
9.286 SpatialStreamPlayer
9.287 SphereShape
9.288 SpinBox
Inherits: Range < Control < CanvasItem < Node < Object
Category: Core
9.288.3 Description
SpinBox is a numerical input text field. It allows entering integers and floats.
9.289 SplitContainer
Inherits: Container < Control < CanvasItem < Node < Object
Inherited By: HSplitContainer, VSplitContainer
Category: Core
9.289.3 Signals
• DRAGGER_VISIBLE = 0
• DRAGGER_HIDDEN = 1
• DRAGGER_HIDDEN_COLLAPSED = 2
9.289.5 Description
Container for splitting two controls vertically or horizontally, with a grabber that allows adjusting the split offset or
ratio.
9.290 SpotLight
Inherits: Light < VisualInstance < Spatial < Node < Object
Category: Core
9.290.2 Description
A SpotLight light is a type of Light node that emits lights in a specific direction, in the shape of a cone. The light is
attenuated through the distance and this attenuation can be configured by changing the energy, radius and attenuation
parameters of Light. TODO: Image of a spotlight.
9.291 Sprite
9.291.3 Signals
• frame_changed ( )
9.291.4 Description
General purpose Sprite node. This Sprite node can show any texture as a sprite. The texture can be used as a spritesheet
for animation, or only a region from a bigger texture can referenced, like an atlas.
Return the texture frame for a sprite-sheet, works when vframes or hframes are greater than 1.
• void set_vframes ( int vframes )
Set the amount of vertical frames and converts the sprite into a sprite-sheet. This is useful for animation.
• int get_vframes ( ) const
Return the amount of vertical frames. See set_vframes.
• void set_hframes ( int hframes )
Set the amount of horizontal frames and converts the sprite into a sprite-sheet. This is useful for animation.
• int get_hframes ( ) const
Return the amount of horizontal frames. See set_hframes.
• void set_modulate ( Color modulate )
Set color modulation for the sprite. All sprite pixels are multiplied by this color. Color may contain rgb values above
1 to achieve a highlight effect.
• Color get_modulate ( ) const
Return color modulation for the sprite. All sprite pixels are multiplied by this color.
9.292 Sprite3D
Inherits: SpriteBase3D < GeometryInstance < VisualInstance < Spatial < Node < Object
Category: Core
9.292.3 Signals
• frame_changed ( )
9.293 SpriteBase3D
Inherits: GeometryInstance < VisualInstance < Spatial < Node < Object
Inherited By: AnimatedSprite3D, Sprite3D
Category: Core
• FLAG_TRANSPARENT = 0
• FLAG_SHADED = 1
• FLAG_MAX = 2
• ALPHA_CUT_DISABLED = 0
• ALPHA_CUT_DISCARD = 1
• ALPHA_CUT_OPAQUE_PREPASS = 2
9.294 SpriteFrames
9.294.3 Description
9.295 StaticBody
Inherits: PhysicsBody < CollisionObject < Spatial < Node < Object
Category: Core
9.295.3 Description
StaticBody implements a static collision Node, by utilizing a rigid body in the PhysicsServer. Static bodies are used
for static collision. For more information on physics body nodes, see PhysicsBody.
9.296 StaticBody2D
Inherits: PhysicsBody2D < CollisionObject2D < Node2D < CanvasItem < Node < Object
Category: Core
9.296.3 Description
Static body for 2D Physics. A static body is a simple body that is not intended to move. They don’t consume any CPU
resources in contrast to a RigidBody2D so they are great for scenaro collision.
A static body can also be animated by using simulated motion mode. This is useful for implementing functionalities
such as moving platforms. When this mode is active the body can be animated and automatically computes linear and
angular velocity to apply in that frame and to influence other bodies.
Alternatively, a constant linear or angular velocity can be set for the static body, so even if it doesn’t move, it affects
other bodies as if it was moving (this is useful for simulating conveyor belts or conveyor wheels).
9.297 StreamPeer
9.297.3 Description
StreamPeer is an abstraction and base class for stream-based protocols (such as TCP or Unix Sockets). It provides an
API for sending and receiving data through streams as raw data or strings.
9.298 StreamPeerSSL
• STATUS_DISCONNECTED = 0
• STATUS_CONNECTED = 1
• STATUS_ERROR_NO_CERTIFICATE = 2
• STATUS_ERROR_HOSTNAME_MISMATCH = 3
9.299 StreamPeerTCP
• STATUS_NONE = 0
• STATUS_CONNECTING = 1
• STATUS_CONNECTED = 2
• STATUS_ERROR = 3
9.299.4 Description
TCP Stream peer. This object can be used to connect to TCP servers, or also is returned by a tcp server.
9.300 StreamPlayer
9.300.3 Signals
• finished ( )
9.300.4 Description
Base class for audio stream playback. Audio stream players inherit from it.
9.301 String
String basename ( )
bool begins_with ( String text )
String c_escape ( )
String c_unescape ( )
String capitalize ( )
int casecmp_to ( String to )
bool empty ( )
String extension ( )
int find ( String what, int from=0 )
int find_last ( String what )
int findn ( String what, int from=0 )
String get_base_dir ( )
String get_file ( )
int hash ( )
int hex_to_int ( )
String insert ( int pos, String what )
bool is_abs_path ( )
bool is_rel_path ( )
bool is_valid_float ( )
bool is_valid_html_color ( )
bool is_valid_identifier ( )
bool is_valid_integer ( )
bool is_valid_ip_address ( )
String json_escape ( )
String left ( int pos )
int length ( )
bool match ( String expr )
bool matchn ( String expr )
RawArray md5_buffer ( )
String md5_text ( )
int nocasecmp_to ( String to )
String ord_at ( int at )
String pad_decimals ( int digits )
String pad_zeros ( int digits )
String percent_decode ( )
String percent_encode ( )
String plus_file ( String file )
String replace ( String what, String forwhat )
String replacen ( String what, String forwhat )
int rfind ( String what, int from=-1 )
int rfindn ( String what, int from=-1 )
String right ( int pos )
StringArray split ( String divisor, bool allow_empty=True )
RealArray split_floats ( String divisor, bool allow_empty=True )
String strip_edges ( )
String substr ( int from, int len )
Continued on next page
9.301.3 Description
This is the built-in string class (and the one used by GDScript). It supports Unicode and provides all necessary means
for string handling. Strings are reference counted and use a copy-on-write approach, so passing them around is cheap
in resources.
• String basename ( )
If the string is a path to a file, return the path to the file without the extension.
• bool begins_with ( String text )
Return true if the strings begins with the given string.
• String c_escape ( )
• String c_unescape ( )
• String capitalize ( )
Return the string in uppercase.
• int casecmp_to ( String to )
Perform a case-sensitive comparison to another string, return -1 if less, 0 if equal and +1 if greater.
• bool empty ( )
Return true if the string is empty.
• String extension ( )
If the string is a path to a file, return the extension.
• int find ( String what, int from=0 )
Find the first occurrence of a substring, return the starting position of the substring or -1 if not found. Optionally, the
initial search index can be passed.
• int find_last ( String what )
Find the last occurrence of a substring, return the starting position of the substring or -1 if not found. Optionally, the
initial search index can be passed.
• int findn ( String what, int from=0 )
Find the first occurrence of a substring but search as case-insensitive, return the starting position of the substring or -1
if not found. Optionally, the initial search index can be passed.
• String get_base_dir ( )
9.302 StringArray
String Array.
9.302.3 Description
String Array. Array of strings. Can only contain strings. Optimized for memory usage, can’t fragment the memory.
9.303 StyleBox
9.303.3 Description
StyleBox is Resource that provides an abstract base class for drawing stylized boxes for the UI. StyleBoxes are used
for drawing the styles of buttons, line edit backgrounds, tree backgrounds, etc. and also for testing a transparency
mask for pointer signals. If mask test fails on a StyleBox assigned as mask to a control, clicks and motion signals will
go through it to the one below.
9.304 StyleBoxEmpty
9.304.2 Description
9.305 StyleBoxFlat
9.305.3 Description
Stylebox of a single color. Displays the stylebox of a single color, alternatively a border with light/dark colors can be
assigned.
9.306 StyleBoxImageMask
9.306.3 Description
This StyleBox is similar to StyleBoxTexture, but only meant to be used for mask testing. It takes an image and applies
stretch rules to determine if the point clicked is masked or not.
9.307 StyleBoxTexture
9.307.3 Description
Texture Based 3x3 scale style. This stylebox performs a 3x3 scaling of a texture, where only the center cell is fully
stretched. This allows for the easy creation of bordered styles.
9.308 SurfaceTool
9.308.3 Description
9.309 TabContainer
Tabbed Container.
9.309.3 Signals
• pre_popup_pressed ( )
• tab_changed ( int tab )
9.309.4 Description
Tabbed Container. Contains several children controls, but shows only one at the same time. Clicking on the top tabs
allows to change the currently visible one.
Children controls of this one automatically.
9.310 Tabs
Tabs Control.
9.310.3 Signals
• ALIGN_LEFT = 0
• ALIGN_CENTER = 1
• ALIGN_RIGHT = 2
• CLOSE_BUTTON_SHOW_ACTIVE_ONLY = 1
• CLOSE_BUTTON_SHOW_ALWAYS = 2
• CLOSE_BUTTON_SHOW_NEVER = 0
9.310.5 Description
Simple tabs control, similar to TabContainer but is only in charge of drawing tabs, not interact with children.
9.311 TCP_Server
TCP Server.
9.311.3 Description
TCP Server class. Listens to connections on a port and returns a StreamPeerTCP when got a connection.
9.312 TestCube
Inherits: GeometryInstance < VisualInstance < Spatial < Node < Object
Category: Core
9.313 TextEdit
9.313.3 Signals
• text_changed ( )
• cursor_changed ( )
• request_completion ( )
9.313.5 Description
TextEdit is meant for editing large, multiline text. It also has facilities for editing code, such as syntax highlighting
support and multiple levels of undo/redo.
• void select_all ( )
Select all the text.
• void select ( int from_line, int from_column, int to_line, int to_column )
Perform selection, from line/column to line/column.
• bool is_selection_active ( ) const
Return true if the selection is active.
• int get_selection_from_line ( ) const
Return the selection begin line.
• int get_selection_from_column ( ) const
Return the selection begin column.
• int get_selection_to_line ( ) const
Return the selection end line.
• int get_selection_to_column ( ) const
Return the selection end column.
• String get_selection_text ( ) const
Return the text inside the selection.
• String get_word_under_cursor ( ) const
• IntArray search ( String flags, int from_line, int from_column, int to_line ) const
Perform a search inside the text. Search flags can be specified in the SEARCH_* enum.
• void undo ( )
Perform undo operation.
• void redo ( )
Perform redo operation.
• void clear_undo_history ( )
Clear the undo history.
• void set_syntax_coloring ( bool enable )
Set to enable the syntax coloring.
• bool is_syntax_coloring_enabled ( ) const
Return true if the syntax coloring is enabled.
• void add_keyword_color ( String keyword, Color color )
Add a keyword and its color.
• void add_color_region ( String begin_key, String end_key, Color color, bool line_only=false )
Add color region (given the delimiters) and its colors.
• void set_symbol_color ( Color color )
Set the color for symbols.
• void set_custom_bg_color ( Color color )
Set a custom background color. A background color with alpha==0 disables this.
• void clear_colors ( )
Clear all the syntax coloring information.
9.314 Texture
9.314.4 Description
A texture works by registering an image in the video hardware, which then can be used in 3D models or 2D Sprite or
GUI Control.
9.315 TextureButton
Inherits: BaseButton < Control < CanvasItem < Node < Object
Category: Core
9.315.3 Description
Button that can be themed with textures. This is like a regular Button but can be themed by assigning textures to it.
This button is intended to be easy to theme, however a regular button can expand (that uses styleboxes) and still be
better if the interface is expect to have internationalization of texts.
Only the normal texture is required, the others are optional.
9.316 TextureFrame
9.316.3 Description
Control frame that simply draws an assigned texture. It can stretch or not. It’s a simple way to just show an image in
a UI.
9.317 TextureProgress
Inherits: Range < Control < CanvasItem < Node < Object
Category: Core
• FILL_LEFT_TO_RIGHT = 0
• FILL_RIGHT_TO_LEFT = 1
• FILL_TOP_TO_BOTTOM = 2
• FILL_BOTTOM_TO_TOP = 3
• FILL_CLOCKWISE = 4
• FILL_COUNTER_CLOCKWISE = 5
9.317.4 Description
ProgressBar implementation that is easier to theme (by just passing a few textures).
• Vector2 get_radial_center_offset ( )
• void set_fill_degrees ( float mode )
• float get_fill_degrees ( )
9.318 Theme
9.318.3 Description
Theme for skinning controls. Controls can be skinned individually, but for complex applications it’s more efficient to
just create a global theme that defines everything. This theme can be applied to any Control, and it and its children
will automatically use it.
Theme resources can be alternatively loaded by writing them in a .theme file, see wiki for more info.
9.319 Thread
Error start ( Object instance, String method, var userdata=NULL, int priority=1 )
String get_id ( ) const
bool is_active ( ) const
Variant wait_to_finish ( )
• PRIORITY_LOW = 0
• PRIORITY_NORMAL = 1
• PRIORITY_HIGH = 2
• Error start ( Object instance, String method, var userdata=NULL, int priority=1 )
• String get_id ( ) const
• bool is_active ( ) const
• Variant wait_to_finish ( )
9.320 TileMap
9.320.3 Signals
• settings_changed ( )
9.320.5 Description
Node for 2D tile-based games. Tilemaps use a TileSet which contain a list of tiles (textures, their rect and a collision)
and are used to create complex grid-based maps.
To optimize drawing and culling (sort of like GridMap), you can specify a quadrant size, so chunks of the map will be
batched together at drawing time.
Set the quadrant size, this optimizes drawing by batching chunks of map at draw/cull time.
Allowed values are integers ranging from 1 to 128.
• int get_quadrant_size ( ) const
Return the quadrant size.
• void set_tile_origin ( int origin )
Set the tile origin to the tile center or its top-left corner (use TILE_ORIGIN_* constants as argument).
• int get_tile_origin ( ) const
Return the tile origin configuration.
• void set_center_x ( bool enable )
Set tiles to be centered in x coordinate. (by default this is false and they are drawn from upper left cell corner).
• bool get_center_x ( ) const
Return true if tiles are to be centered in x coordinate (by default this is false and they are drawn from upper left cell
corner).
• void set_center_y ( bool enable )
Set tiles to be centered in y coordinate. (by default this is false and they are drawn from upper left cell corner).
• bool get_center_y ( ) const
Return true if tiles are to be centered in y coordinate (by default this is false and they are drawn from upper left cell
corner).
• void set_y_sort_mode ( bool enable )
Set the Y sort mode. Enabled Y sort mode means that children of the tilemap will be drawn in the order defined by
their Y coordinate.
A tile with a higher Y coordinate will therefore be drawn later, potentially covering up the tile(s) above it if its sprite
is higher than its cell size.
• bool is_y_sort_mode_enabled ( ) const
Return the Y sort mode.
• void set_collision_use_kinematic ( bool use_kinematic )
Set the tilemap to handle collisions as a kinematic body (enabled) or a static body (disabled).
• bool get_collision_use_kinematic ( ) const
Return whether the tilemap handles collisions as a kinematic body.
• void set_collision_layer ( int mask )
Set the collision layer.
Layers are referenced by binary indexes, so allowable values to describe the 20 available layers range from 0 to 2^20-1.
Return an array of all cells containing a tile from the tileset (i.e. a tile index different from -1).
• Vector2 map_to_world ( Vector2 mappos, bool ignore_half_ofs=false ) const
Return the absolute world position corresponding to the tilemap (grid-based) coordinates given as an argument.
Optionally, the tilemap’s potential half offset can be ignored.
• Vector2 world_to_map ( Vector2 worldpos ) const
Return the tilemap (grid-based) coordinates corresponding to the absolute world position given as an argument.
9.321 TileSet
9.321.3 Description
A TileSet is a library of tiles for a TileMap. It contains a list of tiles, each consisting of a sprite and optional collision
shapes.
Tiles are referenced by a unique integer ID.
9.322 Timer
9.322.3 Signals
• timeout ( )
9.322.5 Description
Timer node. This is a simple node that will emit a timeout callback when the timer runs out. It can optionally be set to
loop.
9.323 ToolButton
Inherits: Button < BaseButton < Control < CanvasItem < Node < Object
Category: Core
9.324 TouchScreenButton
9.324.3 Signals
• released ( )
• pressed ( )
9.325 Transform
3D Transformation.
Transform affine_inverse ( )
Transform inverse ( )
Transform looking_at ( Vector3 target, Vector3 up )
Transform orthonormalized ( )
Transform rotated ( Vector3 axis, float phi )
Transform scaled ( Vector3 scale )
Transform translated ( Vector3 ofs )
var xform ( var v )
var xform_inv ( var v )
Transform Transform ( Vector3 x_axis, Vector3 y_axis, Vector3 z_axis, Vector3 origin )
Transform Transform ( Matrix3 basis, Vector3 origin )
Transform Transform ( Matrix32 from )
Transform Transform ( Quat from )
Transform Transform ( Matrix3 from )
• Matrix3 basis
• Vector3 origin
9.325.4 Description
Transform is used to store transformations, including translations. It consists of a Matrix3 “basis” and Vector3 “origin”.
Transform is used to represent transformations of any object in space. It is similar to a 4x3 matrix.
• Transform affine_inverse ( )
• Transform inverse ( )
Returns the inverse of the transform.
• Transform looking_at ( Vector3 target, Vector3 up )
• Transform orthonormalized ( )
• Transform rotated ( Vector3 axis, float phi )
• Transform scaled ( Vector3 scale )
• Transform translated ( Vector3 ofs )
• var xform ( var v )
Transforms vector “v” by this transform.
• var xform_inv ( var v )
Inverse-transforms vector “v” by this transform.
• Transform Transform ( Vector3 x_axis, Vector3 y_axis, Vector3 z_axis, Vector3 origin )
• Transform Transform ( Matrix3 basis, Vector3 origin )
• Transform Transform ( Matrix32 from )
• Transform Transform ( Quat from )
• Transform Transform ( Matrix3 from )
9.326 Translation
Language Translation.
9.326.3 Description
Translations are resources that can be loaded/unloaded on demand. They map a string to another string.
9.327 TranslationServer
Inherits: Object
Category: Core
Server that manages all translations. Translations can be set to it and removed from it.
9.328 Tree
void clear ( )
TreeItem create_item ( TreeItem parent=Object() )
TreeItem get_root ( )
void set_column_min_width ( int column, int min_width )
void set_column_expand ( int column, bool expand )
int get_column_width ( int column ) const
void set_hide_root ( bool enable )
TreeItem get_next_selected ( TreeItem from )
TreeItem get_selected ( ) const
int get_selected_column ( ) const
int get_pressed_button ( ) const
void set_select_mode ( int mode )
void set_columns ( int amount )
int get_columns ( ) const
TreeItem get_edited ( ) const
int get_edited_column ( ) const
Rect2 get_custom_popup_rect ( ) const
Rect2 get_item_area_rect ( TreeItem item, int column=-1 ) const
void ensure_cursor_is_visible ( )
void set_column_titles_visible ( bool visible )
bool are_column_titles_visible ( ) const
void set_column_title ( int column, String title )
String get_column_title ( int column ) const
Vector2 get_scroll ( ) const
void set_hide_folding ( bool hide )
bool is_folding_hidden ( ) const
9.328.3 Signals
• item_activated ( )
• multi_selected ( Object item, int column, bool selected )
• custom_popup_edited ( bool arrow_clicked )
• item_collapsed ( Object item )
• item_edited ( )
• item_selected ( )
• cell_selected ( )
• button_pressed ( Object item, int column, int id )
• SELECT_SINGLE = 0
• SELECT_ROW = 1
• SELECT_MULTI = 2
• void clear ( )
• TreeItem create_item ( TreeItem parent=Object() )
• TreeItem get_root ( )
• void set_column_min_width ( int column, int min_width )
• void set_column_expand ( int column, bool expand )
• int get_column_width ( int column ) const
• void set_hide_root ( bool enable )
• TreeItem get_next_selected ( TreeItem from )
• TreeItem get_selected ( ) const
• int get_selected_column ( ) const
• int get_pressed_button ( ) const
• void set_select_mode ( int mode )
• void set_columns ( int amount )
• int get_columns ( ) const
• TreeItem get_edited ( ) const
• int get_edited_column ( ) const
• Rect2 get_custom_popup_rect ( ) const
• Rect2 get_item_area_rect ( TreeItem item, int column=-1 ) const
• void ensure_cursor_is_visible ( )
9.329 TreeItem
Inherits: Object
Category: Core
• CELL_MODE_STRING = 0
• CELL_MODE_CHECK = 1
• CELL_MODE_RANGE = 2
• CELL_MODE_ICON = 3
• CELL_MODE_CUSTOM = 4
• void move_to_top ( )
• void move_to_bottom ( )
9.330 Tween
9.330.3 Signals
• TWEEN_PROCESS_FIXED = 0
• TWEEN_PROCESS_IDLE = 1
• TRANS_LINEAR = 0
• TRANS_SINE = 1
• TRANS_QUINT = 2
• TRANS_QUART = 3
• TRANS_QUAD = 4
• TRANS_EXPO = 5
• TRANS_ELASTIC = 6
• TRANS_CUBIC = 7
• TRANS_CIRC = 8
• TRANS_BOUNCE = 9
• TRANS_BACK = 10
• EASE_IN = 0
• EASE_OUT = 1
• EASE_IN_OUT = 2
• EASE_OUT_IN = 3
• bool reset_all ( )
• bool stop ( Object object, String key )
• bool stop_all ( )
• bool resume ( Object object, String key )
• bool resume_all ( )
• bool remove ( Object object, String key )
• bool remove_all ( )
• bool seek ( float time )
• float tell ( ) const
• float get_runtime ( ) const
• bool interpolate_property ( Object object, String property, var initial_val, var final_val, float times_in_sec, int
trans_type, int ease_type, float delay=0 )
• bool interpolate_method ( Object object, String method, var initial_val, var final_val, float times_in_sec, int
trans_type, int ease_type, float delay=0 )
• bool interpolate_callback ( Object object, float times_in_sec, String callback, var arg1=NULL, var
arg2=NULL, var arg3=NULL, var arg4=NULL, var arg5=NULL )
• bool interpolate_deferred_callback ( Object object, float times_in_sec, String callback, var arg1=NULL, var
arg2=NULL, var arg3=NULL, var arg4=NULL, var arg5=NULL )
• bool follow_property ( Object object, String property, var initial_val, Object target, String target_property, float
times_in_sec, int trans_type, int ease_type, float delay=0 )
• bool follow_method ( Object object, String method, var initial_val, Object target, String target_method, float
times_in_sec, int trans_type, int ease_type, float delay=0 )
• bool targeting_property ( Object object, String property, Object initial, String initial_val, var final_val, float
times_in_sec, int trans_type, int ease_type, float delay=0 )
• bool targeting_method ( Object object, String method, Object initial, String initial_method, var final_val, float
times_in_sec, int trans_type, int ease_type, float delay=0 )
9.331 UndoRedo
Inherits: Object
Category: Core
9.332 VBoxContainer
Inherits: BoxContainer < Container < Control < CanvasItem < Node < Object
Category: Core
9.332.2 Description
9.333 VButtonArray
Inherits: ButtonArray < Control < CanvasItem < Node < Object
Category: Core
9.333.2 Description
9.334 Vector2
float angle ( )
float angle_to ( Vector2 to )
float angle_to_point ( Vector2 to )
Vector2 cubic_interpolate ( Vector2 b, Vector2 pre_a, Vector2 post_b, float t )
float distance_squared_to ( Vector2 to )
float distance_to ( Vector2 to )
float dot ( Vector2 with )
Vector2 floor ( )
Vector2 floorf ( )
float get_aspect ( )
float length ( )
float length_squared ( )
Vector2 linear_interpolate ( Vector2 b, float t )
Vector2 normalized ( )
Vector2 reflect ( Vector2 vec )
Vector2 rotated ( float phi )
Vector2 slide ( Vector2 vec )
Vector2 snapped ( Vector2 by )
Vector2 tangent ( )
Vector2 Vector2 ( float x, float y )
• float x
• float y
• float width
• float height
• float angle ( )
Returns the result of atan2 when called with the Vector’s x and y as parameters (Math::atan2(x,y)).
Be aware that it therefore returns an angle oriented clockwise with regard to the (0, 1) unit vector, and not an angle ori-
ented counter-clockwise with regard to the (1, 0) unit vector (which would be the typical trigonometric representation
of the angle when calling Math::atan2(y,x)).
• float angle_to ( Vector2 to )
Returns the angle in radians between the two vectors.
• float angle_to_point ( Vector2 to )
Returns the angle in radians between the line connecting the two points and the x coordinate.
• Vector2 cubic_interpolate ( Vector2 b, Vector2 pre_a, Vector2 post_b, float t )
Cubicly interpolates between this Vector and “b”, using “pre_a” and “post_b” as handles, and returning the result at
position “t”.
• float distance_squared_to ( Vector2 to )
Returns the squared distance to vector “b”. Prefer this function over “distance_to” if you need to sort vectors or need
the squared distance for some formula.
• float distance_to ( Vector2 to )
Returns the distance to vector “b”.
• float dot ( Vector2 with )
Returns the dot product with vector “b”.
• Vector2 floor ( )
Remove the fractional part of x and y.
• Vector2 floorf ( )
• float get_aspect ( )
Returns the ratio of X to Y.
• float length ( )
Returns the length of the vector.
• float length_squared ( )
Returns the squared length of the vector. Prefer this function over “length” if you need to sort vectors or need the
squared length for some formula.
• Vector2 linear_interpolate ( Vector2 b, float t )
Returns the result of the linear interpolation between this vector and “b”, by amount “t”.
• Vector2 normalized ( )
Returns a normalized vector to unit length.
• Vector2 reflect ( Vector2 vec )
Like “slide”, but reflects the Vector instead of continuing along the wall.
• Vector2 rotated ( float phi )
Rotates the vector by “phi” radians.
• Vector2 slide ( Vector2 vec )
Slides the vector by the other vector.
• Vector2 snapped ( Vector2 by )
Snaps the vector to a grid with the given size.
• Vector2 tangent ( )
Returns a perpendicular vector.
• Vector2 Vector2 ( float x, float y )
Constructs a new Vector2 from the given x and y.
9.335 Vector2Array
An Array of Vector2.
9.335.3 Description
9.336 Vector3
Vector3 abs ( )
Vector3 ceil ( )
Vector3 cross ( Vector3 b )
Vector3 cubic_interpolate ( Vector3 b, Vector3 pre_a, Vector3 post_b, float t )
float distance_squared_to ( Vector3 b )
float distance_to ( Vector3 b )
float dot ( Vector3 b )
Vector3 floor ( )
Vector3 inverse ( )
float length ( )
float length_squared ( )
Vector3 linear_interpolate ( Vector3 b, float t )
int max_axis ( )
int min_axis ( )
Vector3 normalized ( )
Vector3 reflect ( Vector3 by )
Vector3 rotated ( Vector3 axis, float phi )
Vector3 slide ( Vector3 by )
Vector3 snapped ( float by )
Vector3 Vector3 ( float x, float y, float z )
• float x
• float y
• float z
• AXIS_X = 0 — Enumerated value for the X axis. Returned by functions like max_axis or min_axis.
• AXIS_Y = 1 — Enumerated value for the Y axis.
• AXIS_Z = 2 — Enumerated value for the Z axis.
9.336.5 Description
Vector3 is one of the core classes of the engine, and includes several built-in helper functions to perform basic vector
math operations.
• Vector3 abs ( )
Returns a new vector with all components in absolute values (e.g. positive).
• Vector3 ceil ( )
Returns a new vector with all components rounded up.
9.337 Vector3Array
An Array of Vector3.
9.337.3 Description
9.338 VehicleBody
Inherits: PhysicsBody < CollisionObject < Spatial < Node < Object
Category: Core
9.339 VehicleWheel
9.340 VideoPlayer
9.341 VideoStream
9.342 VideoStreamTheora
9.343 Viewport
9.343.3 Signals
• size_changed ( )
• RENDER_TARGET_UPDATE_DISABLED = 0
• RENDER_TARGET_UPDATE_ONCE = 1
• RENDER_TARGET_UPDATE_WHEN_VISIBLE = 2
• RENDER_TARGET_UPDATE_ALWAYS = 3
9.343.5 Description
A Viewport creates a different view into the screen, or a sub-view inside another viewport. Children 2D Nodes will
display on it, and children Camera 3D nodes will render on it too.
Optionally, a viewport can have its own 2D or 3D world, so they don’t share what they draw with other viewports.
If a viewport is a child of a Control, it will automatically take up its same rect and position, otherwise they must be set
manually.
Viewports can also choose to be audio listeners, so they generate positional audio depending on a 2D or 3D camera
child of it.
Also, viewports can be assigned to different screens in case the devices have multiple screens.
Finally, viewports can also behave as render targets, in which case they will not be visible unless the associated texture
is used to draw.
9.344 ViewportSprite
9.345 VisibilityEnabler
• ENABLER_FREEZE_BODIES = 1
• ENABLER_PAUSE_ANIMATIONS = 0
• ENABLER_MAX = 2
9.346 VisibilityEnabler2D
Inherits: VisibilityNotifier2D < Node2D < CanvasItem < Node < Object
Category: Core
• ENABLER_FREEZE_BODIES = 1
• ENABLER_PAUSE_ANIMATIONS = 0
• ENABLER_PAUSE_PARTICLES = 2
• ENABLER_PARENT_PROCESS = 3
• ENABLER_PARENT_FIXED_PROCESS = 4
• ENABLER_MAX = 5
9.347 VisibilityNotifier
9.347.3 Signals
• enter_screen ( )
• enter_camera ( Object camera )
• exit_screen ( )
• exit_camera ( Object camera )
9.348 VisibilityNotifier2D
9.348.3 Signals
• enter_screen ( )
• enter_viewport ( Object viewport )
• exit_screen ( )
• exit_viewport ( Object viewport )
9.349 VisualInstance
9.350 VisualServer
Inherits: Object
Category: Core
RID texture_create ( )
RID texture_create_from_image ( Image arg0, int arg1=7 )
void texture_set_flags ( RID arg0, int arg1 )
int texture_get_flags ( RID arg0 ) const
int texture_get_width ( RID arg0 ) const
int texture_get_height ( RID arg0 ) const
RID shader_create ( int mode=0 )
void shader_set_mode ( RID shader, int mode )
RID material_create ( )
void material_set_shader ( RID shader, RID arg1 )
RID material_get_shader ( RID arg0 ) const
void material_set_param ( RID arg0, String arg1, var arg2 )
void material_get_param ( RID arg0, String arg1 ) const
void material_set_flag ( RID arg0, int arg1, bool arg2 )
bool material_get_flag ( RID arg0, int arg1 ) const
void material_set_blend_mode ( RID arg0, int arg1 )
int material_get_blend_mode ( RID arg0 ) const
void material_set_line_width ( RID arg0, float arg1 )
float material_get_line_width ( RID arg0 ) const
RID mesh_create ( )
void mesh_add_surface ( RID arg0, int arg1, Array arg2, Array arg3, bool arg4=-1 )
void mesh_surface_set_material ( RID arg0, int arg1, RID arg2, bool arg3=false )
RID mesh_surface_get_material ( RID arg0, int arg1 ) const
int mesh_surface_get_array_len ( RID arg0, int arg1 ) const
int mesh_surface_get_array_index_len ( RID arg0, int arg1 ) const
int mesh_surface_get_format ( RID arg0, int arg1 ) const
int mesh_surface_get_primitive_type ( RID arg0, int arg1 ) const
void mesh_remove_surface ( RID arg0, int arg1 )
int mesh_get_surface_count ( RID arg0 ) const
Continued on
• NO_INDEX_ARRAY = -1
• CUSTOM_ARRAY_SIZE = 8
• ARRAY_WEIGHTS_SIZE = 4
• MAX_PARTICLE_COLOR_PHASES = 4
• MAX_PARTICLE_ATTRACTORS = 4
• MAX_CURSORS = 8
• TEXTURE_FLAG_MIPMAPS = 1
• TEXTURE_FLAG_REPEAT = 2
• TEXTURE_FLAG_FILTER = 4
• TEXTURE_FLAG_CUBEMAP = 2048
• TEXTURE_FLAGS_DEFAULT = 7
• CUBEMAP_LEFT = 0
• CUBEMAP_RIGHT = 1
• CUBEMAP_BOTTOM = 2
• CUBEMAP_TOP = 3
• CUBEMAP_FRONT = 4
• CUBEMAP_BACK = 5
• SHADER_MATERIAL = 0
• SHADER_POST_PROCESS = 2
• MATERIAL_FLAG_VISIBLE = 0
• MATERIAL_FLAG_DOUBLE_SIDED = 1
• MATERIAL_FLAG_INVERT_FACES = 2
• MATERIAL_FLAG_UNSHADED = 3
• MATERIAL_FLAG_ONTOP = 4
• MATERIAL_FLAG_MAX = 7
• MATERIAL_BLEND_MODE_MIX = 0
• MATERIAL_BLEND_MODE_ADD = 1
• MATERIAL_BLEND_MODE_SUB = 2
• MATERIAL_BLEND_MODE_MUL = 3
• FIXED_MATERIAL_PARAM_DIFFUSE = 0
• FIXED_MATERIAL_PARAM_DETAIL = 1
• FIXED_MATERIAL_PARAM_SPECULAR = 2
• FIXED_MATERIAL_PARAM_EMISSION = 3
• FIXED_MATERIAL_PARAM_SPECULAR_EXP = 4
• FIXED_MATERIAL_PARAM_GLOW = 5
• FIXED_MATERIAL_PARAM_NORMAL = 6
• FIXED_MATERIAL_PARAM_SHADE_PARAM = 7
• FIXED_MATERIAL_PARAM_MAX = 8
• FIXED_MATERIAL_TEXCOORD_SPHERE = 3
• FIXED_MATERIAL_TEXCOORD_UV = 0
• FIXED_MATERIAL_TEXCOORD_UV_TRANSFORM = 1
• FIXED_MATERIAL_TEXCOORD_UV2 = 2
• ARRAY_VERTEX = 0
• ARRAY_NORMAL = 1
• ARRAY_TANGENT = 2
• ARRAY_COLOR = 3
• ARRAY_TEX_UV = 4
• ARRAY_BONES = 6
• ARRAY_WEIGHTS = 7
• ARRAY_INDEX = 8
• ARRAY_MAX = 9
• ARRAY_FORMAT_VERTEX = 1
• ARRAY_FORMAT_NORMAL = 2
• ARRAY_FORMAT_TANGENT = 4
• ARRAY_FORMAT_COLOR = 8
• ARRAY_FORMAT_TEX_UV = 16
• ARRAY_FORMAT_BONES = 64
• ARRAY_FORMAT_WEIGHTS = 128
• ARRAY_FORMAT_INDEX = 256
• PRIMITIVE_POINTS = 0
• PRIMITIVE_LINES = 1
• PRIMITIVE_LINE_STRIP = 2
• PRIMITIVE_LINE_LOOP = 3
• PRIMITIVE_TRIANGLES = 4
• PRIMITIVE_TRIANGLE_STRIP = 5
• PRIMITIVE_TRIANGLE_FAN = 6
• PRIMITIVE_MAX = 7
• PARTICLE_LIFETIME = 0
• PARTICLE_SPREAD = 1
• PARTICLE_GRAVITY = 2
• PARTICLE_LINEAR_VELOCITY = 3
• PARTICLE_ANGULAR_VELOCITY = 4
• PARTICLE_LINEAR_ACCELERATION = 5
• PARTICLE_RADIAL_ACCELERATION = 6
• PARTICLE_TANGENTIAL_ACCELERATION = 7
• PARTICLE_INITIAL_SIZE = 9
• PARTICLE_FINAL_SIZE = 10
• PARTICLE_INITIAL_ANGLE = 11
• PARTICLE_HEIGHT = 12
• PARTICLE_HEIGHT_SPEED_SCALE = 13
• PARTICLE_VAR_MAX = 14
• LIGHT_DIRECTIONAL = 0
• LIGHT_OMNI = 1
• LIGHT_SPOT = 2
• LIGHT_COLOR_DIFFUSE = 0
• LIGHT_COLOR_SPECULAR = 1
• LIGHT_PARAM_SPOT_ATTENUATION = 0
• LIGHT_PARAM_SPOT_ANGLE = 1
• LIGHT_PARAM_RADIUS = 2
• LIGHT_PARAM_ENERGY = 3
• LIGHT_PARAM_ATTENUATION = 4
• LIGHT_PARAM_MAX = 10
• SCENARIO_DEBUG_DISABLED = 0
• SCENARIO_DEBUG_WIREFRAME = 1
• SCENARIO_DEBUG_OVERDRAW = 2
• INSTANCE_MESH = 1
• INSTANCE_MULTIMESH = 2
• INSTANCE_PARTICLES = 4
• INSTANCE_LIGHT = 5
• INSTANCE_ROOM = 6
• INSTANCE_PORTAL = 7
• INSTANCE_GEOMETRY_MASK = 30
• INFO_OBJECTS_IN_FRAME = 0
• INFO_VERTICES_IN_FRAME = 1
• INFO_MATERIAL_CHANGES_IN_FRAME = 2
• INFO_SHADER_CHANGES_IN_FRAME = 3
• INFO_SURFACE_CHANGES_IN_FRAME = 4
• INFO_DRAW_CALLS_IN_FRAME = 5
• INFO_USAGE_VIDEO_MEM_TOTAL = 6
• INFO_VIDEO_MEM_USED = 7
• INFO_TEXTURE_MEM_USED = 8
• INFO_VERTEX_MEM_USED = 9
9.350.4 Description
Server for anything visible. The visual server is the API backend for everything visible. The whole scene system
mounts on it to display.
The visual server is completely opaque, the internals are entirely implementation specific and cannot be accessed.
• RID texture_create ( )
• RID texture_create_from_image ( Image arg0, int arg1=7 )
• void texture_set_flags ( RID arg0, int arg1 )
• int texture_get_flags ( RID arg0 ) const
• int texture_get_width ( RID arg0 ) const
• int texture_get_height ( RID arg0 ) const
• RID shader_create ( int mode=0 )
• void shader_set_mode ( RID shader, int mode )
• RID material_create ( )
• void material_set_shader ( RID shader, RID arg1 )
• RID material_get_shader ( RID arg0 ) const
• void material_set_param ( RID arg0, String arg1, var arg2 )
• void material_get_param ( RID arg0, String arg1 ) const
• void material_set_flag ( RID arg0, int arg1, bool arg2 )
• bool material_get_flag ( RID arg0, int arg1 ) const
• void material_set_blend_mode ( RID arg0, int arg1 )
• int material_get_blend_mode ( RID arg0 ) const
• void material_set_line_width ( RID arg0, float arg1 )
• float material_get_line_width ( RID arg0 ) const
• RID mesh_create ( )
• void mesh_add_surface ( RID arg0, int arg1, Array arg2, Array arg3, bool arg4=-1 )
• void mesh_surface_set_material ( RID arg0, int arg1, RID arg2, bool arg3=false )
• RID mesh_surface_get_material ( RID arg0, int arg1 ) const
• int mesh_surface_get_array_len ( RID arg0, int arg1 ) const
• int mesh_surface_get_array_index_len ( RID arg0, int arg1 ) const
• int mesh_surface_get_format ( RID arg0, int arg1 ) const
• int mesh_surface_get_primitive_type ( RID arg0, int arg1 ) const
• void mesh_remove_surface ( RID arg0, int arg1 )
• int mesh_get_surface_count ( RID arg0 ) const
• RID multimesh_create ( )
• void multimesh_set_mesh ( RID arg0, RID arg1 )
• void multimesh_set_aabb ( RID arg0, AABB arg1 )
• void multimesh_instance_set_transform ( RID arg0, int arg1, Transform arg2 )
• void multimesh_instance_set_color ( RID arg0, int arg1, Color arg2 )
• RID multimesh_get_mesh ( RID arg0 ) const
• void canvas_item_add_texture_rect_region ( RID arg0, Rect2 arg1, RID arg2, Rect2 arg3, Color
arg4=Color(1,1,1,1), bool arg5=false )
• void canvas_item_add_style_box ( RID arg0, Rect2 arg1, RID arg2, RealArray arg3, Color arg4=Color(1,1,1,1)
)
• void canvas_item_add_circle ( RID arg0, Vector2 arg1, float arg2, Color arg3 )
• void viewport_set_canvas_transform ( RID arg0, RID arg1, Matrix32 arg2 )
• void canvas_item_clear ( RID arg0 )
• void canvas_item_raise ( RID arg0 )
• void cursor_set_rotation ( float arg0, int arg1 )
• void cursor_set_texture ( RID arg0, Vector2 arg1, int arg2 )
• void cursor_set_visible ( bool arg0, int arg1 )
• void cursor_set_pos ( Vector2 arg0, int arg1 )
• void black_bars_set_margins ( int left, int top, int right, int bottom )
• void black_bars_set_images ( RID left, RID top, RID right, RID bottom )
• RID make_sphere_mesh ( int arg0, int arg1, float arg2 )
• void mesh_add_surface_from_planes ( RID arg0, Array arg1 )
• void draw ( )
• void sync ( )
• void free ( RID arg0 )
• void set_default_clear_color ( Color arg0 )
• int get_render_info ( int arg0 )
9.351 VScrollBar
Inherits: ScrollBar < Range < Control < CanvasItem < Node < Object
Category: Core
Vertical version of ScrollBar, which goes from left (min) to right (max).
9.352 VSeparator
Inherits: Separator < Control < CanvasItem < Node < Object
Category: Core
9.352.2 Description
Vertical version of Separator. It is used to separate objects horizontally, though (but it looks vertical!).
9.353 VSlider
Inherits: Slider < Range < Control < CanvasItem < Node < Object
Category: Core
Vertical slider.
9.353.2 Description
Vertical slider. See Slider. This one goes from left (min) to right (max).
9.354 VSplitContainer
Inherits: SplitContainer < Container < Control < CanvasItem < Node < Object
Category: Core
9.354.2 Description
Vertical split container. See SplitContainer. This goes from left to right.
9.355 WeakRef
9.356 WindowDialog
Inherits: Popup < Control < CanvasItem < Node < Object
Inherited By: AcceptDialog
Category: Core
9.356.3 Description
Windowdialog is the base class for all window-based dialogs. It’s a by-default toplevel Control that draws a window
decoration and allows motion and resizing.
9.357 World
9.357.3 Description
Class that has everything pertaining to a world. A physics space, a visual scenario and a sound space. Spatial nodes
register their resources into the current world.
9.358 World2D
RID get_canvas ( )
RID get_space ( )
RID get_sound_space ( )
Physics2DDirectSpaceState get_direct_space_state ( )
9.358.3 Description
Class that has everything pertaining to a 2D world. A physics space, a visual scenario and a sound space. 2D nodes
register their resources into the current 2D world.
• RID get_canvas ( )
• RID get_space ( )
• RID get_sound_space ( )
• Physics2DDirectSpaceState get_direct_space_state ( )
9.359 WorldEnvironment
9.360 XMLParser
int read ( )
int get_node_type ( )
String get_node_name ( ) const
String get_node_data ( ) const
int get_node_offset ( ) const
int get_attribute_count ( ) const
String get_attribute_name ( int idx ) const
String get_attribute_value ( int idx ) const
bool has_attribute ( String name ) const
String get_named_attribute_value ( String name ) const
String get_named_attribute_value_safe ( String name ) const
bool is_empty ( ) const
int get_current_line ( ) const
void skip_section ( )
int seek ( int pos )
int open ( String file )
int open_buffer ( RawArray buffer )
• NODE_NONE = 0
• NODE_ELEMENT = 1
• NODE_ELEMENT_END = 2
• NODE_TEXT = 3
• NODE_COMMENT = 4
• NODE_CDATA = 5
• NODE_UNKNOWN = 6
• int read ( )
• int get_node_type ( )
• String get_node_name ( ) const
• String get_node_data ( ) const
• int get_node_offset ( ) const
• int get_attribute_count ( ) const
• String get_attribute_name ( int idx ) const
• String get_attribute_value ( int idx ) const
• bool has_attribute ( String name ) const
• String get_named_attribute_value ( String name ) const
9.361 YSort
Languages
10.1 GDScript
10.1.1 Introduction
GDScript is a high level, dynamically typed programming language used to create content. It uses a syntax that is very
similar to the Python language (blocks are indent-based) and its goal is to be very optimal and tightly integrated with
the engine, allowing great flexibility for content creation and integration.
10.1.2 History
Initially, Godot was designed to support multiple scripting languages (this ability still exists today). However, only
GDScript is in use right now. There is a little history behind this.
In the early days, the engine used the Lua scripting language. Lua is fast, but creating bindings to an object oriented
system (by using fallbacks) was complex and slow and took an enormous amount of code. After some experiments
with Python, it also proved difficult to embed.
The last third party scripting language that was used for shipped games was Squirrel, but it was dropped as well. At
that point, it became evident that Godot would work more optimally by using a built-in scripting language, as the
following barriers were met:
• Godot embeds scripts in nodes, most languages are not designed with this in mind.
• Godot uses several built-in data types for 2D and 3D math, script languages do not provide this, and binding
them is inefficient.
• Godot uses threads heavily for lifting and initializing data from the net or disk, script interpreters for common
languages are not friendly to this.
• Godot already has a memory management model for resources, most script languages provide their own, which
resulted in duplicate effort and bugs.
• Binding code is always messy and results in several failure points, unexpected bugs and generally low maintain-
ability.
Finally, GDScript was written as a custom solution. The language and interpreter for it ended up being smaller than the
binding code itself for Lua and Squirrel, and equally as functional. With time, having a built-in language has proven
to be a huge advantage.
791
Godot Engine Documentation, Release latest
10.1.3 Example
Some people can learn better by just taking a look at the syntax, so here’s a simple example of how it looks.
# a file is a class!
# inheritance
extends BaseClass
# member variables
var a = 5
var s = "Hello"
var arr = [1, 2, 3]
var dict = {"key":"value", 2:3}
# constants
const answer = 42
const thename = "Charly"
var v2 = Vector2(1, 2)
var v3 = Vector3(1, 2, 3)
# function
for i in range(20):
print(i)
while(param2 != 0):
param2 -= 1
# subclass
class Something:
var a = 10
# constructor
func _init():
print("constructed!")
var lv = Something.new()
print(lv.a)
If you have previous experience with statically typed languages such as C, C++, or C# but never used a dynamically
typed one, it is advised you read this tutorial: [[GDScript (More Efficiently)]].
10.1.4 Language
Identifiers
Any string that restricts itself to alphabetic characters (a to z and A to Z), digits (0 to 9) and _ qualifies as an identifier.
Additionally, identifiers must not begin with a digit. Identifiers are case-sensitive (foo is different from FOO).
Keywords
The following is the list of keywords supported by the language. Since keywords are reserved words (tokens), they
can’t be used as identifiers.
Operators
The following is the list of supported operators and their precedence (TODO, change since this was made to reflect
python operators)
Operator Description
x[index] Subscription, Highest Priority
x.attribute Attribute Reference
extends Instance Type Checker
~ Bitwise NOT
-x Negative
*/% Multiplication / Division / Remainder
+- Addition / Subtraction
<< >> Bit Shifting
& Bitwise AND
^ Bitwise XOR
| Bitwise OR
< > == != >= <= Comparisons
in Content Test
! not Boolean NOT
and && Boolean AND
or || Boolean OR
= += -= *= /= %= &= |= Assignment, Lowest Priority
Literals
Literal Type
45 Base 10 integer
0x8F51 Base 16 (hex) integer
3.14, 58.1e-10 Floating point number (real)
"Hello", "Hi" Strings
"""Hello, Dude""" Multiline string
@"Node/Label" NodePath or StringName
Comments
Anything from a # to the end of the line is ignored and is considered a comment.
# This is a comment
Multi-line comments can be created using “”” (three quotes in a row) at the beginning and end of a block of text.
""" Everything on these
lines is considered
a comment """
null
null is a data type that contains no information, nothing assigned, and it’s just empty. It can only be set to one value:
null.
bool
int
The integer data type can only contain integer numbers, (both negative and positive).
float
String
A sequence of characters in Unicode format. Strings can contain the standard C escape sequences.
Vector2
2D vector type containing x and y fields. Can alternatively access fields as width and height for readability. Can
also be accessed as array.
Rect2
2D Rectangle type containing two vectors fields: pos and size. Alternatively contains an end field which is
pos+size.
Vector3
3D vector type containing x, y and z fields. This can also be accessed as an array.
Matrix32
Plane
3D Plane type in normalized form that contains a normal vector field and a d scalar distance.
Quat
Quaternion is a datatype used for representing a 3D rotation. It’s useful for interpolating rotations.
AABB
Axis Aligned bounding box (or 3D box) contains 2 vectors fields: pos and size. Alternatively contains an end field
which is pos+size. As an alias of this type, Rect3 can be used interchangeably.
Matrix3
3x3 matrix used for 3D rotation and scale. It contains 3 vector fields (x, y and z) and can also be accessed as an array
of 3D vectors.
Transform
Color
Color data type contains r, g, b, and a fields. It can also be accessed as h, s, and v for hue/saturation/value.
Image
Contains a custom format 2D image and allows direct access to the pixels.
NodePath
Compiled path to a node used mainly in the scene system. It can be easily assigned to, and from, a String.
RID
Object
InputEvent
Events from input devices are contained in very compact form in InputEvent objects. Due to the fact that they can be
received in high amounts from frame to frame they are optimized as their own data type.
Array
Generic sequence of objects. Its size can be changed to anything and starts from index 0.
var arr=[]
arr=[1, 2, 3]
arr[0] = "Hi!"
Arrays are allocated linearly in memory, so they are fast, but very large arrays (more than tens of thousands of elements)
may cause fragmentation.
There are specialized arrays (listed below) for some built-in data types which do not suffer from this and use less
memory, but they are atomic and generally run a little slower, so they are only justified for very large amount of data.
Dictionary
Lua-style table syntax is also supported, given that it’s easier to write and read:
var d = {
somekey = 2,
otherkey = [2,3,4],
morekey = "Hello"
}
ByteArray
IntArray
FloatArray
StringArray
Vector2Array
Vector3Array
ColorArray
10.1.6 Data
Variables
Variables can exist as class members or local to functions. They are created with the var keyword and may, optionally,
be assigned a value upon initialization.
var a # data type is null by default
var b = 5
var c = 3.8
var d = b + c # variables are always initialized in order
Constants
Constants are similar to variables, but must be constants or constant expressions and must be assigned on initialization.
const a = 5
const b = Vector2(20, 20)
const c = 10 + 20 # constant expression
const d = Vector2(20, 30).x # constant expression: 20
const e = [1, 2, 3, 4][0] # constant expression: 1
const f = sin(20) # sin() can be used in constant expressions
const g = x + 20 # invalid; this is not a constant expression!
Functions
Functions always belong to a class. The scope priority for variable look-up is: local→class member→global. self
is provided as an option for accessing class members, but is not always required (and must not be defined as the first
parameter, like in Python). For performance reasons, functions are not considered class members, so they can’t be
referenced directly. A function can return at any point. The default return value is null.
func myfunction(a, b):
print(a)
print(b)
return a + b # return is optional; without it null is returned
Statements are standard and can be assignments, function calls, control flow structures, etc (see below). ; as a
statement separator is entirely optional.
if/else/elif
Simple conditions are created by using the if/else/elif syntax. Parenthesis around statements is allowed, but not
required. Given the nature of the tab-based indentation, elif can be used instead of else:/if: to maintain a level of
indentation.
if [expression]:
statement(s)
elif [expression]:
statement(s)
else:
statement(s)
while
Simple loops are created by using while syntax. Loops can be broken using break or continued using continue:
while [expression]:
statement(s)
for
To iterate through a range, such as an array or table, a for loop is used. For loops store the index in the loop variable
on each iteration.
for i in range(3):
statement # similar to [0, 1, 2] but does not allocate an array
for i in range(1,3):
statement # similar to [1, 2] but does not allocate an array
for i in range(2,8,2):
statement # similar to [2, 4, 6] but does not allocate an array
To call a function on a base class (that was overridden in the current one), prepend . to the function name:
.basefunc()
However, remember that functions such as _init, and most notifications such as _enter_tree, _exit_tree,
_process, _fixed_process, etc. are called in all base classes automatically, so this should be only for calling
functions you write yourself.
Classes
By default, the body of a script file is an unnamed class and it can only be referenced externally as a resource or file.
Class syntax is meant to be very compact and can only contain member variables or functions. Static functions are
allowed, but not static members (this is in the spirit of thread safety since scripts can be initialized in separate threads
without the user knowing). In the same way, member variables (including arrays and dictionaries) are initialized every
time an instance is created.
func print_value_of_a():
print(a)
Inheritance
A class file can inherit from a global class, another file or a subclass inside another file. Multiple inheritance is not
allowed. The extends syntax is used. Follows is 3 methods of using extends:
# extend from some class (global)
extends SomeClass
Inheritance Testing
It’s possible to check if an instance inherits from a given class. For this the extends keyword can be used as an
operator instead:
const enemy_class = preload("enemy.gd") # cache the enemy class
# [...]
Constructor
A class can have an optional constructor; a function named _init that is called when the class is instanced.
When inheriting, parent constructors are called automatically (no need to call ._init()). If a parent constructor
takes arguments, they are passed like this:
func _init(args).(parentargs):
pass
Sub Classes
func _init():
var sc = SomeSubClass.new() #instance by calling built-in new
sc.print_value_of_a()
Classes as Objects
It may be desired at some point to load a class from a file and then instance it. Since the global scope does not exist,
classes must be loaded as a resource. Instancing is done by calling the new function in a class object:
# load the class (loaded every time the script is instanced)
var MyClass = load("myclass.gd")
# alternatively, using the preload() function preloads the class at compile time
var MyClass2 = preload("myclass.gd")
func _init():
var a = MyClass.new()
a.somefunction()
Exports
Class members can be exported. This means their value gets saved along with a scene. If class members have
initializers to constant expressions, they will be available for editing in the property editor. Exporting is done by using
the export keyword:
extends Button
One of the fundamental benefits of exporting member variables is to have them visible in the property editor. This
way artists and game designers can modify values that later influence how the program runs. For this, a special export
syntax is provided for more detail in the exported variables:
# if the exported value assigns a constant or constant expression, the type will be inferred and used
# export can take a basic data type as an argument which will be used in the editor
export(int, "Warrior", "Magician", "Thief") var character_class # (editor will set them as 0, 1 and
export(String, "Rebecca", "Mary", "Leah") var character_name
# strings as paths
# using paths in the global filesystem is also possible, but only in tool scripts (see further below)
export(String, FILE, GLOBAL, "*.png") var tool_image # string is a path to a PNG file in the global f
export(String, DIR, GLOBAL) var tool_dir # string is a path to a directory in the global filesystem
# multiline strings
export(String, MULTILINE) var text # display a large window to edit strings with multiple lines
export(float, EASE) var transition_speed # display a visual representation of the ease() function whe
It must be noted that even if the script is not being run while at the editor, the exported properties are still editable (see
below for “tool”).
Integers used as bit flags can store multiple true/false (boolean) values in one property. By using the export hint int,
FLAGS, they can be set from the editor:
export(int, FLAGS) var spell_elements = ELEMENT_WIND | ELEMENT_WATER # individually edit the bits of
Restricting the flags to a certain number of named flags is also possible. The syntax is very similar to the enumeration
syntax:
export(int, FLAGS, "Fire", "Water", "Earth", "Wind") var spell_elements = 0 # set any of the given fl
In this example, Fire has value 1, Water has value 2, Earth has value 4 and Wind corresponds to value 8. Usually,
constants should be defined accordingly (e.g. const ELEMENT_WIND = 8 and so on).
Using bit flags requires some understanding of bitwise operations. If in doubt, boolean variables should be exported
instead.
Exporting Arrays
Exporting arrays works too but there is a restriction. While regular arrays are created local to every instance, exported
arrays are shared between all instances. This means that editing them in one instance will cause them to change in all
other instances. Exported arrays can have initializers, but they must be constant expressions.
# Exported array, shared between all instances.
# Default value must be a constant expression.
var b = [a,2,3]
Static Functions
A function can be declared static. When a function is static it has no access to the instance member variables or self.
This is mainly useful to make libraries of helper functions:
static func sum2(a, b):
return a + b
Setters/Getters
It is often useful to know when an member variable changed. It may also be desired to encapsulate its access. For
this, GDScript provides a setter_/_getter helper using the setget keyword.
Just add it at the end of the variable definition line like this:
If the value of myinteger is modified externally (not from local usage in the class), the setter function will be called
beforehand. The setter must, then, decide what to do with the new value. The setter function looks like this:
func myinteger_changed(newvalue):
myinteger=newvalue
A setter and a getter can be used together too, just define both of them:
var myvar setget myvar_set,myvar_get
func myvar_set(newvalue):
myvar=newvalue
func myvar_get():
return myvar # getter must return a value
This is especially useful when exporting variables to editor in tool scripts or plugins, for validating input.
Note: As mentioned before, local access will not trigger the setter and getter. For example:
func _init():
#does not trigger setter/getter
myinteger=5
print(myinteger)
#triggers setter/getter
self.myinteger=5
print(self.myinteger)
Tool Mode
Scripts, by default, don’t run inside the editor and only the exported properties can be changed. In some cases it is
desired that they do run inside the editor (as long as they don’t execute game code or manually avoid doing so). For
this, the tool keyword exists and must be placed at the top of the file:
tool
extends Button
func _ready():
print("Hello")
Memory Management
If a class inherits from [[Class:Reference]], then instances will be freed when no longer in use. No garbage collector
exists, just simple reference counting. By default, all classes that don’t define inheritance extend Reference. If this
is not desired, then a class must inherit [[Class:Object]] manually and must call instance.free(). To avoid reference
cycles that can’t be freed, a weakref function is provided for creating weak references.
Function References
Functions can’t be referenced because they are not treated as class members. There are two alternatives to this, though.
The call function or the funcref helper.
instance.call("funcname", args) # call a function by name
Signals
It is often desired to send a notification that something happened in an instance. GDScript supports creation of built-in
Godot signals. Declaring a signal in GDScript is easy, in the body of the class, just write:
# no arguments
signal your_signal_name
# with arguments
signal your_signal_name_with_args(a,b)
These signals, just like regular signals, can be connected in the editor or from code. Just take the instance of a class
where the signal was declared and connect it to the method of another instance:
func _callback_no_args():
print("Got callback!")
func _callback_args(a,b):
print("Got callback with args! a: ",a," and b: ",b)
func _at_some_func():
instance.connect("your_signal_name",self,"callback_no_args")
instance.connect("your_signal_name_with_args",self,"callback_args")
It is also possible to bind arguments to a signal that lacks them with your custom values:
func _at_some_func():
instance.connect("your_signal_name_with_args",self,"callback_no_args",[22,"hello"])
This is very useful when a signal from many objects is connected to a single callback and the sender must be identified:
func _button_pressed(which):
print("Button was pressed: ",which.get_name())
func _ready():
for b in get_node("buttons").get_children():
b.connect("pressed",self,"_button_pressed",[b])
Coroutines
GDScript has some support for coroutines via the yield built-in function. The way it works is very simple: Calling
yield() will immediately return from the current function, with the current frozen state of the same function as
the return value. Calling resume on this resulting object will continue execution and return whatever the function
returns. Once resumed the state object becomes invalid. Here is an example:
func myfunc():
print("hello")
yield()
print("world")
func _ready():
var y = myfunc()
#function state saved in 'y'
print("my dear")
y.resume()
# 'y' resumed and is now an invalid state
Will print:
hello
my dear
world
It is also possible to pass values between yield() and resume(), for example:
func myfunc():
print("hello")
print( yield() )
return "cheers!"
func _ready():
var y = myfunc()
#function state saved in 'y'
print( y.resume("world") )
# 'y' resumed and is now an invalid state
Will print:
hello
world
cheers!
The real strength of using yield is when combined with signals. yield can accept two parameters, an object and a
signal. When the signal is activated, execution will return. Here are some examples:
#resume execution the next frame
yield( get_tree(), "idle_frame" )
10.2.1 About
This tutorial aims to be a quick reference for how to use GDScript more efficiently. It focuses in common cases specific
to the language, but also covers a lot related to using dynamically typed languages.
It’s meant to be specially useful for programmers without previous or little experience of dynamically typed languages.
GDScript is a Dynamically Typed language. As such, it’s main advantages are that:
• Language is very simple to learn.
• Most code can be written and changed quickly and without hassle.
• Less code written means less errors & mistakes to fix.
• Easier to read the code (less clutter).
• No compilation is required to test.
• Run-Time is tiny.
• [[API:Duck-Typing]] and [[API:Polymorphism]] by nature.
While the main cons are:
• Less performance than statically typed languages.
• More difficult to refactor (symbols can’t be traced)
• Some errors that would typically be detected at compile time in statically typed languages only appear while
running the code (because expression parsing is more strict).
• Less flexibility for code-completion (some variable types are only known at run-time).
This, translated to reality, means that Godot+GDScript are a combination designed to games very quickly and effi-
ciently. For games that are very computationally intensive and can’t benefit from the engine built-in tools (such as the
Vector types, Physics Engine, Math library, etc), the possibility of using C++ is present too. This allows to still create
the entire game in GDScript and add small bits of C++ in the areas that need a boost.
All variables in a dynamicaly typed language are “variant”-like. This means that their type is not fixed, and is only
modified through assignment. Example:
Static:
int a; // value uninitialized
a=5; // this is valid
a="Hi!"; // this is invalid
Dynamic:
var a # null by default
a=5 # valid, 'a' becomes an integer
a="Hi!" # valid, 'a' changed to a string
As Function Arguments:
Functions are of dynamic nature too, which means they can be called with different arguments, for example:
Static:
void print_value(int value)
{
printf("value is %i\\n",value);
}
[..]
print_value(55); // valid
print_value("Hello"); // invalid
Dynamic:
func print_value(value):
print(value)
[..]
print_value(55) # valid
print_value("Hello") # valid
In static languages such as C or C++ (and to some extent Java and C#), there is a distinction between a variable and a
pointer/reference to a variable. The later allows the object to be modified by other functions by passing a reference to
the original one.
In C# or Java, everything not a built-in type (int, float, sometimes String) is always a pointer or a reference. Refer-
ences are also garbage-collected automatically, which means they are erased when no onger used. Dynamically typed
languages tend to use this memory model too. Some Examples:
// C++
void use_class(SomeClass *instance) {
instance->use();
}
void do_something() {
Java:
@Override
public final void use_class(SomeClass instance) {
instance.use();
}
GDScript:
func use_class(instance); #does not care about class type
instance.use() # will work with any class that has a ".use()" method.
func do_something():
var instance = SomeClass.new() # created as reference
use_class(instance) # pass as reference
#will be unreferenced and deleted
In GDScript, only base types (int, float, string and the vector types) are passed by value to functions (value is copied).
Everything else (instances, arrays, dictionaries, etc) is passed as reference. Classes that inherit [[API:Reference]]
(the default if nothing is specified) will be freed when not used, but manual memory management is allowed too if
inheriting manualy from [[API:Object]].
10.2.3 Arrays
Arrays in dynamically typed languages can contain many different mixed datatypes inside and are always dynamic
(can be resized at any time). Example:
int *array = new int[4]; //create array
array[0]=10; //initialize manually
array[1]=20; //can't mix types
array[2]=40;
array[3]=60;
//can't resize
use_array(array); //passed as pointer
delete[] array; //must be freed
//or
std::vector array;
array.resize(4);
array[0]=10; //initialize manually
array[1]=20; //can't mix types
array[2]=40;
array[3]=60;
array.resize(3); //can be resized
use_array(array); //passed reference or value
//freed when stack ends
GDScript:
var array = [10, "hello", 40, 60] # simple, and can mix types
array.resize(3) # can be resized
use_array(array) # passed as reference
#freed when no longer in use
In dynamically typed languages, arrays can also double as other datatypes, such as lists:
var array = []
array.append(4)
array.append(5)
array.pop_front()
or unordered sets:
var a = 20
if a in [10,20,30]:
print("We have a Winner!")
10.2.4 Dictionaries
Dictionaries are always a very powerful in dynamically typed languages. Most programmers that come from statically
typed languages (such as C++ or C#) ignore their existence and make their life unnecessarily more difficult. This
datatype is generally not present in such languages (or only on limited form).
Dictionaries can map any value to any other value with complete disregard for the datatype used as either key or value.
Contrary to popular belief, they are very efficient because they can be implemented with hash tables. They are, in fact,
so efficient that languages such as Lua will go as far as implementing arrays as dictionaries.
Example of Dictionary:
var d = { "name":"john", "age":22 } # simple syntax
print("Name: ", d["name"], " Age: ", d["age"] )
Dictionaries are also dynamic, keys can be added or removed at any point at little cost:
d["mother"]="Rebecca" # addition
d["age"]=11 # modification
d.erase("name") #removal
In most cases, two-dimensional arrays can often be implemented more easily with dictionaries. Here’s a simple
battleship game example:
#battleship game
const SHIP=0
const SHIP_HIT=1
const WATER_HIT=2
var board={}
func initialize():
board[Vector(1,1)]=SHIP
board[Vector(1,2)]=SHIP
board[Vector(1,3)]=SHIP
func missile(pos):
func game():
initialize()
missile( Vector2(1,1) )
missile( Vector2(5,8) )
missile( Vector2(2,3) )
Dictionaries can also be used as data markup or quick structures. While GDScript dictionaries resemble python
dictionaries, it also supports Lua style syntax an indexing, which makes it very useful for writing initial states and
quick structs:
# same example, lua-style support
# this syntax is a lot more readable and usable
var d = {
name="john",
age=22
}
print("Name: ", d.name, " Age: ", d.age ) # used "." based indexing
# indexing
d.nother="rebecca" #this doesn't work (use syntax below to add a key:value pair)
d["mother"]="rebecca" #this works
d.name="caroline" # if key exists, assignment does work, this is why it's like a quick struct.
[..]
for(int i=0;i<50;i++)
{
printf("value: %s\\n",i,strings[i]);
}
//Even in STL:
Container datatypes (arrays and dictionaries) are iterable. Dictionaries allow iterating the keys:
for key in dict:
print(key," -> ",dict[key])
Some examples:
for(int i=0;i<10;i++) {}
for(int i=5;i<10;i++) {}
for(int i=5;i<10;i+=2) {}
Translate to:
for i in range(10):
for i in range(5,10):
for i in range(5,10,2):
becomes
for i in range(10,0,-1):
10.2.6 While
var i=0
while(i
One of the most difficult concepts to grasp when moving from a statically typed language to a dynamic one is Duck
Typing. Duck typing makes overall code design much simpler and straightforward to write, but it’s not obvious how it
works.
As an example, imagine a situation where a big rock is falling down a tunnel, smashing everything on it’s way. The
code for the rock, in a statically typed language would be something like:
void BigRollingRock::on_object_hit(Smashable *entity)
{
entity->smash();
}
This, way, everything that can be smashed by a rock would have to inherit Smashable. If a character, enemy, piece
of furniture, small rock were all smashable, they would need to inherit from the class Smashable, possibly requiring
multiple inheritance. If multiple inheritance was undesired, then they would have to inherit a common class like Entity.
Yet, it would not be very elegant to add a virtual method “smash()” to Entity only if a few of them can be smashed.
With dynamically typed languages, this is not a problem. Duck typing makes sure you only have to define a smash()
function where required and that’s it. No need to consider inheritance, base classes, etc.
func _on_object_hit(object):
object.smash()
And that’s it. If the object that hit the big rock has a smash() method, it will be called. No need for inheritance or
polymorphysm. Dynamically typed languages only care about the instance having the desired method or member, not
what it inherits or the class type. The definition of Duck Typing should make this clearer:
“When I see a bird that walks like a duck and swims like a duck and quacks like a duck, I call that bird a duck”
In this case, it translates to:
“If the object can be smashed, don’t care what it is, just smash it.”
Yes, we should call it Hulk typing instead. Anyway though, there exists the possibility of the object being hit not
having a smash() function. Some dynamically typed languages simply ignore a method call when it doesn’t exist (like
Objective C), but GDScript is more strict, so checking if the function exists is desirable:
func _on_object_hit(object):
if (object.has_method("smash")):
object.smash()
Then, simply define that method and anything the rock touches can be smashed.
10.3.1 Introduction
Godot uses a simplified shader language (almost a subset of GLSL). Shaders can be used for:
• Materials
• Post-Processing
• 2D
and are divided in Vertex, Fragment and Light sections.
10.3.2 Language
Typing
The language is statically type and supports only a few operations. Arrays, classes, structures, etc are not supported.
Several built-in datatypes are provided:
Data Types
DataType Description
void Void
bool boolean (true or false)
float floating point
vec2 2-component vector, float subindices (x,y or r,g )
vec3 3-component vector, float subindices (x,y,z or r,g,b )
vec4, color 4-component vector, float subindices (x,y,z,w or r,g,b,a )
mat2 2x2 matrix, vec3 subindices (x,y)
mat3 3x3 matrix, vec3 subindices (x,y,z)
mat4 4x4 matrix, vec4 subindices (x,y,z,w)
texture texture sampler, can only be used as uniform
cubemap cubemap sampler, can only be used as uniform
Syntax
The syntax is similar to C, with statements ending in ; , and comments as // and /* */.
Example:
float a = 3;
vec3 b;
b.x = a;
Swizzling
Constructors
Constructors take the regular amount of elements, but can also accept less if the element has more subindices, for
example:
vec3 a = vec3( 1, vec2(2,3) );
vec3 b = vec3( a );
vec3 c = vec3( vec2(2,3), 1 );
vec4 d = vec4( a, 5 );
mat3 m = mat3( a,b,c );
Conditionals
Uniforms
A variable can be declared as uniform. In this case, it’s value will come from outside the shader (it will be the
responsibility of the material or whatever using the shader to provide it).
uniform vec3 direction;
uniform color tint;
Functions
Simple support for functions is provided. Functions can’t access uniforms or other shader variables.
vec3 addtwo( vec3 a, vec3 b) {
return a+b;
}
vec3 c = addtwo(vec3(1,1,1)+vec3(2,2,2));
|. Function |. Description |
| float sin( float ) | Sine |
| float cos( float ) | Cosine |
| float tan( float ) | Tangent |
| float asin( float ) | arc-Sine |
| float acos( float ) | arc-Cosine |
| float atan( float ) | arc-Tangent |
Depending on the shader type, several built-in variables are available, listed as follows:
Material - VertexShader
|. Variable |. Description |
| const vec3 SRC_VERTEX | Model-Space Vertex |
| const vec3 SRC_NORMAL | Model-Space Normal |
| const vec3 SRC_TANGENT | Model-Space Tangent |
| const float SRC_BINORMALF | Direction to Compute Binormal |
| vec3 VERTEX | View-Space Vertex |
| vec3 NORMAL | View-Space Normal |
| vec3 TANGENT | View-Space Tangent |
| vec3 BINORMAL | View-Space Binormal |
| vec2 UV | UV |
| vec2 UV2 | UV2 |
Material - FragmentShader
Variable Description
const vec3 VERTEX View-Space vertex
const vec4 POSITION View-Space Position
const vec3 NORMAL View-Space Normal
const vec3 TANGENT View-Space Tangent
const vec3 BINORMAL View-Space Binormal
const vec3 NORMALMAP Alternative to NORMAL, use for normal texture output.
const vec3 Complementary to the above, allows changing depth of normalmap.
NORMALMAP_DEPTH
const vec2 UV UV
const vec2 UV2 UV2
const color COLOR Vertex Color
const vec4 VAR1 Varying 1
const vec4 VAR2 Varying 2
const vec2 SCREEN_UV Screen Texture Coordinate (for using with texscreen)
const float TIME Time (in seconds)
const vec2 POINT_COORD UV for point, when drawing point sprites.
out vec3 DIFFUSE Diffuse Color
out vec4 DIFFUSE_ALPHA Diffuse Color with Alpha (using this sends geometry to alpha pipeline)
out vec3 SPECULAR Specular Color
out vec3 EMISSION Emission Color
out float SPEC_EXP Specular Exponent (Fragment Version)
out float GLOW Glow
out mat4 Inverse camera matrix, can be used to obtain world coords (see example
INV_CAMERA_MATRIX below).
Material - LightShader
Variable Description
const vec3 NORMAL View-Space normal
const vec3 LIGHT_DIR View-Space Light Direction
const vec3 EYE_VEC View-Space Eye-Point Vector
const vec3 DIFFUSE Material Diffuse Color
const vec3 LIGHT_DIFFUSE Light Diffuse Color
const vec3 SPECULAR Material Specular Color
const vec3 LIGHT_SPECULAR Light Specular Color
const float SPECULAR_EXP Specular Exponent
const vec1 SHADE_PARAM Generic Shade Parameter
const vec2 POINT_COORD Current UV for Point Sprite
out vec2 LIGHT Resulting Light
const float TIME Time (in seconds)
|. Variable |. Description |
| const vec2 SRC_VERTEX | CanvasItem space vertex. |
| vec2 UV | UV |
| out vec2 VERTEX | Output LocalSpace vertex. |
| out vec2 WORLD_VERTEX | Output WorldSpace vertex. (use this or the one above) |
| color COLOR | Vertex Color |
| out vec4 VAR1 | Varying 1 Output |
| out vec4 VAR2 | Varying 2 Output |
| out float POINT_SIZE | Point Size (for points) |
| const mat4 WORLD_MATRIX | Object World Matrix |
| const mat4 EXTRA_MATRIX | Extra (user supplied) matrix via CanvasItem.draw_set_transform(). Identity by
default. |
| const mat4 PROJECTION_MATRIX | Projection Matrix (model coords to screen).|
| const float TIME | Time (in seconds) |
|. Variable |. Description |
| const vec4 SRC_COLOR | Vertex color |
| const vec4 POSITION | Screen Position |
| vec2 UV | UV |
| out color COLOR | Output Color |
| out vec3 NORMAL | Optional Normal (used for 2D Lighting) |
| out vec3 NORMALMAP | Optional Normal in standard normalmap format (flipped y and Z from 0 to 1) |
| out float NORMALMAP_DEPTH | Depth option for above normalmap output, default value is 1.0 |
| const texture TEXTURE | Current texture in use for CanvasItem |
| const vec2 TEXTURE_PIXEL_SIZE | Pixel size for current 2D texture |
| in vec4 VAR1 | Varying 1 Output |
| in vec4 VAR2 | Varying 2 Output |
| const vec2 SCREEN_UV| Screen Texture Coordinate (for using with texscreen) |
|. Variable |. Description |
| const vec4 POSITION | Screen Position |
| in vec3 NORMAL | Input Normal |
| in vec2 UV | UV |
| in color COLOR | Input Color |
| const texture TEXTURE | Current texture in use for CanvasItem |
| const vec2 TEXTURE_PIXEL_SIZE | Pixel size for current 2D texture |
| in vec4 VAR1 | Varying 1 Output |
| in vec4 VAR2 | Varying 2 Output |
| const vec2 SCREEN_UV| Screen Texture Coordinate (for using with texscreen) |
| const vec2 POINT_COORD | Current UV for Point Sprite |
| const float TIME| Time (in seconds) |
| vec2 LIGHT_VEC | Vector from light to fragment, can be modified to alter shadow computation. |
| const float LIGHT_HEIGHT | Height of Light |
| const color LIGHT_COLOR | Color of Light |
| out vec4 LIGHT | Light Ouput (shader is ignored if this is not used) |
10.3.5 Examples
Material that reads a texture, a color and multiples them, fragment program:
uniform color modulate;
uniform texture source;
10.3.6 Notes
* Do not use DIFFUSE_ALPHA unless you really intend to use transparency. Transparent materials must be sorted
by depth and slow down the rendering pipeline. For opaque materials, just use DIFFUSE.
* Do not use DISCARD unless you really need it. Discard makes rendering slower, specially on mobile devices.
* TIME may reset after a while (may last an hour or so), it’s meant for effects that vary over time.
* In general, every built-in variable not used results in less shader code generated, so writing a single giant shader
with a lot of code and optional scenarios is often not a good idea.
10.4 Locales
This is the list of supported locales and variants in the engine. It’s based on the Unix standard locale strings:
10.5.1 Introduction
[[RichTextLabel]] allows to display complex text markup in a control. It has a built-in API for generating the markup,
but can also parse a BBCode.
10.5.2 Setting Up
For RichTextLabel to work properly, it must be set-up. This means loading the intended fonts in the releavant proper-
ties:
10.5.3 Reference
CHAPTER 11
Cheat sheets
11.1.1 2D Viewport
11.1.2 3D Viewport
11.2.1 Object
11.2.2 Reference
attachment:sources.zip
Compiling
12.1.1 Scons
Godot uses Scons to build. We love it, we are not changing it for anything else. We are not even sure other build
systems are up to the task of building Godot. We constantly get requests to move the build system to CMake, or Visual
Studio, but this is not going to happen. There are many reasons why we have chosen SCons over other alternatives
and are listed as follows:
• Godot can be compiled for a dozen different platforms. All PC platforms, all mobile platforms, many consoles,
and many web-based platforms (such as HTML5 and Chrome PNACL).
• Developers often need to compile for several of the platforms at the same time, or even different targets of the
same platform. They can’t afford reconfiguring and rebuilding the project each time. SCons can do this with no
sweat, without breaking the builds.
• SCons will never break a build no matter how many changes, configurations, additions, removals etc. You have
more chances to die struck by lightning than needing to clean and rebuild in SCons.
• Godot build process is not simple. Several files are generated by code (binders), others are parsed (shaders), and
others need to offer customization (plugins). This requires complex logic which is easier to write in an actual
programming language (like Python) rather than using a mostly macro-based language only meant for building.
• Godot build process makes heavy use of cross compiling tools. Each platform has a specific detection process,
and all these must be handled as specific cases with special code written for each.
So, please get at least a little familiar with it if you are planning to build Godot yourself.
Godot’s build system will begin by detecting the platforms it can build for. If not detected, the platform will simply
not appear on the list of available platforms. The build requirements for each platform are described in the rest of this
tutorial section.
Scons is invoked by just calling scons.
However, this will do nothing except list the available platforms, for example:
user@host:~/godot$ scons
scons: Reading SConscript files ...
No valid target platform selected.
The following were detected:
831
Godot Engine Documentation, Release latest
android
server
javascript
windows
x11
To build a platform (for example, x11), run with the platform= (or just p= to make it short) argument:
user@host:~/godot$ scons platform=x11
This will start the build process, which will take a while. If you want scons to build faster, use the -j parameter to
specify how many cores will be used for the build. Or just leave it using one core, so you can use your computer for
something else :)
Example for using 4 processes:
user@host:~/godot$ scons platform=x11 -j 4
The resulting binaries will be placed in the bin/ subdirectory, generally with this naming convention:
godot..[opt].[tools/debug]..
For the previous build attempt the result would look like this:
user@host:~/godot$ ls bin
bin/godot.x11.tools.64
This means that the binary is for X11, is not optimized, has tools (the whole editor) compiled in, and is meant for 64
bits.
A Windows binary with the same configuration will look like this.
C:\\GODOT> DIR BIN/
godot.windows.tools.64.exe
Just copy that binary to wherever you like, as it self-contains the project manager, editor and all means to execute the
game. However, it lacks the data to export it to the different platforms. For that the export templates are needed (which
can be either downloaded from http://www.godotengine.org, or you can build them yourself).
Aside from that, there are a few standard options that can be set in all build targets, and will be explained as follows.
12.1.4 Tools
Tools are enabled by default in al PC targets (Linux, Windows, OSX), disabled for everything else. Disabling tools
produces a binary that can run projects but that does not include the editor or the project manager.
scons platform= tools=yes/no
12.1.5 Target
This flag appends ”.debug” suffix (for debug), or ”.tools” (for debug with tools enables). When optimization is enabled
(release) it appends the ”.opt” suffix.
12.1.6 Bits
Bits is meant to control the CPU or OS version intended to run the binaries. It works mostly on desktop platforms and
ignored everywhere else.
• 32: Build binaries for 32 bits platform.
• 64: Build binaries for 64 bits platform.
• default: Built whatever the build system feels is best. On Linux this depends on the host platform (if not cross
compiling), while on Windows and Mac it defaults to produce 32 bits binaries unless 64 bits is specified.
scons platform= bits=default/32/64
This flag appends ”.32” or ”.64” suffixes to resulting binaries when relevant.
Official export templates are downloaded from the Godot Engine site: http://www.godotengine.org. However, you
might want to build them yourself (in case you want newer ones, you are using custom modules, or simply don’t trust
your own shadow).
If you download the official export templates package and unzip it, you will notice that most are just optimized binaries
or packages for each platform:
android_debug.apk
android_release.apk
javascript_debug.zip
javascript_release.zip
linux_server_32
linux_server_64
linux_x11_32_debug
linux_x11_32_release
linux_x11_64_debug
linux_x11_64_release
osx.zip
version.txt
windows_32_debug.exe
windows_32_release.exe
windows_64_debug.exe
windows_64_release.exe
windows_debug.exe
windows_release.exe
To create those yourself, just follow the instructions detailed for each platform in this same tutorial section. Each
platform explains how to create it’s own template.
If you are working for multiple platforms, OSX is definitely the best host platform for cross compilation, since you
can cross-compile for almost every target (except for winrt). Linux and Windows come in second place, but Linux has
the advantage of being the easier platform to set this up.
12.2.1 Requirements
Python adds the interpreter (python.exe) to the path. It usually installs in C:\\Python (or C:\\Python[Version]). SCons
installs inside the python install and provides a .bat file called “scons.bat”. The location of this file can be added to the
path or it can simply be copied to C:\\Python together with the interpreter executable.
12.2.3 Compiling
Start a Visual Studio command prompt (it sets up environment variables needed by SCons to locate the compiler and
SDK), go to the root dir of the engine source code and type:
C:\\godot> scons platform=windows
If all goes well, the resulting binary executable will be placed in C:\\godot\\bin\\godot.windows.tools.exe. This ex-
ecutable file contains the whole engine and runs without any dependencies. Executing it will bring up the project
manager.
For most projects, using only scripting is enough but when development in C++ is needed, for creating modules or
extending the engine, working with an IDE is usually desirable. The visual studio command prompt calls a .bat file that
sets up environment variables (vcvarsall.bat). To build the whole engine from a single command outside the command
prompt, the following should be called in a .bat file:
C:\\path_to_sdk\\vcvarsall.bat && scons bin/godot.windows.tools.exe
NOTE: It seems the latest Visual Studio does not include a desktop command prompt (No, Native tools for x86 is not
it). The only way to build it seems to be by running:
"C:\\Program Files (x86)\\Microsoft Visual Studio 12.0\\VC\\vcvarsall.bat" && c:\\python27\\scons p=w
If you are a Linux or Mac user, you need to install mingw32 and mingw-w64. Under Ubuntu or Debian, just run the
following commands:
apt-get install mingw32 mingw-w64
If you are using other distro, scons will check for the following binaries:
i586-mingw32msvc-gcc
i686-w64-mingw32-gcc
If the binaries are named or located somewhere else, export the following env variables:
export MINGW32_PREFIX="/path/to/i586-mingw32msvc-"
export MINGW64_PREFIX="/path/to/i686-w64-mingw32-"
To make sure you are doing things correctly, executing the following in the shell should result in a working compiler:
user@host:~$ ${MINGW32_PREFIX}gcc
gcc: fatal error: no input files
Windows export templates are created by compiling Godot as release, with the following flags:
(for 32 bits, using Mingw32 command prompt or Visual Studio command prompt)
C:\\godot> scons platform=windows tools=no target=release bits=32
C:\\godot> scons platform=windows tools=no target=release_debug bits=32
However, if you are writing your custom modules or custom C++ code, you might instead want to configure your
binaries as custom export templates here:
You don’t even need to copy them, you can just reference the resulting files in the bin\\ directory of your Godot source
folder, so the next time you build you automatically have the custom templates referenced.
12.3.1 Requirements
For compiling under Linux or other Unix variants, the following is required:
• GCC or LLVM
• Python 2.7+ (3.0 is untested as of now).
• SCons build system.
• X11 and MESA development Libraries
• Xinerama Libraries
• ALSA development libraries
• Freetype (for the editor)
• OpenSSL (for HTTPS and TLS)
If you wish to have Joypad support, libevdev-dev and libudev-dev are required.
apt-get install libevdev-dev libudev-dev
12.3.2 Compiling
Start a terminal, go to the root dir of the engine source code and type:
user@host:~/godot$ scons platform=x11
If all goes well, the resulting binary executable will be placed in the “bin” subdirectory. This executable file contains
the whole engine and runs without any dependencies. Executing it will bring up the project manager.
To build Linux export templates, run the build system with the following parameters:
(32 bits)
user@host:~/godot$ scons platform=x11 tools=no target=release bits=32
user@host:~/godot$ scons platform=x11 tools=no target=release_debug bits=32
(64 bits)
user@host:~/godot$ scons platform=x11 tools=no target=release bits=64
user@host:~/godot$ scons platform=x11 tools=no target=release_debug bits=64
Note that cross compiling for the opposite bits (64/32) as your host platform in linux is quite difficult and might need
a chroot environment.
In Ubuntu, compilation works without a chroot but some libraries (.so) might be missing from /usr/lib32. Symlinking
the missing .so files from /usr/lib results in a working build.
To create standard export templates, the resulting files must be copied to:
/home/youruser/.godot/templates
However, if you are writing your custom modules or custom C++ code, you might instead want to configure your
binaries as custom export templates here:
You don’t even need to copy them, you can just reference the resulting files in the bin/ directory of your Godot source
folder, so the next time you build you automatically have the custom templates referenced.
12.4.1 Requirements
For compiling under Linux or other Unix variants, the following is required:
• Python 2.7+ (3.0 is untested as of now).
• SCons build system.
• XCode
12.4.2 Compiling
Start a terminal, go to the root dir of the engine source code and type:
user@host:~/godot$ scons platform=osx
If all goes well, the resulting binary executable will be placed in the “bin” subdirectory. This executable file contains
the whole engine and runs without any dependencies. Executing it will bring up the project manager. There is a .app
template to put the binary into in tools/Godot.app.
12.4.3 Cross-compiling
It is possible to compile for OS X in a Linux environment (and maybe also in Windows with Cygwin). For that you
will need OSXCross for being able to use OS X as target. First, follow the instructions to install it:
# Clone the OSXCross repository (https://github.com/tpoechtrager/osxcross) somewhere in your machine (or down-
load a Zip file and extract it somewhere). E.g.
~$ git clone https://github.com/tpoechtrager/osxcross.git /home/myuser/sources/osxcross
Now you can compile with SCons like you normally would:
~/godot$ scons platform=osx
12.5.1 Note
For most cases, using the built-in deployer and export templates is good enough. Compiling the Android APK manu-
ally is mostly useful for custom builds or custom packages for the deployer.
Also, you still need to do all the steps mentioned in the [[Exporting for Android]] tutorial before attempting your
custom export template.
12.5.2 Requirements
To set those environment variables on Windows, press Windows+R, type “control system”, then click on Advanced
system settings in the left pane, then click on Environment variables on the window that appears.
To set those environment variables on Linux, use export ANDROID_HOME=/path/to/android-sdk and
export ANDROID_NDK_ROOT=/path/to/android-ndk.
12.5.4 Compiling
This should result in a regular .so in \\bin folder as if it was compiled with flags: tools=no target=debug.
The resulting file will be huge because it will contain all debug symbols, so for next builds, using
target=release_debug or target=release is recommended.
Copy the .so to the libs/armeabi Android folder (or symlink if you are in Linux or OSX). Note: Git does not support
empty directories so you will have to create it if it does not exist:
C:\\godot> mkdir platform/android/java/libs
C:\\godot> mkdir platform/android/java/libs/armeabi
Remember that only one of libgodot_android.so must exist for each platform, for each build type (release, debug, etc),
it must be replaced.
Note: The file inside libs/armeabi must be renamed to “libgodot_android.so”, or else unsatisfied link error will
happen at runtime.
If you also want to include support for x86 Android, add the following compile flag: x86=yes , then copy/symlink
the resulting folder to the x86 folder:
C:\\godot> cp bin/libgodot.android..x86.so platform/android/java/libs/x86/libgodot_android.so
This will create a fat binary that works in both platforms, but will add about 6 megabytes to the APK.
12.5.5 Toolchain
We usually try to keep the Godot Android build code up to date, but Google changes their toolchain versions very often,
so if compilation fails due to wrong toolchain version, go to your NDK directory and check the current numbers, then
set the following environment variables:
NDK_TOOLCHAIN (by default set to "arm-eabi-4.4.0")
NDK_TARGET (by default set to "arm-linux-androideabi-4.8")
To compile the APK, go to the Java folder and run ant, either for debug or release build:
In the java/bin subfolder, the resulting apk can be used as export template.
Note: If you reaaaally feel oldschool, you can copy your entire game (or symlink) to the assets/ folder of the Java
project (make sure engine.cfg is in assets/) and it will work, but you lose all the benefits of the export system (scripts
are not byte-compiled, textures not converted to Android compression, etc. so it’s not a good idea).
Godot needs the freshly compiled APK as export templates. It opens the APK, changes a few things inside, adds your
file and spits it back. It’s really handy! (and required some reverse engineering of the format).
Compiling the standard export templates is done by calling scons with the following arguments:
(debug)
C:\\godot> scons platform=android target=release_debug
C:\\godot> cp bin/libgodot_android.opt.debug.so platform/android/java/libs/armeabi
C:\\godot> cd platform/android/java
C:\\godot\\platform\\android\\java> ant release
(release)
C:\\godot> scons platform=android target=release
C:\\godot> cp bin/libgodot_android.opt.so platform/android/java/libs/armeabi
C:\\godot> cd platform/android/java
C:\\godot\\platform\\android\\java> ant release
(same as before)
They must be copied to your templates folder with the following names:
android_debug.apk
android_release.apk
However, if you are writing your custom modules or custom C++ code, you might instead want to configure your
APKs as custom export templates here:
You don’t even need to copy them, you can just reference the resulting file in the bin\\ directory of your Godot source
folder, so the next time you build you automatically have the custom templates referenced.
12.5.8 Troubleshooting
Android might complain the application is not correctly installed. If so, check the following:
• Check that the debug keystore is properly generated.
• Check that jarsigner is from JDK6.
If it still fails, open a command line and run logcat:
C:\\android-sdk\\platform-tools> adb logcat
And check the output while the application is installed. Reason for failure should be presented there.
Seek assistance if you can’t figure it out.
If the application runs but exits immediately, there might be one of the following reasons:
• libgodot_android.so is not in libs/armeabi
• Device does not support armv7 (try compiling yourself for armv6)
• Device is Intel, and apk is compiled for ARM.
In any case, adb logcat should also show the cause of the error.
12.6.1 Requirements
• SCons (you can get it from macports, you should be able to run scons in a terminal when installed)
• Xcode with the iOS SDK and the command line tools.
12.6.2 Compiling
Open a Terminal, go to the root dir of the engine source code and type:
$ scons p=iphone bin/godot.iphone.debug
for a release build (check platform/iphone/detect.py for the compiler flags used for each configuration).
Alternatively, you can run
$ scons p=isim bin/godot.isim.tools
12.6.3 Run
The procedure for this is somewhat complex and requires a lot of steps, but once you have the environment properly
configured it will be easy to compile Godot for iOS anytime you want.
12.7.1 Disclaimer
While it is possible to compile for iOS on a Linux environment, Apple is very restrictive about the tools to be used
(specially hardware-wise), allowing pretty much only their products to be used for development. So this is not official.
However, a statement from Apple in 2010 says they relaxed some of the App Store review guidelines to allow any
tool to be used, as long as the resulting binary do not download any code, which means it should be OK to use the
procedure described here and cross-compiling the binary.
12.7.2 Requirements
darling-dmg
# Build it:
$ cd darling-dmg
$ mkdir build
$ cd build
$ cmake .. -DCMAKE_BUILD_TYPE=RELEASE
$ make -j 4 # The number is the amount of cores your processor have, for faster build
$ cd ../..
$ mkdir -p iPhoneSDK/iPhoneOS9.1.sdk
$ cp -r xcode/Xcode.app/Contents/Developer/Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS.sdk/*
$ cp -r xcode/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/include/c++/* iPho
$ fusermount -u xcode # unmount the image
Toolchain
# Build cctools:
$ git clone https://github.com/tpoechtrager/cctools-port.git
$ cd cctools-port/usage_examples/ios_toolchain
$ ./build.sh /path/iPhoneOS9.1.sdk.tar.xz arm64
# Copy the tools to a nicer place. Note that the SCons scripts for building will look under usr/bin inside the directory
you provide for the toolchain binaries, so you must copy to such subdirectory, akin to the following commands:
$ mkdir -p /home/user/iostoolchain/usr
$ cp -r target/bin /home/user/iostoolchain/usr/
Once you’ve done the above steps, you should keep two things in your environment: the built toolchain and the
iPhoneOS SDK directory. Those can stay anywhere you want since you have to provide their paths to the SCons build
command.
# For the iPhone platform to be detected, you need the OSXCROSS_IOS environment variable defined to anything.
$ export OSXCROSS_IOS=anything
# Now you can compile for iPhone using SCons like the standard Godot way, with some additional arguments to
provide the correct paths:
$ scons -j 4 platform=iphone bits=32 target=release_debug IPHONESDK="/path/to/iPhoneSDK" IPHONEPATH="
$ scons -j 4 platform=iphone bits=64 target=release_debug IPHONESDK="/path/to/iPhoneSDK" IPHONEPATH="
Apple requires a fat binary with both achitectures (armv7 and arm64) in a single file. To do this, use the
arm-apple-darwin11-lipo executable. The following example assumes you are in the root Godot source
directory:
$ /path/to/iostoolchain/usr/bin/arm-apple-darwin11-lipo -create bin/godot.iphone.opt.debug.32 bin/god
This page documents the current state of the “winrt” platform, used to support “Windows Store Apps” for Windows
8.1, and Windows Phone 8.1 apps using Microsoft’s new “Universal” APIs.
12.8.1 Requirements
• Windows 8
• SCons (see [[Compiling for Windows]] for more details)
• Visual Studio 2013 for Windows (but not “for Windows Desktop”). Tested on “Microsoft Visual Studio Express
2013 for Windows Version 12.0.31101.00 Update 4”.
12.8.2 Compiling
The platform can compile binaries for both Windows 8.1 and Windows Phone 8.1. The architecture is decided by the
environment variable “PLATFORM”.
Windows 8.1
• You should get an executable file inside bin/ named according to your build options, for the architecture “x64”,
for example “godot.winrt.tools.x64.exe”.
• You should get an executable file inside bin/ named according to your build options, for the architecture “arm”,
for example “godot.winrt.tools.arm.exe”.
12.8.3 Running
On Visual studio, create a new project using any of the “Unversal App” templates found under Visual C++ -> Store
Apps -> Universal Apps. “Blank App” should be fine.
On the “Solution Explorer” box, you should have 3 sections, “App.Windows (Windows 8.1)”, “App.WindowsPhone
(Windows Phone 8.1)” and “App.Shared”. You need to add files to each section:
App.Shared
Add a folder named “game” containing your game content (can be individual files or your data.pck). Remember to set
the “Content” property of each file to “True”, otherwise your files won’t get included in the package.
App.Windows
• Add your windows executable, and all the .dll files found on platform/winrt/x64/bin on the godot source. Re-
member to also set the “Content” property.
• Find the file “Package.appxmanifest”. Right click on it and select “Open with...” then “XML (Text) Editor”
from the list.
• Find the “Application” section, and add (or modify) the “Executable” property with the name of your .exe.
Example:
<Application Id=”App” Executable=”godot.winrt.tools.x64.exe” EntryPoint=”App_Windows.App”>
App.WindowsPhone
Repeat all the steps from App.Windows, using your arm executable and the dlls found in platform/winrt/arm/bin.
Remember to set the “Content” property for all the files.
Use the green “Play” button on the top to run. The drop down menu next to it should let you choose the project
(App.Windows or App.WindowsPhone) and the device (“Local Machine”, “Device” for an attached phone, etc).
12.8.4 Angle
ANGLE precompiled binaries are provided on platform/winrt/x64 and platform/winrt/arm. They are built from
MSOpenTech’s “future-dev” branch, found here: https://github.com/MSOpenTech/angle. The visual studio ‘solu-
tions’ used are found on “projects/winrt/windows/angle.sln” and “projects/winrt/windowsphone/angle.sln”.
• Audio
• Semaphores
• Keyboard input
• Proper handling of screen rotation
• Proper handling of other events such as focus lost, back button, etc.
• Packaging and deploying to devices from the editor.
• Adding Angle to our tree and compiling it from there. The same source could also be used to build for Windows
(and use Angle instead of native GL, which will be more compatible with graphics hardware)
12.8.6 Packages
• There are a bunch of tools on “powershell” that deal with packages that might be relevant:
http://technet.microsoft.com/library/dn448373.aspx
• When running a Windows 8.1 app on “Local Machine” from Visual studio, the app seems to run from an
uncompressed directory on the filesystem in an arbitrary location (ie. outside of the proper directory where
Apps are installed), but there is some special registry entry made for it, so we know it’s possible to skip the
packaging step to run locally (in the case of very big games this can be useful).
The following is almost the same exact script we use to build all the export templates that go to the site. If you want
to build or roll them yourself, this might be of use.
(note: mac stuff is missing)
#This script is intended to run on Linux or OSX. Cygwin might work.
#Build templates
# Android
**IMPORTANT REPLACE THIS BY ACTUAL VALUES**
export ANDROID_HOME=/home/to/android-sdk
export ANDROID_NDK_ROOT=/home/to/android-ndk
#Android Release
#Android Debug
# EMScripten
#. /path/to/bbndk/bbndk-env.sh
#scons -j 4 platform/bb10/godot_bb10_opt.qnx.armle target=release
#cp platform/bb10/godot_bb10_opt.qnx.armle platform/bb10/bar
# BUILD ON MAC
[...]
mkdir -p release
cp doc/base/classes.xml .
release/linux_server.64 -doctool classes.xml
cd demos
rm -f godot_demos.zip
zip -r godot_demos *
cd ..
cd tools/export/blender25
zip -r bettercollada *
mv bettercollada.zip ../../..
cd ../../..
Advanced
This page introduces the global organization of Godot Engine’s source code.
If you are writing or correcting bugs affecting Godot Engine editor, remember that the binary runs the launcher first,
which runs the editor in another process. Thus, you need to run the editor directly by passing the -e argument to
Godot Engine editor’s binary.
Godot has a rich set of classes and template that make for its core, and everything is built upon them.
This reference will try to list them in order for their better understanding.
Definitions
Godot uses the standard C98 datatypes, such as uint8_t, uint32_t, int64_t, etc. which are nowadays sup-
ported by every compiler. Reinventing the wheel for those is not fun, as it makes code more difficult to read.
In general, care is not taken to use the most efficient datatype for a given task unless using large structures or arrays.
int is used through most of the code unless neccesary. This is done because nowadays every device has at least a 32
bits bus and can do such operations in one cycle. It makes code more readable too.
For files or memory sizes, size_t is used, which is warranted to be 64 bits.
For Unicode characters, CharType instead of wchar_t is used, because many architectures have 4 bytes long wchar_t,
where 2 bytes might be desired. However, by default, this has not been forced and CharType maps directly to wchar_t.
References:
• core/typedefs.h
853
Godot Engine Documentation, Release latest
Memory model
PC is a wonderful architecture. Computers often have gigabytes of RAM, terabytes of storage and gigahertz of CPU,
and when an application needs more resources the OS will just swap out the inactive ones. Other architectures (like
mobile or consoles) are in general more limited.
The most common memory model is the heap, where an application will request a region of memory, and the
underlying OS will try to fit it somewhere and return it. This often works best and is very flexible, but over time and
with abuse, this can lead to segmentation.
Segmentation slowly creates holes that are too small for most common allocations, so that memory is wasted. There
is a lot of literature about heap and segmentation, so this topic will not be developed further here. Modern Operating
Systems use paged memory, which helps mitigate the problem of segmentation but doesn’t solve it.
However, in many studies and tests, it is shown that given enough memory, if the maximum allocation size is below a
given threshold in proportion to the maximum heap size and proportion of memory intended to be unused, segmenta-
tion will not be a problem over time as it will remain constant. In other words, just leave 10-20% of your memory free
and perform all small allocations and you are fine.
Godot ensures that all objects that can be allocated dynamically are small (less than a few kb at most). But what
happens if an allocation is too large (like an image or mesh geometry or large array)? In this case Godot has the option
to use a dynamic memory pool. This memory needs to be locked to be accessed, and if an allocation runs out of
memory, the pool will be rearranged and compacted on demand. Depending on the need of the game, the programmer
can configure the dynamic memory pool size.
Allocating memory
Godot has many tools for tracking memory usage in an game, specially during debug. Because of this, the regular C
and C++ library calls should not be used. Instead, a few other ones are provided.
For C-style allocation, Godot provides a few macros:
memalloc()
memrealloc()
memfree()
These are equivalent to the usual malloc, realloc, free of the standard library.
For C++-style allocation, special macros are provided:
memnew( Class / Class(args) )
memdelete( instance )
DVector is just a standard vector class, it can be accessed using the [] operator, but that’s probably slow for large
amount of accesses (as it has to lock internally). A few helpers exist for this:
DVector::Read r = dvector.read()
int someint = r[4]
and
DVector::Write w = dvector.write()
w[4]=22;
respectively. These allow fast read/write from DVectors and keep it locked until they go out of scope.
References:
• core/os/memory.h
• core/dvector.h
Containers
References:
• core/vector.h
• core/list.h
• core/set.h
• core/map.h
String
Godot also provides a String class. This class has a huge amoun of features, full Unicode support in all the functions
(like case operations) and utf8 parsing/exracting, as well as helpers for conversion and visualization.
References:
• core/ustring.h
StringName
StringNames are like a String, but they are unique. Creating a StringName from a string results in a unique internal
pointer for all equal strings. StringNames are really useful for using strings as identifier, as comparing them is
basically comparing a pointer.
Creation of a StringName (specially a new one) is slow, but comparison is fast.
References:
• core/string_db.h
Math types
There are several linear math types available in the core/math directory, they are basically just that.
References:
• core/math
NodePath
This is a special datatype sued for storing paths in a scenetree and referencing them fast.
References:
• core/path_db.h
RID
RIDs are resource IDs. Servers use these to reference data stored in them. RIDs are opaque, meaning that the data
they reference can’t be accessed directly. RIDs are unique, even for different types of referenced data.
References:
• core/rid.h
About
Variant is the most important datatype of Godot, it’s the most important class in the engine. A Variant takes up only
20 bytes and can store almost any engine datatype inside of it. Variants are rarely used to hold information for long
periods of time, instead they are used mainly for communication, editing, serialization and generally moving data
around.
A Variant can:
• Store almost any datatype
• Perform operations between many variants (GDScript uses Variant as it’s atomic/native datatype).
• Be hashed, so it can be compared quickly to over variants
• Be used to convert safely between datatypes
• Be used to abstract calling methods and their arguments (Godot exports all it’s functions through variants)
• Be used to defer calls or move data between threads.
• Be serialized as binary and stored to disk, or transfered via network.
• Be serialized to text and use it for printing values and editable settings.
• Work as an exported property, so the editor can edit it universally.
• Be used for dictionaries, arrays, parsers, etc.
Basically, thanks to the Variant class, writing Godot itself was a much, much easier task, as it allows for highly
dynamic constructs not common of c++ with little effort. Become a friend of Variant today.
References:
• core/variant.h
Both are implemented using variants. A Dictionary can match any datatype used as key to any other datatype. An
Array just holds an array of Variants. Of course, a Variant can also hold a Dictionary and an Array inside, making it
even more flexible.
Both have a shared mode and a COW mode. Scripts often use them in shared mode (meaning modifications to a
container will modify all references to it), or COW mode (modifications will always alter the local copy, making a
copy of the internal data if necessary, but will not affect the other copies). In COW mode, Both Dictionary and Array
are thread-safe, otherwise a Nutex should be created to lock if multi thread access is desired.
References:
• core/dictionary.h
• core/array.h
Object is the base class for almost everything. Most classes in Godot inherit directly or indirectly from it. Objects
provide reflection and editable properties, and declaring them is a matter of using a single macro like this.
class CustomObject : public Object {
obj2 = obj->cast_to<OtherType>(); // converting between types, this also works without RTTI enabled.
References:
• core/object.h
Registering an Object
ObjectTypeDB is a static class that hold the entire list of registered classes that inherit from object, as well as dynamic
bindings to all their methods properties and integer constants.
Classes are registered by calling:
ObjectTypeDB::register_type()
Registering it will allow the type to be instanced by scripts, code, or creating them again when deserializing.
Registering as virtual is the same but it can’t be instanced.
ObjectTypeDB::register_virtual_type()
Object derived classes can override a static function static void _bind_methods(), when one class is regis-
tered, this static function is called to register all the object methods, properties, constants, etc. It’s only called once. If
an Object derived class is instanced but has not been registered, it will be registered as virtual automatically.
Inside _bind_methods, there are a couple of things that can be done. Registering functions is one:
ObjectTypeDB::register_method(_MD("methodname","arg1name","arg2name"),&MyCustethod);
_MD is a macro that convers “methodname” to a stringname for more efficiency. Argument names are used for instro-
spection, but when compiling on release, the macro ignores them, so the strings are unused and optimized away.
Check _bind_methods of Control or Object for more examples.
If just adding modules and functionality that is not expected to be documented as throughly, the _MD() macro can
safely be ignore and a string passing the name can be passed for brevity.
References:
• core/object_type_db.h
Constants
For these to work when binding to methods, the enum must be declared convertible to int, for this a macro is provided:
VARIANT_ENUM_CAST( MyClass::SomeMode); // now functions that take SomeMode can be bound.
Properties (set/get)
For example:
PropertyInfo(Variant::INT,"amount",PROPERTY_HINT_RANGE,"0,49,1",PROPERTY_USAGE_EDITOR)
This is an integer property, named “amount”, hint is a range, range goes from 0 to 49 in steps of 1 (integers). It is only
usable for the editor (edit value visually) but wont be serialized.
or
PropertyInfo(Variant::STRING,"modes",PROPERTY_HINT_ENUM,"Enabled,Disabled,Turbo")
This is a string property, can take any string but the editor will only allow the defined hint ones. Since no hint flags
were specified, the default ones are PROPERTY_USAGE_STORAGE and PROPERTY_USAGE_EDITOR.
There are plenty of hints and usages available in object.h, give them a check.
Properties can also work like C# properties and be accessed from script using indexing, but ths usage is generally
discouraged, as using functions is preferred for legibility. Many properties are also bound with categories, such as
“animation/frame” which also make indexing imposssible unless using operator [].
From _bind_methods(), properties can be created and bound as long as a set/get functions exist. Example:
ADD_PROPERTY( PropertyInfo(Variant::INT,"amount"), _SCS("set_amount"), _SCS("get_amount") )
This creates the property using the setter and the getter. _SCS is a macro that creates a StringName efficiently.
An additional method of creating properties exists when more flexibility is desired (i.e. adding or removing properties
on context):
The following functions can be overriden in an Object derived class, they are NOT virtual, DO NOT make them virtual,
they are called for every override and the previous ones are not invalidated (multilevel call).
void _get_property_info(List *r_props); //return list of propertes
bool _get(const StringName& p_property, Variany& r_value) const; //return true if property was found
bool _set(const StringName& p_property, const Variany& p_value); //return true if property was found
This is also a little less efficient since p_property must be compared against the desired names in serial order.
Dynamic casting
Godot provides dynamic casting between Object Derived classes, for example:
void somefunc(Object *some_obj) {
If cast fails, NULL is returned. This system uses RTTI, but it also works fine (although a bit slower) when RTTI is
disabled. This is useful on platforms where a very small binary size is ideal, such as HTML5 or consoles (with low
memory footprint).
Signals
Objects can have a set of signals defined (similar to Delegates in other languages). Connecting to them is rather easy:
obj->connect(,target_instance,target_method)
//for example
obj->connect("enter_tree",this,"_node_entered_tree")
References
Reference inherits from Object and holds a reference count. It is the base for reference counted object types. Declaring
them must be done using Ref<> template. For example.
class MyReference: public Reference {
OBJ_TYPE( MyReference ,Reference);
};
myref is reference counted. It will be freed when no more Ref<> templates point to it.
References:
• core/reference.h
Resources:
Resource inherits from Reference, so all resources are reference counted. Resources can optionally contain a path,
which reference a file on disk. This can be set with resource.set_path(path). This is normally done by the
resource loader though. No two different resources can have the same path, attempt to do so will result in an error.
Resources without a path are fine too.
References:
• core/resource.h
Resource loading
If a reference to that resource has been loaded previously and is in memory, the resource loader will return that
reference. This means that there can be only one resource loaded from a file referenced on disk at the same time.
• resourceinteractiveloader (TODO)
References:
• core/io/resource_loader.h
Resource saving
Instance will be saved. Sub resources that have a path to a file will be saved as a reference to that resource. Sub
resources without a path will be bundled with the saved resource and assigned sub-IDs, like “res://somereource.res::1”.
This also helps to cache them when loaded.
References:
• core/io/resource_saver.h
Modules
Godot allows extending the engine in a modular way. New modules can be created and then enabled/disabled. This
allows for adding new engine functionality at every level without modifying the core, which can be split for use and
reuse in different modules.
Modules are located in them modules/ subdirectory of the build system. By default, two modules exist, GDScript
(which, yes it’s not part of the core engine), and the GridMap. As many new modules as desired can be created and
combined, and the SCons build system will take care of it transparently.
What for?
While it’s recommended that most of a game is written in scripting (as it is an enormous time saver), it’s perfectly
possible to use C++ instead. Adding C++ modules can be useful in the following scenarios:
• Binding an external library to Godot (like Bullet, Physx, FMOD, etc).
• Optimize critical parts of a game.
• Adding new functionality to the engine and/or editor.
• Porting an existing game.
• Write a whole, new game in C++ because you can’t live without C++.
Before creating a module, make sure to download the source code of Godot and manage to compile it. There are
tutorials in the wiki for this.
To create a new module, the first step is creating a directory inside modules. If you want to maintain the module
separately, you can checkout a different VCS into modules and use it.
The example module will be called “sumator”, and is placed inside the Godot source tree (C:\\godot refers to wherever
the Godot sources are located):
c:\\godot> cd modules
c:\\godot\\modules> mkdir sumator
c:\\godot\\modules> cd sumator
c:\\godot\\modules\\sumator>
#include "reference.h"
int count;
protected:
static void _bind_methods();
public:
Sumator();
};
#endif
#include "sumator.h"
count+=value;
}
void Sumator::reset() {
count=0;
}
return count;
}
void Sumator::_bind_methods() {
ObjectTypeDB::bind_method("add",&Sumator::add);
ObjectTypeDB::bind_method("reset",&Sumator::reset);
ObjectTypeDB::bind_method("get_total",&Sumator::get_total);
}
Sumator::Sumator() {
count=0;
}
Then, the new class needs to be registered somehow, so two more files need to be created:
register_types.h
register_types.cpp
void register_sumator_types();
void unregister_sumator_types();
/* yes, the word in the middle must be the same as the module folder name */
/* register_types.cpp */
#include "register_types.h"
#include "object_type_db.h"
#include "sumator.h"
void register_sumator_types() {
ObjectTypeDB::register_type<Sumator>();
}
void unregister_sumator_types() {
//nothing to do here
}
Next, we need to create a SCsub so the build system compiles this module:
# SCsub
Import('env')
And finally, the configuration file for the module, this is a simple python script that must be named ‘config.py’
# config.py
def can_build(platform):
return True
def configure(env):
pass
The module is asked if it’s ok to build for the specific platform (in this case, True means it will build for every
platform).
The second function allows to customize the build process for the module, like adding special compiler flags, options
etc. (This can be done in SCSub, but configure(env) is called at a previous stage). If unsure, just ignore this.
And that’s it. Hope it was not too complex! Your module should look like this:
godot/modules/sumator/config.py
godot/modules/sumator/sumator.h
godot/modules/sumator/sumator.cpp
godot/modules/sumator/register_types.h
godot/modules/sumator/register_types.cpp
godot/modules/sumator/SCsub
You can then zip it and share the module with everyone else. When building for every platform (instructions in the
previous section), your module will be included.
Using your newly created module is very easy, from any script you can do:
var s = Sumator.new()
s.add(10)
s.add(20)
s.add(30)
print( s.get_total() )
s.reset()
Summing up
As you see, it’s really easy to develop Godot in C++. Just write your stuff normally and remember to:
• use OBJ_TYPE macro for inheritance, so Godot can wrap it
• use _bind_methods to bind your functions to scripting, and to allow them to work as callbacks for signals.
But this is not all, depending what you do, you will be greeted with some surprises.
• If you inherit from [[API:Node]] (or any derived node type, such as Sprite), your new class will appear in the
editor, in the inheritance tree in the “Add Node” dialog.
• If you inherit from [[API:Resource]], it will appear int the resource list, and all the exposed properties can be
serialized when saved/loaded.
• By this same logic, you can extend the Editor and almost any area of the engine.
Introduction
Making videogames portable is all fine and dandy, until mobile gaming monetization shows up.
This area is complex, usually a mobile game that monetizes needs special connections to a server for stuff such as:
• Analytics
• In-app purchases
• Receipt validation
• Install tracking
• Ads
• Video ads
• Cross promotion
• In-game soft & hard currencies
• Promo codes
• A/B testing
• Login
• Cloud saves
• Leaderboards and scores
• User support & feedback
• Posting to Facebook, Twitter, etc.
• Push notifications
Oh yeah, developing for mobile is a lot of work. On iOS, you can just write a C++ module and take advantage of the
C++/ObjC intercommunication, so this is rather easy.
For C++ developers Java is a pain, the build system is severely bloated and interfacing it with C++ through JNI (Java
Native Interface) is more pain that you don’t want even for your worst enemy.
Maybe REST?
Most of these APIs allow communication via REST+JSON APIs. Godot has great support for HTTP, HTTPS and
JSON, so consider this as an option that works in every platform. Only write the code once and you are set to go.
Popular engines that have half the share of apps published on mobile get special plugins written just for them. Godot
does not have that luxury yet. So, if you write a REST implementation of a SDK for Godot, please share it with the
community.
Android module
Writing an Android module is similar to [[Custom modules in C++]], but needs a few more steps.
Make sure you are familiar with building your own [[Compiling for Android|Android export templates]], as well as
creating [[Custom modules in C++|custom modules]].
config.py
In the config.py for the module, some extra functions are provided for convenience. First, it’s often wise to detect if
android is being built and only enable building in this case:
def can_build(plat):
return plat=="android"
If more than one platform can be built (typical if implementing the module also for iOS), check manually for Android
in the configure functions:
def can_build(plat):
return plat=="android" or plat=="iphone"
def configure(env):
if env['platform'] == 'android':
#androd specific code
Java singleton
An android module will usually have a singleton class that will load it, this class inherits from
Godot.SingletonBase. A singleton object template follows:
//namespace is wrong, will eventually change
package com.android.godot;
// you might want to try initializing your singleton here, but android
// threads are weird and this runs in another thread, so you usually have to do
activity.runOnUiThread(new Runnable() {
public void run() {
//useful way to get config info from engine.cfg
String key = GodotLib.getGlobal("plugin/api_key");
SDK.initializeHere();
}
});
Calling back to Godot from Java is a little more difficult. The instance ID of the script must be known first, this is
obtained by calling get_instance_ID() on the script. This returns an integer that can be passed to Java.
From Java, use the calldeferred function to communicate back with Godot. Java will most likely run in a separate
thread, so calls are deferred:
GodotLib.calldeferred(, "", new Object[]{param1,param2,etc});
Add this singleton to the build of the project by adding the following to config.py:
def can_build(plat):
return plat=="android" or plat=="iphone"
def configure(env):
if env['platform'] == 'android':
# will copy this to the java folder
env.android_module_file("MySingleton.java")
#env.android_module_file("MySingleton2.java") call again for more files
AndroidManifest
Some SDKs need custom values in AndroidManifest.xml. Permissions can be edited from the godot exporter so there
is no need to add those, but maybe other functionalities are needed.
Create the custom chunk of android manifest and put it inside the module, add it like this:
def can_build(plat):
return plat=="android" or plat=="iphone"
def configure(env):
if env['platform'] == 'android':
# will copy this to the java folder
env.android_module_file("MySingleton.java")
env.android_module_manifest("AndroidManifestChunk.xml")
SDK library
So, finally it’s time to add the SDK library. The library can come in two flavors, a JAR file or an Android project for
ant. JAR is the easiest to integrate, just put it in the module directory and add it:
def can_build(plat):
return plat=="android" or plat=="iphone"
def configure(env):
if env['platform'] == 'android':
# will copy this to the java folder
env.android_module_file("MySingleton.java")
env.android_module_manifest("AndroidManifestChunk.xml")
env.android_module_library("MyLibrary-3.1.jar")
SDK project
When this is an Android project, things usually get more complex. Copy the project folder inside the module directory
and configure it:
c:\\godot\\modules\\mymodule\\sdk-1.2> android -p . -t 15
As of this writing, godot uses minsdk 10 and target sdk 15. If this ever changes, should be reflected in the manifest
template:
https://github.com/okamstudio/godot/blob/master/platform/android/AndroidManifest.xml.template
Then, add the module folder to the project:
def can_build(plat):
return plat=="android" or plat=="iphone"
def configure(env):
if env['platform'] == 'android':
# will copy this to the java folder
env.android_module_file("MySingleton.java")
env.android_module_manifest("AndroidManifestChunk.xml")
env.android_module_source("sdk-1.2","")
Building
As you probably modify the contents of the module, and modify your .java inside the module, you need the module to
be built with the rest of Godot, so compile android normally.
c:\\godot> scons p=android
This will cause your module to be included, the .jar will be copied to the java folder, the .java will be copied to the
sources folder, etc. Each time you modify the .java scons must be called.
Afterwards, just build the ant project normally:
This should generate the apk used as export template properly, as defined in the [[Compiling for Android|Android
build instructions]].
Usually to generate the apk, again both commands must be run in sequence:
c:\\godot> scons p=android
c:\\godot\\platform\\android\\java> ant release
To use the Module from GDScript, first enable the singleton by adding the following line to engine.cfg:
[android]
modules="com/android/godot/MySingleton"
More than one singleton module can be enable by separating with comma:
[android]
modules="com/android/godot/MySingleton,com/android/godot/MyOtherSingleton"
Then just request the singleton Java object from Globals like this:
#in any file
var singleton=null
func _init():
singleton = Globals.get_singleton("MySingleton")
print( singleton.myFunction("Hello") )
Troubleshooting
Future
Godot has an experimental Java API Wrapper that allows to use the entire Java API fro GDScript.
It’s simple to use and it’s used like this:
class = JavaClassWrapper.wrap()
This is most likely not functional yet, if you want to test it and help us make it work, contact us through the developer
mailing list.
Introduction
Godot has a simple serialization API based on Variant. It’s used for converting data types to an array of bytes efficiently.
This API is used in the functions get_var and store_var of the [[API:File|File class]] as well as the packet APIs
for [[API:PacketPeer]]. This format is not used for binary scenes and resources.
Packet specification
The packet is designed to be always padded to 4 bytes. All values are little endian encoded.
All packets have a 4 byte header representing an integer, specifying the type of data:
Type Value
0 null
1 bool
2 integer
3 float
4 string
5 vector2
6 rect2
7 vector3
8 matrix32
9 plane
10 quaternion
11 aabb (rect3)
12 matrix3x3
13 transform (matrix 4x3)
14 color
15 image
16 node path
17 rid (unsupported)
18 object (unsupported)
19 input event
20 dictionary
21 array
22 byte array
23 int array
24 float array
25 string array
26 vector2 array
27 vector3 array
28 color array
Following this is the actual packet contents, which varies for each type of packet:
0: null
1: bool
2: integer
3: float
4: string
5: vector2
6: rect2
7: vector3
8: matrix32
9: plane
10: quaternion
12: matrix3x3
14: color
15: image
20: dictionary
Then what follows is, for amount of “elements”, pairs of key and value, one after the other, using
this same format.
21: array
Then what follows is, for amount of “elements”, values one after the other, using
this same format.
13.3 Misc
Some developers like using the command line extensively. Godot is designed to be friendly to them, so here are the
steps for working entirely from the command line. Given the engine relies on little to no external libraries, initialization
times are pretty fast, making it suitable for this workflow.
Path
It is recommended that your godot binary is in your PATH environment variable, so it can be executed easily from any
place by typing godot. You can do so on Linux by placing the Godot binary in /usr/local/bin and making
sure it is called godot.
Creating a project
Creating a project from the command line is simple, just navigate the shell to the desired place and just make an
engine.cfg file exist, even if empty.
user@host:~$ mkdir newgame
user@host:~$ cd newgame
user@host:~/newgame$ touch engine.cfg
Running the editor is done by executing godot with the -e flag. This must be done from within the project directory,
or a subdirectory, otherwise the command is ignored and the project manager appears.
user@host:~/newgame$ godot -e
If a scene has been created and saved, it can be edited later by running the same code with that scene as argument.
user@host:~/newgame$ godot -e scene.xml
Erasing a scene
Godot is friends with your filesystem, and will not create extra metadata files, simply use rm to erase a file. Make sure
nothing references that scene, or else an error will be thrown upon opening.
user@host:~/newgame$ rm scene.xml
To run the game, simply execute Godot within the project directory or subdirectory.
user@host:~/newgame$ godot
When a specific scene needs to be tested, pass that scene to the command line.
user@host:~/newgame$ godot scene.xml
Debugging
Catching errors in the command line can be a difficult task because they just fly by. For this, a command line debugger
is provided by adding -d. It works for both running the game or a simple scene.
user@host:~/newgame$ godot -d
Exporting
Exporting the project from the command line is also supported. This is specially useful for continuous integration
setups. The version of Godot that is headless (server build, no video) is ideal for this.
The platform names recognized by the -export switch are the same as displayed in the export wizard of the editor.
To get a list of supported platforms from the command line, just try exporting to a non-recognized platform and the
full listing of platforms your configuration supports will be shown.
To export a debug version of the game, use the -export_debug switch instead of -export. Their parameters and
usage are the same.
Running a script
It is possible to run a simple .gd script from the command line. This feature is specially useful in very large projects,
for batch conversion of assets or custom import/export.
The script must inherit from SceneTree or MainLoop.
func _init():
print("Hello!")
quit()
If no engine.cfg exists at the path, current path is assumed to be the current working directory (unless -path is
specified).
Godot allows changing the font for the editor, and the font for the code editor. Both need to be in .fnt format, so they
need to be imported somewhere using the [[Importing fonts|font import tool]].
Then copy or do whathever you want with the font, as long as the location does not change, and set the relevant
property in Editor Settings. Code editor font is refreshed automatically, but the editor needs to be restarted for the new
global font to take effect.
At the moment, there are 2 iOS APIs partially implemented, GameCenter and Storekit. Both use the same model of
asynchronous calls explained below.
Asynchronous methods
When requesting an asynchronous operation, the method will look like this:
The parameter will usually be a Dictionary, with the information necessary to make the request, and the call will have
2 phases. First, the method will immediately return an Error value. If the Error is not ‘OK’, the call operation is
completed, with an error probably caused locally (no internet connection, API incorrectly configured, etc). If the error
value is ‘OK’, a response event will be produced and added to the ‘pending events’ queue. Example:
func on_purchase_pressed():
var result = InAppStore.purchase( { "product_id": "my_product" } )
if result == OK:
animation.play("busy") # show the "waiting for response" animation
else:
show_error()
Remember that when a call returns OK, the API will always produce an event through the pending_event interface,
even if it’s an error, or a network timeout, etc. You should be able to, for example, safely block the interface waiting
for a reply from the server. If any of the APIs don’t behave this way it should be treated as a bug.
The pending event interface consists of 2 methods:
• get_pending_event_count() Returns the number of pending events on the queue.
• Variant pop_pending_event() Pops the first event from the queue and returns it.
Store Kit
Implemented in platform/iphone/in_app_store.mm
The Store Kit API is accessible through the “InAppStore” singleton (will always be available from gdscript). It is
initialized automatically. It has 2 methods for purchasing:
• Error purchase(Variant p_params);
• Error request_product_info(Variant p_params);
and the pending_event interface
int get_pending_event_count();
Variant pop_pending_event();
purchase
Parameters Takes a Dictionary as a parameter, with one field, product_id, a string with your product id. Exam-
ple:
Response event The response event will be a dictionary with the following fields:
On error:
{
"type": "purchase",
"result": "error",
"product_id": "the product id requested"
}
On success:
{
"type": "purchase",
"result": "ok",
"product_id": "the product id requested"
}
request_product_info
Parameters Takes a Dictionary as a parameter, with one field, product_ids, a string array with a list of product
ids. Example:
var result = InAppStore.request_product_info( { "product_ids": ["my_product1", "my_product2"] } )
Response event The response event will be a dictionary with the following fields:
{
"type": "product_info",
"result": "ok",
"invalid_ids": [ list of requested ids that were invalid ],
"ids": [ list of ids that were valid ],
"titles": [ list of valid product titles (corresponds with list of valid ids) ],
"descriptions": [ list of valid product descriptions ] ,
"prices": [ list of valid product prices ],
"localized_prices": [ list of valid product localized prices ],
}
Game Center
Implemented in platform/iphone/game_center.mm
The Game Center API is available through the “GameCenter” singleton. It has 6 methods:
• Error post_score(Variant p_score);
• Erroraward_achievement(Variant p_params);
• Error reset_achievements();
• Error request_achievements();
• Error request_achievement_descriptions();
• Error show_game_center(Variant p_params);
plus the standard pending event interface.
post_score
Response event The response event will be a dictionary with the following fields:
On error:
{
"type": "post_score",
"result": "error",
"error_code": the value from NSError::code,
"error_description": the value from NSError::localizedDescription,
}
On success:
{
"type": "post_score",
"result": "ok",
}
award_achievement
Response event The response event will be a dictionary with the following fields:
On error:
{
"type": "award_achievement",
"result": "error",
"error_code": the error code taken from NSError::code,
}
On success:
{
"type": "award_achievement",
"result": "ok",
}
reset_achievements
Response event The response event will be a dictionary with the following fields:
On error:
{
"type": "reset_achievements",
"result": "error",
"error_code": the value from NSError::code
}
On success:
{
"type": "reset_achievements",
"result": "ok",
}
request_achievements
Request all the Game Center achievements the player has made progress on. The function takes no parameters.
Response event The response event will be a dictionary with the following fields:
On error:
{
"type": "achievements",
"result": "error",
"error_code": the value from NSError::code
}
On success:
{
"type": "achievements",
"result": "ok",
"names": [ list of the name of each achievement ],
"progress": [ list of the progress made on each achievement ]
}
request_achievement_descriptions
Request the descriptions of all existing Game Center achievements regardless of progress. The function takes no
parameters.
Response event The response event will be a dictionary with the following fields:
On error:
{
"type": "achievement_descriptions",
"result": "error",
"error_code": the value from NSError::code
}
On success:
{
"type": "achievement_descriptions",
"result": "ok",
"names": [ list of the name of each achievement ],
"titles": [ list of the title of each achievement ]
"unachieved_descriptions": [ list of the description of each achievement when it is unachieved ]
"achieved_descriptions": [ list of the description of each achievement when it is achieved ]
"maximum_points": [ list of the points earned by completing each achievement ]
"hidden": [ list of booleans indicating whether each achievement is initially visible ]
"replayable": [ list of booleans indicating whether each achievement can be earned more than once ]
}
show_game_center
Displays the built in Game Center overlay showing leaderboards, achievements, and challenges.
Response event The response event will be a dictionary with the following fields:
On close:
{
"type": "show_game_center",
"result": "ok",
}
Multi-platform games
When working on a multi-platform game, you won’t always have the “GameCenter” singleton available (for example
when running on PC or Android). Because the gdscript compiler looks up the singletons at compile time, you can’t
just query the singletons to see and use what you need inside a conditional block, you need to also define them as valid
identifiers (local variable or class member). This is an example of how to work around this in a class:
var GameCenter = null # define it as a class member
func post_score(p_score):
if GameCenter == null:
return
GameCenter.post_score( { "value": p_score, "category": "my_leaderboard" } )
func check_events():
while GameCenter.get_pending_event_count() > 0:
# do something with events here
pass
func _ready():
# check if the singleton exists
if Globals.has_singleton("GameCenter"):
GameCenter = Globals.get_singleton("GameCenter")
# connect your timer here to the "check_events" function
Contributing
This page describes the typical workflow of the bug triage team aka bugsquad when handling issues and pull re-
quests on Godot’s GitHub repository. It is bound to evolve together with the bugsquad, so do not hesitate to propose
modifications to the following guidelines.
14.1.1 Labels
885
Godot Engine Documentation, Release latest
The categories are used for general triage of the issues. They can be combined in some way when relevant, e.g. an
issue can be labelled Bug, Confirmed and High priority at the same time if it’s a critical bug that was confirmed by
several users, or Feature request and Needs discussion if it’s a non-consensual feature request, or one that is not precise
enough to be worked on.
Topics:
• Buildsystem: relates to building issues, either linked to the SCons buildsystem or to compiler peculiarities.
• Core: anything related to the core engine. It might be further split later on as it’s a pretty big topic.
• Demos: relates to the official demos.
• GDScript: relates to GDScript.
• Porting: relates to some specific platforms.
• Rendering engine: relates to the 2D and 3D rendering engines.
• User interface: relates to the UI design.
Issues would typically correspond to only one topic, though it’s not unthinkable to see issues that fit two bills. The
general idea is that there will be specialized contributors teams behind all topics, so they can focus on the issues
labelled with their team topic.
Bug reports concerning the website or the documentation should not be filed in GitHub but in the appropriate tool in
OpenProject, therefore such issues should be closed and archived once they have been moved to their rightful platform.
14.1.2 Milestones
Milestones correspond to planned future versions of Godot for which there is an existing roadmap. Issues that fit in
the said roadmap should be filed under the corresponding milestone; if they don’t correspond to any current roadmap,
they should be set to Later. As a rule of thumb, an issue corresponds to a given milestone if it concerns a feature that
is new in the milestone, or a critical bug that can’t be accepted in any future stable release, or anything that Juan wants
to work on right now :)
This page describes the rules to follow if you want to contribute Godot Engine by writing documentation or translating
existing documentation.
A good documentation is well written in plain English and well-formed sentences. It is clear and objective.
A documentation page is not a tutorial page. We differentiate these concepts by these definitions :
• tutorial : a page aiming at explaining how to use one or more concepts in Godot Editor in order to achieve a
specific goal with a learning purpose (ie. “make a simple 2d Pong game”, “apply forces to an object”...)
• documentation : a page describing precisely one and only one concept at the time, if possible exhaustively (ie.
the list of methods of the Sprite class for example).
You are free to write the kind of documentation you wish, as long as you respect the following rules.
Creating a new documentation page or tutorial page is easy. The following rules must be respected :
• Choose a short and explicit title
• Respect the grammar and orthography
• Make use of the [[Wiki syntax]]
Try to structure your page in order to enable users to include a page directly in another page or even forum posts
using the include wiki syntax. For example, the syntax to include the page you are reading is :
!{{include(Documentation writing and translating guidelines)}}.
Titles
Also, avoid American CamelCase titles: titles’ first word should begin with a capitalized letter, and every following
word should not. Thus, this is a good example:
• Insert your title here And this is a bad example:
• Insert Your Title Here
Only project names (and people names) should have capitalized first letter. This is good:
• Starting up with Godot Engine and this is bad:
• Starting up with godot engine
If you intend to create a new page in your language, you are asked to firstly create the corresponding English page if
it doesn’t already exist. Do it even if you will not write it yourself, just leave it blank. Only then, create the
corresponding page in your own language. Maybe later, another contributor will translate your new page to English.
Remember : even if Godot aims at being accessible to everyone, English is the most frequent language for
documentation.
You are very welcome to translate existing pages from English to your language, or from your language to English.
If these guidelines were respected, an English page already exists for every page of this wiki, even if it is empty. To
translate an existing page, please follow these few rules :
• Respect the grammar and orthography
Please add a line at the very beginning of your translation, linking to the English base page you translate from :
Traduction de ![[Godot Engine:Creating 2D Games]]
The previous link is of the form ![[:]] which enables you to add a link to a page located in an other project. Here,
“Godot Engine” is the English project.
You are welcome to correct mistakes or styles to respect these guidelines. However, in case of important changes,
please do not start a discussion on this page : use the forum, create a new topic with a link to the incriminated page
and start discussing there about your remarks.
14.2.6 Licence
This wiki and every page it contains is published under the terms of the Creative Commons BY 3.0 license.
| Container | | |
| Control | | |
| ConvexPolygonShape | | |
| ConvexPolygonShape2D | Ovnuniarchos | Finished | |
| CubeMap | | |
| Curve2D | Ovnuniarchos | Finished |
| Curve3D | Ovnuniarchos | Finished |
| DampedSpringJoint2D | | |
| Dictionary | | |
| DirectionalLight | | |
| Directory | vnen | Started | 11/10/2015 |
| EditorFileDialog | | |
| EditorImportPlugin | | |
| EditorPlugin | | |
| EditorScenePostImport | | |
| EditorScript | | |
| Environment | | |
| EventPlayer | | |
| EventStream | | |
| EventStreamChibi | | |
| File | | |
| FileDialog | | |
| FixedMaterial | | |
| Font | | |
| FuncRef | | |
| GDFunctionState | | |
| GDNativeClass | | |
| GDScript | | |
| Generic6DOFJoint | | |
| Geometry | | |
| GeometryInstance | | |
| Globals | | |
| GraphEdit | StraToN | Finished | | may need a tutorial. I’ll think about it. |
| GraphNode | StraToN | Finished | | may need a tutorial. I’ll think about it. |
| GridContainer | | |
| GridMap | | |
| GrooveJoint2D | | |
| HBoxContainer | | |
| HButtonArray | | |
| HScrollBar | | |
| HSeparator | | |
| HSlider | | |
| HSplitContainer | | |
| HTTPClient | | |
| HingeJoint | | |
| IP | | |
| IP_Unix | | |
| Image | | |
| ImageTexture | | |
| ImmediateGeometry | | |
| Input | | |
| InputDefault | | |
| InputEvent | | |
| InputEventAction | | |
| InputEventJoyButton | | |
| InputEventJoyMotion | | |
| InputEventKey | | |
| InputEventMouseButton | | |
| InputEventMouseMotion | | |
| InputEventScreenDrag | | |
| InputEventScreenTouch | | |
| InputMap | | |
| IntArray | | |
| InterpolatedCamera | | |
| ItemList | | |
| Joint | | |
| Joint2D | | |
| KinematicBody | | |
| KinematicBody2D | Ovnuniarchos | Started | 2015/11/23 | |
| Label | | |
| LargeTexture | | |
| Light | | |
| Light2D | | |
| LightOccluder2D | | |
| LineEdit | | |
| LineShape2D | Ovnuniarchos | Finished | |
| MainLoop | | |
| MarginContainer | | |
| Marshalls | | |
| Material | | |
| MaterialShader | | |
| MaterialShaderGraph | | |
| Matrix3 | | |
| Matrix32 | | |
| MenuButton | | |
| Mesh | | |
| MeshDataTool | | |
| MeshInstance | | |
| MeshLibrary | | |
| MultiMesh | | |
| MultiMeshInstance | | |
| Mutex | | |
| Navigation | | |
| Navigation2D | | |
| NavigationMesh | | |
| NavigationMeshInstance | | |
| NavigationPolygon | | |
| NavigationPolygonInstance | | |
| Nil | | |
| Node | | |
| Node2D | | |
| NodePath | | |
| OS | | |
| Object | | |
| OccluderPolygon2D | | |
| OmniLight | | |
| OptionButton | | |
| PCKPacker | | |
| PHashTranslation | | |
| PackedDataContainer | | |
| PackedDataContainerRef | | |
| PackedScene | | |
| PacketPeer | | |
| PacketPeerStream | | |
| PacketPeerUDP | | |
| Panel | | |
| PanelContainer | | |
| ParallaxBackground | | |
| ParallaxLayer | | |
| ParticleAttractor2D | | |
| Particles | | |
| Particles2D | | |
| Patch9Frame | | |
| Path | Ovnuniarchos | Finished |
| Path2D | Ovnuniarchos | Finished |
| PathFollow | Ovnuniarchos | Finished |
| PathFollow2D | Ovnuniarchos | Finished |
| PathRemap | | |
| Performance | | |
| Physics2DDirectBodyState | | |
| Physics2DDirectBodyStateSW | | |
| Physics2DDirectSpaceState | | |
| Physics2DServer | | |
| Physics2DServerSW | | |
| Physics2DShapeQueryParameters | | |
| Physics2DShapeQueryResult | | |
| Physics2DTestMotionResult | | |
| PhysicsBody | | |
| PhysicsBody2D | Ovnuniarchos | Finished | 2015/12/22 | |
| PhysicsDirectBodyState | | |
| PhysicsDirectBodyStateSW | | |
| PhysicsDirectSpaceState | | |
| PhysicsServer | | |
| PhysicsServerSW | | |
| PhysicsShapeQueryParameters | | |
| PhysicsShapeQueryResult | | |
| PinJoint | | |
| PinJoint2D | | |
| Plane | | |
| PlaneShape | | |
| Polygon2D | | |
| PolygonPathFinder | | |
| Popup | | |
| PopupDialog | | |
| PopupMenu | | |
| PopupPanel | | |
| Portal | | |
| Position2D | | |
| Position3D | | |
| ProgressBar | | |
| ProximityGroup | | |
| Quad | | |
| Quat | | |
| RID | | |
| Range | | |
| RawArray | | |
| RayCast | | |
| RayCast2D | eska | Started | 2015-10-16 |
| RayShape | | |
| RayShape2D | Ovnuniarchos | Finished | |
| RealArray | | |
| Rect2 | bojidar_bg | Finished |
| RectangleShape2D | Ovnuniarchos | Finished | |
| Reference | | |
| ReferenceFrame | | |
| RegEx | Ovnuniarchos | Finished | 2015-11-03 |
| RemoteTransform2D | eska | Started | 2015-10-16 |
| RenderTargetTexture | | |
| Resource | | |
| ResourceImportMetadata | | |
| ResourceInteractiveLoader | | |
| ResourceLoader | | |
| ResourcePreloader | | |
| ResourceSaver | | |
| RichTextLabel | | |
| RigidBody | | |
| RigidBody2D | Ovnuniarchos | Started | 2015/11/23 | |
| Room | | |
| RoomBounds | | |
| Sample | Akien | Finished |
| SampleLibrary | Akien | Finished |
| SamplePlayer | Akien | Finished |
| SamplePlayer2D | Akien | Finished |
| SceneTree | | |
| Script | | |
| ScrollBar | | |
| ScrollContainer | | |
| SegmentShape2D | Ovnuniarchos | Finished | |
| Semaphore | | |
| Separator | | |
| Shader | | |
| ShaderGraph | | |
| ShaderMaterial | | |
| Shape | | |
| Shape2D | Ovnuniarchos | Finished | |
| Skeleton | | |
| Slider | | |
| SliderJoint | | |
| SoundPlayer2D | Akien | Not started |
| SoundRoomParams | Akien | Not started |
| Spatial | Akien | Not started |
| SpatialPlayer | Akien | Not started |
| SpatialSamplePlayer | Akien | Not started |
| SpatialSound2DServer | Akien | Not started |
| SpatialSound2DServerSW | Akien | Not started |
| SpatialSoundServer | Akien | Not started |
| SpatialSoundServerSW | Akien | Not started |
| SpatialStreamPlayer | Akien | Not started |
| SphereShape | | |
| SpinBox | | |
| SplitContainer | | |
| SpotLight | | |
| Sprite | | |
| Sprite3D | | |
| SpriteBase3D | | |
| SpriteFrames | | |
| StaticBody | | |
| StaticBody2D | Ovnuniarchos | Started | 2015/11/23 | |
| StreamPeer | | |
| StreamPeerSSL | | |
| StreamPeerTCP | | |
| StreamPlayer | | |
| String | | |
| StringArray | | |
| StyleBox | | |
| StyleBoxEmpty | | |
| StyleBoxFlat | | |
| StyleBoxImageMask | | |
| StyleBoxTexture | | |
| SurfaceTool | | |
| TCP_Server | | |
| TabContainer | | |
| Tabs | | |
| TestCube | | |
| TextEdit | | |
| Texture | | |
| TextureButton | | |
| TextureFrame | | |
| TextureProgress | | |
| Theme | | |
| Thread | | |
| TileMap | Akien | Finished |
| TileSet | Akien | Finished |
| Timer | Akien | Finished |
| ToolButton | | |
| TouchScreenButton | | |
| Transform | | |
| Translation | | |
| TranslationServer | | |
| Tree | | |
| TreeItem | | |
| Tween | | |
| UndoRedo | | |
| VBoxContainer | | |
| VButtonArray | | |
| VScrollBar | | |
| VSeparator | | |
| VSlider | | |
| VSplitContainer | | |
| Vector2 | bojidar_bg | Finished |
| Vector2Array | bojidar_bg | Finished |
| Vector3 | bojidar_bg | Finished |
| Vector3Array | bojidar_bg | Finished |
| VehicleBody | | |
| VehicleWheel | | |
| VideoPlayer | | |
| VideoStream | | |
| Viewport | | |
| ViewportSprite | | |
| VisibilityEnabler | | |
| VisibilityEnabler2D | | |
| VisibilityNotifier | | |
| VisibilityNotifier2D | | |
| VisualInstance | | |
| VisualServer | | |
| WeakRef | | |
| WindowDialog | | |
| World | | |
| World2D | | |
| WorldEnvironment | | |
| XMLParser | | |
| YSort | eska | Started | 2015-10-16 |
| bool | | |
| float | | |
| int | | |
Godot Engine provides an important number of classes that you can make use of to create your games. However, the
[[Reference|reference]] that lists all these classes with their methods is quite incomplete. We need your kind help to
fill this reference. This page will explain you how.
> Please note: we aim at filling completely this reference in English first. Please do not start translating it for the
moment.
[[List of classes and documenters]]
First of all, you need to fork the Godot Engine on your own GitHub repository.
You will then need to clone the master branch of Godot Engine in order to work on the most recent version of the
engine, including all of its features.
git clone https://github.com/your_name/godot.git
Then, create a new git branch that will contain your changes.
git checkout -b reference-edition
The branch you just created is identical to current master branch of Godot Engine. It already contains a doc/ folder,
with the currently written reference.
When classes are modified in the source code, the documentation template might become outdated. To make sure that
you are editing an up-to-date version, you first need to compile Godot (you can follow the [[Introduction to the Godot
buildsystem]] page), and then run the following command (assuming 64-bit Linux):
./bin/godot.x11.tools.64 -doctool doc/base/classes.xml
The doc/base/classes.xml should then be up-to-date with current Godot Engine features. You can then check what
changed (or not) using the git diff command. If there are changes to other classes than the one you are planning
to document, please commit those changes first before starting to edit the template:
git add doc/base/classes.xml
git commit -m "Sync classes reference template with current code base"
Once your modifications are finished, push your changes on your GitHub repository:
git add doc/base/classes.xml
git commit -m "Explain your modifications."
git push
When it’s done, you can ask for a Pull Request on the GitHub UI of your Godot fork.
First of all, check the [[List of classes and documenters]]. Try to work on classes not already assigned nor filled.
This file is produced by Godot Engine. It is used by the editor, for example in the Help window (F1, Shift+F1).
You can edit this file using your favourite text editor.
Base node for 2D system. Node2D contains a position, rotation and scale, which is used to position an
As you can see, some methods in this class have no description (i.e. there is no text between their marks). This can
also happen for the description and the brief_description of the class, but in our case they are already filled. Let’s edit
the description of the rotate() method:
Rotates the node of "degrees" degrees.
That’s all!
You simply have to write any missing text between these marks:
•
•
•
Describe clearly and shortly what it does. You can include an example of use if needed. Avoid grammar faults.
Not a problem. Leave it behind for now, and don’t forget to notify the missing methods when you request a pull of
your changes. Another editor will take care of it.
If you wonder what a method does, you can still have a look at its implementation in Godot Engine’s source code on
GitHub. Also, if you have a doubt, feel free to ask on the Forums and on IRC (freenode, #godotengine).