* Map is unnecessary and inefficient in almost every case.
* Replaced by the new HashMap.
* Renamed Map to RBMap and Set to RBSet for cases that still make sense
(order matters) but use is discouraged.
There were very few cases where replacing by HashMap was undesired because
keeping the key order was intended.
I tried to keep those (as RBMap) as much as possible, but might have missed
some. Review appreciated!
The stack now contains three special addresses that should no be copied
to the state, since it contains references that creates cycles. They can
be recreated when the function is resumed.
This commit also removes the clearing of stack from the
GDScriptFunctionState destructor, since it should be cleared when the
function exits. The state stack should only be cleared manually if the
instance is freed before the state resumes (which is already being
done). Otherwise this would destruct the stack twice, causing crashes.
Adds a new, cleaned up, HashMap implementation.
* Uses Robin Hood Hashing (https://en.wikipedia.org/wiki/Hash_table#Robin_Hood_hashing).
* Keeps elements in a double linked list for simpler, ordered, iteration.
* Allows keeping iterators for later use in removal (Unlike Map<>, it does not do much
for performance vs keeping the key, but helps replace old code).
* Uses a more modern C++ iterator API, deprecates the old one.
* Supports custom allocator (in case there is a wish to use a paged one).
This class aims to unify all the associative template usage and replace it by this one:
* Map<> (whereas key order does not matter, which is 99% of cases)
* HashMap<>
* OrderedHashMap<>
* OAHashMap<>
* Changed to use the same stages as extensions.
* Makes the initialization more coherent, helping solve problems due to lack of stages.
* Makes it easier to port between module and extension.
* removed the DRIVER initialization level (no longer needed).
Previously, there was an issue where the gdscript analyzer incorrectly
riased a validation error for code that had a default Dictionary, Array,
or custom type.
These typedefs don't save much typing compared to the full `Ref<Resource>`
and `Ref<RefCounted>`, yet they sometimes introduce confusion among
new contributors.
Didn't commit all the changes where it wants to initialize a struct
with `{}`. Should be reviewed in a separate PR.
Option `IgnoreArrays` enabled for now to be conservative, can be
disabled to see if it proposes more useful changes.
Also fixed manually a handful of other missing initializations / moved
some from constructors.
Rewrites the definition of how the function works.
Reworks the style of the examples and adds a negative range example.
Changes the while loop to a range loop in the array backwards example.
Convert method signature parameters to const where it is possible
# Conflicts:
# drivers/gles3/rasterizer_canvas_gles3.cpp
# drivers/gles3/rasterizer_canvas_gles3.h
# editor/plugins/animation_state_machine_editor.cpp
# editor/plugins/animation_state_machine_editor.h
Autoloaded scripts should always inherit from Node. When you run a
project that tries to autoload a script which doesn’t inherit from Node,
then Godot gives an error.
Before this change, the error said “Script does not inherit a Node”.
That error message is a little bit misleading. If a class inherits a
Node, then one of its superclasses has a Node. If a class inherits
_from_ Node, then one of its superclasses is Node. This change corrects
that mistake.
Fixes#59884.
* Ability to create script languages from GDExtension
* Some additions to gdnative_extension.h to make this happen
* Moved the GDExtension binder to core
This now allows creating scripting languages from GDExtension, with the same ease as if it was a module. It replaces the old PluginScript from Godot 3.x.
Warning: GodotCPP will need to be updated to support this (it may be a bit of work as ScriptInstance needs to be created over there again).
* GDScriptLanguage::complete_code already adds parentheses to function calls, and does this a lot smarter than the language server right now.
* Instead of the previous naive approach we now reuse the same logic as the internal editor.
* For this to have any effect we also have to send the `insertText` field already during the completionRequest and not only during resolve.
* Very old macros from the time Godot was created.
* Limited arguments to 5 (then later changed to 8) in many places.
* They were replaced by C++11 Variadic Templates.
* Renamed methods that take argument pointers to have a "p" suffix. This was used in some places and not in others, so made it standard.
* Also added a dereference check for Variant*. Helped catch a couple of bugs.
To guarantee polymorphism, a method signature must be compatible with
the parent. This checks if:
1. Return type is the same.
2. The subclass method takes at least the same amount of parameters.
3. The matching parameters have the same type.
4. If the subclass takes more parameters, all of the extra ones have a
default value.
5. If the superclass has default values, so must have the subclass.
There's a few test cases to ensure this holds up.
This makes it easier to spot syntax errors when editing the
class reference. The schema is referenced locally so validation
can still work offline.
Each class XML's schema conformance is also checked on GitHub Actions.
Since enums resolve to a dictionary at runtime, calling dictionary
methods on an enum type is a valid use case. This ensures this is true
by adding test cases. This also makes enum values be treated as ints
when used in operations.
- Fix compilation issues by disabling warnings on release builds. This
also strips warnings from expected result before the comparison to
avoid false mismatches.
- Add a `#debug-only` flag to tests. Must be the first line of the test
script. Those won't run with release builds. Can be used for test
cases that rely on checks only available on debug builds.
This makes sure that assigning values to enum-typed variables are
consistent. Same enum is always valid, different enum is always
invalid (without casting) and assigning `int` creates a warning
if there is no casting.
There are new test cases to ensure this behavior doesn't break in
the future.
This commit adds a condition to VariantCaster that casts Variants of type OBJECT to any type T, if T is derived from Object.
This change enables a fair bit of code cleanup. First, the Variant implicit cast operators for Node and Control can be removed, which allows for some invalid includes to be removed. Second, helper methods in Tree whose sole purpose was to cast arguments to TreeItem * are no longer necessary.
A few small changes also had to be made to other files, due to the changes cascading down all the includes.
Found via `codespell -q 3 -S ./thirdparty,*.po,./DONORS.md -L ackward,ang,ans,ba,beng,cas,childs,childrens,dof,doubleclick,expct,fave,findn,gird,hist,inh,inout,leapyear,lod,nd,numer,ois,ony,paket,ro,seeked,sinc,switchs,te,uint,varn,vew`
Format switch
Added a case for constant subscripts
Fixed default value hinting for the enum type
Removed is_null checking for value
Added a case for dictionary
