Allow implementing Object::_validate_property() from GDExtension
#81515
Conversation
dsnopek
force-pushed
the
gdextension-validate-property-object
branch
from
62a5a2c
to
abef8e3
Compare
|
I just snuck a tiny bug fix for PR #81261 into this PR. When working on that one, I didn't realize that there was a mismatch for the data type used for |
| @@ -527,6 +527,27 @@ void Object::get_property_list(List<PropertyInfo> *p_list, bool p_reversed) cons | |||
| void Object::validate_property(PropertyInfo &p_property) const { | |||
| _validate_propertyv(p_property); | |||
|
|
|||
| if (_extension && _extension->validate_property) { | |||
| // GDExtension uses a StringName rather than a String for property name. | |||
| StringName prop_name = p_property.name; | |||
There was a problem hiding this comment.
Regarding these StringName manipulations, should we share similar concerns to other extended Object methods around here? There is this whole preprocessor pragma guard every time we cast to GDExtensionStringNamePtr. We don't do it here, not directly at least, but maybe we should? And maybe we should add the same guards?
There was a problem hiding this comment.
I think the #pragma stuff is about the constness in those casts. In the code added in this PR, we don't need to mess with const, so I think it's OK? And running locally with GCC, I don't see any warnings.
There was a problem hiding this comment.
Alright, sounds fine then. I don't really understand this myself, so I'm trusting you :)
|
Thanks! |
This is a follow-up to PR #75778 (exposed
_validate_property()to scripting) and PR #81261 (exposed_validate_property()to scripting languages defined in GDExtesion)This PR exposes
_validate_property()on normal object types defined by GDExtension and registered inClassDBHere is companion PR godotengine/godot-cpp#1239 for godot-cpp - it includes a unit test that validates that this functionality works.