Using variables from other scenes (Godot) - godot

I have two scenes (Profile and Login) and I want to use the variable new_profile from Profile and use it in Login but when I autoload Profile and mark it as a singleton it doesn't let me run the program and gives me an index error and says that the nodes referenced in my variables do not exist even though before I autoloaded it, it worked perfectly fine

You would use a singleton to achieve this task. Godot calls it, autoload. Goto Project -> Project Settings -> Autoload. Click on the folder icon. Add your gdscript file with the variables needed. That will auto populate the field directly to the right. You can edit this name as you wish. Then click on add. When you play your program, those variables will be accessible by any script by way of this code. Godot official explanation and code available here.
PlayerVariables (added to Autoload)
var player_vars = get_node("/root/PlayerVariables")
player_vars.health -= 10
Or
PlayerVariables.health -= 10

Related

What is the ideal way to change the value of a progress bar with on-screen button press (Godot)

I'm new to coding and Godot and I need help with changing the value of my progress bar from the script inside my button node. The error I get is 'get_node: (Node not found: "/root/ProgressBar" (absolute path attempted from "/root/Node2D/Button").)' I want every time the button is '_pressed()' to increment health by 10 which would change 'value' in ProgressBar to that value.
extends Button
var health= 0
func _ready():
pass
func _process(delta):
pass
func _pressed():
health += 10
print(health)
get_node("/root/ProgressBar").set_value(health)
The path /root/ProgressBar is unlikely to be correct. Usually, as a child of root, you would have the current scene, and as child of that something else (So the path could be something like /root/MySceneName/ProgressBar). And relative paths work too… Usually, we would solve this by means of figuring out the correct path…
But things have changed (For Godot 3.5 and newer). What you are going to do is open the script while the scene is open, and then drag - with CTRL pressed - the Control you want to access (i.e. the ProgressBar) on script, outside of any method (func).
By doing that Godot will generate a line of code that looks something like this:
onready var progress_bar: ProgressBar = $"Path/To/ProgressBar"
In other words, Godot will figure out the path you need, and set up a variable you can use. Godot will set the variable to the Node you dragged as part of the initialization of you script. Assuming all goes well.
You might get an error saying that the script is not used in the scene. Perhaps you have the wrong script open, or you use the script in multiple places and Godot got confused… Either way, close the script and open it from the scene, and it should work.
And, of course, you can use the variable to set its properties, for example:
progress_bar.value = health
You might also be interested in scene unique names. On the contextual menu of the Nodes in the Scene panel you can select "% Access As Scene Unique Name", this will allow you to access the Node with this syntax:
$"%ProgressBar"
Regardless of where it is in the scene tree. Except, be aware that the name must be unique for that scene (you won't be able to do that with two Nodes that have the same name).
This has the advantage that if later you change where the Node is in the scene tree (for example to add some Container to organize your Controls), you don't have to update the path to where you use it. Instead it should continue to work, as long as you don't change the name.
Let that be yet another reason to pick good Node names.
And yes, dragging the Node to the script also works with these.
The problem with your path is, that your missing the name of your main node in it.
So let's say your node tree looks like this:
MainNode
ProgressBar
Button
To get the ProgressBar from anywhere you would need to use get_node("/root/MainNode/ProgressBar")

How to make the functions done for group nodes in a scene not change after changing scene?

I have put nodes in a group. And I have put a func on the scene script to make changes to the nodes in group. In that func,I made the nodes to queue free and stuffs like that. But when I change scene and come back to the previous scene,the queued free is back again,it is not queued free anymore. How do I make it not change even after changing scene?
Why information is lost
When you change the current scene with change_scene or change_scene_to the current scene is unloaded, and the new one is loaded and instanced.
Perhaps it helps to conceptualize that the instance of the scene is not the same as the scene in storage. So what happens is like opening a file in an editor, modifying and not saving it.
Alright, that is one solution if you really want to go that route: you can save a scene.
First create a PackedScene, for example:
var packed_scene := PackedScene.new()
Then tell it to pack the root node of the scene, for example:
packed_scene.pack(self)
Note: any Nodes that don't have their owner property set to the Node you passed here will be ignored. And no, add_child does not set the owner property.
If you have been paying attention, you know you can give a PackedScene to change_scene_to... But how do you keep it around when the scene changes?
There are a few ways. And the thing is: if we can keep information around, we might not need to save the scene.
Keep things around in autoloads
The scripts and scene you add to your autoloads (singleton) in project settings stay there (unless you explicitly remove them) even when you change scene.
Thus, a very simple way to keep information around is to have a script with some variables that you can write and read from anywhere in your project. And you can do that with an autoload.
Keep things around in resources
Godot caches resources. When you load the same resource in multiple places, you actually get the same object.
Well, you know you can create a custom Resource class. To do that, on the context menu of the FileSystem panel select "New Script", and in the script make a class that extends Resource, and give it a class_name.
Then you can create resources of that Resource from the context menu of the FielSystem panel by selecting "New Resource" and picking your Resource class when asked.
Everywhere you load one of those, you are going to get the same object. Yes, even if it is in another scene. So you can add variables to your Resource class, write them in one scene and read them in another.
I explain a more concrete example elsewhere.
Keep things around in storage
You can write to a file and read from a file. For example you can do this:
# save
var file = File.new()
file.open("user://a.sav", File.WRITE)
file.store_pascal_string(var2str(data))
file.close()
# load
file.open("user://a.sav", File.READ)
data = str2var(file.get_pascal_string())
file.close()
Or, if what you want to store is a Resource (be it a PackedScene, or some other Resource class including a custom one), you can use ResourceSaver and ResourceLoader:
# save
ResourceSaver.save("user://a.tres", resource)
# load
resource = ResourceSaver.load("user://a.tres")
Of course, you can also load resources with load, or preload. You may also be interested in Background Loading.
By the way, if you are going to save player progress, having all the player progress data in a single object makes sense. And if you are going to have all the data you are keeping all the player progress data in a single place, it makes sense it would be somewhere it is accesible all the time and stays around even when you change scenes, thus: put the player progress data in an autoload. Bonus: you can put the functions to save and load to a file there too.
Don't change scenes
Since changing scenes brings problems - consider another solution: don't.
You can load an scene:
var packed_scene := load("res://scene.tscn") as PackedScene
Then make an instance of it:
var scene := packed_scene.instance()
Then add it to your current scene tree:
add_child(scene)
Yes, it is a Node! It also means you can…
scene.queue_free()
Or you can simply remove it with remove_child which does not free it, so you can add it back later.
So you would be in control of what gets loaded or unloaded and when. Which is useful to keep stuff around (e.g. the UI, the player character, etc...). A drawback of doing it the way I describe here is get_tree().current_scene would not be useful to you anymore. See also Change scenes manually.

How can I edit the WorldMenu in Pharo

How can I remove or add an entry to the WorldMenu in Pharo at run time?
For example I have a menu option that loads extra tools for working with web tools. Running the code setting up these tools would include adding items to the menu to stop and start the web service. I don't want these stop and start items in normal use but the code setting up the items would be in the image.
I have seen and used the method in this question However this adds the item when the code is loaded.
Let me clarify what #Peter and I mean
Choose any class, and in the class side add your #menuCommandOn: method on the lines of
menuCommandOn: aBuilder
<worldMenu>
self showsItem ifTrue: [
(aBuilder item: self itemToken)
order: 0.1;
action: [self performItemAction]]
This way, even though the method would be invoked every time the world menu is about to pop up, it will add the menu item only if the logic behind #showsItem enables it. Notice that the dynamic nature of the menu doesn't require you to remove menu items, instead you simply do not add them. In your case such a logic should reflect the availability of the web service.
The #itemToken message send is a placeholder for the Symbol you want to use to identify the item. In other words, you would probably want to inline it as a literal rather than sending the #itemToken message. This Symbol will be used as the item label.
For further optional configuration features take a look at other implementors of #menuCommandOn:.

Adding item to all view menus

In Eclipse, it is possible to extend to any pop-up menu using the following special location URI:
popup:org.eclipse.ui.popup.any
Is it possible to do the same for the view menus (the little triangle in each view)? (I am developing an Eclipse RCP application, not extending the IDE. The purpose for this item is to help debugging and testing of the application.).
The short answer: no, that is not possible, unless you add the items by hand for all views.
The slightly longer answer: yes, if you
manually go through the view registry
create a new AbstractContributionFactory with the location URI "menu:<view-id>" and whatever createContributionItems(...) you want for each. (Alternatively you can create AbstractMenuAdditionCacheEntry or even better MenuAdditionCacheEntry - though these are internal and requires a little extra work... The last allows you to have a centrally defined menus extension point that is used for all these additions.)
add these to the workbench using IMenuService.addContributionFactory(AbstractContributionFactory factory) (use IWorkbenchWindow.getService(IMenuService.class) to acquire the menu service).
I haven't tested the above recipe, but I have used something very similar in an application...

Magento _prepareLayout() called 5 times to many

** New EDIT **
so what I'm trying to do is this.
I want the to add new form elements generated by my module on the product view of the following url
http://magento.example.com/catalog/product/view/id/46
ultimately these elements will be determined to show up by a related table in my module
I expected that if I extended Mage_Catalog_Block_Product_View in my module as shown below I would be able to create a block in the product form that would contain such form fields, only if he are in the related table in my module
so I created a test.phtml file in
app/design/frontend/default/default/templates/<module>/test.phtml
then as you can see in my the View.php file described bellow I built the block and displayed it in the product view.
It did appear but 5 times too many. from the answers below this is normal so that answers the question as to why the it shows up five times but leaves the question what is the proper way to proceecd since this plan is not going to work
** End New Edit **
in my module I call _prepareLayout() and it does this 5 times when i pull up the page
here's my code
in
/app/code/local/Namespace/Module/Product/Veiw.php
class <Namespace>_<module>_Block_Product_View extends Mage_Catalog_Block_Product_View {
protected function _toHtml() {
return parent::_toHtml();
}
public function _prepareLayout() {
$block = $this->getLayout()->createBlock(
'Mage_Core_Block_Template',
'my_block_name_here',
array('template' => '<module>/test.phtml')
);
if ($block){
$this->getLayout()->getBlock('content')->insert($block)->toHtml();
}else{
echo "no block";
}
return parent::_prepareLayout();
}
}
NOTE:
I just noticed this also takes away the price availability qty and add to cart button. which is also a problem
EDIT
First I want to thank you all for your answers. Second i want to give you more context
the reason for choosing to do this in the module is that I don't want the block to show up on every product . What i have is a table of what I'll call custom options containing properties of the product sort of like hair color height weight etc and depending on what set of properties are attached to the product (if any) will depend on what html content will show up on the page.
so in one case it my get a drop down menu and in another case it may get an input box. the other very important piece is that this must be setup so that I can give the end result out as a module that can be installed and not worrry that it won't show up if someone upgrades there magento
that said does it still make sense to do this all in the xml file ?
It seems to me that your code is overriding a core Magento module in order to achieve what could be easily done in the layout xml configuration. I would strongly recommend the follwing:
Use the built-in configuration mechanisms (e.g. layout xml - read Alan's excellent tutorial here) instead of writing code whenever possible.
Don't override the core code
if you must change the behaviour of the core code, use an Observer rather than Rewrite/Override
if you absolutely must Override, always call parent::whatever()
For example, if you create a <module>.xml layout file in your theme (app/design/frontend/default/<theme>/layout), you could use the following code:
<catalog_product_view>
<reference name="content">
<block type="module/block" name"my_block_name_here" template="module/test.phtml"/>
</reference>
</catalog_product_view>
You would then need to use a getChildHtml('my_block_name_here'); call within your phtml to position the block.
So unless there is other functionality happening inside your _prepareLayout, there's no need to override the core, or even to override the default catalog.xml.
EDIT (small edit above)
So now in your Block (I would recommend that you call it Namespace_Module_Block_Product_Customattributes or something like that), you are not overriding the core Product_View block, but merely processing your logic for what html widgets to use to render your custom attributes. Leave the rest of the tier prices, add to cart, other generic product block code, etc to Magento to work out.
If you are worried about the upgrade path for your module's users, you should definitely NOT be overriding core code. Use the configuration approach and very selectively introduce code that "plays nice" with the system rather than try to boss it around with overrides.
I took a look at a stock Magento install of CE 1.4.1, and unmodified the _prepareLayout method is called six times when loading the URL
http://magento.example.com/catalog/product/view/id/46
That's because the class is instantiated six times. So that's the correct behavior.
As for the vanishing element, I can'y say for sure, but your override to _prepareLayout doesn't appear to either
Do the same things as Mage_Catalog_Block_Product_View::_prepareLayout
Call parent::_prepareLayout();
When you override a class in a Magento you're replacing an existing class with your own. If you change a method, you're responsible for that old code being run.
It's not clear what you're trying to accomplish here. You should consider breaking your problem down into smaller problems, and then posting one (or more) "I tried X, expected Y, and got Z" type questions. As written no one's going to be able to answer your question.

Resources