Messages

... I'm not convinced it's much of an improvement. It might be less work for the computer, but it was harder to write, and I think harder to follow. And the only lines saved are the {}s.

TBH I'm very much surprised by such view, I have a completely opposite opinion on this.
For me it's not even the question of performance, but a question of clearly declaring a purpose. IMO the second variant has a more clear intent than the second.
For the same reason, it would be easier to write for me personally, I would just do that automatically by reflex.

But I would suggest to make a function that does "Transfer energy" from object 1 to object 2.

On a topic of ChatGPT, one big reason why I am biased against using auto-generated code is that if a person does not have enough knowledge, they would not be able to tell if the code is correct, but more importantly - if the code is suitable. Because even if the code does compile and seemingly work for the first time, that does not mean that it will work always and do precisely what you need.

I strongly advise learning what the code means if it's taken from someones examples in the internet, or generated by machine, not just put blindly into a program.

I was thinking about the easiest way to do this and came up with a somewhat weird solution:

1. put the crossed out words in the background image
2. create a room object for each hidden object as usual but compose the sprite of the object image and the non-crossed out version already in the correct position, so that when the object is placed in the room, the box with the name covers the crossed out name
3. in each object's "any click" handler, hide the object, which will also reveal the crossed out name
4. block being able to also click the words by putting a transparent but clickable GUI on top

I would advise against baking texts with the image for at least two reasons:
1. Difficult to change if you spot a mistake or want to change the text looks.
2. Difficult to translate if you add translations to your game.
(in both cases you probably would have to resort to creating dynamic sprites with texts drawn by a script command)

Furthermore, merging object's image and text on the image means that after you do you can't adjust individual object and their text position separately, you will have to remake whole merged sprite. If later in development you realize that you want to change the looks of GUI, you will have to remake ALL the combined sprites.
And, indeed, as cat said, in such case you are stuck with the fixed text's position, you won't be able to position text elsewhere at runtime.

EDIT: There are other concerns with such "merge everything together" approach:
* you won't be able to "hide" or move the GUI at runtime.
* you won't be able to have "scrolling rooms", or any camera effects at all (like zoom in/out).

So by following this "easy" solution you are going to make other things harder, if not impossible to have.

Small question on that approach, does C# use references or does it copy stuff? Like, if the Schema changed, do I need to set the schema again to CustomPropertyDescriptor or it only has a reference, so it is directly pointing to the right thing anyway?

IIRC classes are passed by references and structs by copying, but it is also possible to pass structs by reference somehow.
https://learn.microsoft.com/en-us/dotnet/csharp/language-reference/builtin-types/struct

Since Schema is a class, then it will be assigned by reference, and reference counted too.

Right, the issue is mainly the game owns the properties schema, and not the custom properties, so I don't know yet how connect these things. More an ownership problem. If the custom properties owned their schema I would have an easier time propagating them upwards. Or if the custom property itself had a "Type" property in itself - it only has value and name.

First thing that comes to mind is for CustomPropertyDescriptor class to have a static property that links to the Schema, which will be assigned by the editor on init. Similar to how editors and lists of values are assigned to some type converters and property editor attributes.

EDIT: but that's just one idea. Maybe CustomProperties class should have a link to Schema instead. Then it would be able to pass one into created Descriptor, or pass only a Type.

Since custom properties allow spaces I am once again fighting the fight for PascalCase auto-coversion....

Removing invalid characters might work. I would advise against auto-changing the letter case though, as that's a matter of a personal code style.

Ok, I checked the code now, it's amazingly little amount of code required for this to work. To put it very simply, you implement certain interface and return a dynamically collected arrays of stuff.

If you'd search for "virus" on the technical forums, you would find numerous threads that mention this problem.

https://www.adventuregamestudio.co.uk/forums/beginners-technical-questions/anitivirus-blocks-build-exe/msg636522651/#msg636522651
https://www.adventuregamestudio.co.uk/forums/beginners-technical-questions/compiled-game-being-quarantined-as-virus/msg636604522/#msg636604522
https://www.adventuregamestudio.co.uk/forums/advanced-technical-forum/virus-alert/msg636648929/#msg636648929
https://www.adventuregamestudio.co.uk/forums/beginners-technical-questions/keep-getting-this-error-and-exe-doesn-t-appear-when-compiling-game-solved/msg636579299/#msg636579299
https://www.adventuregamestudio.co.uk/forums/beginners-technical-questions/compiling-errors-due-to-virus-scanner/msg636660355/#msg636660355

Even as old as year 2009:
https://www.adventuregamestudio.co.uk/forums/advanced-technical-forum/my-creations-and-ags-exe-s-have-been-deemed-as-viruses/msg499391/#msg499391


Antiviruses do not like AGS for some reason. We had a vague idea about which program code could trigger them, but we cannot remove this code, at least not from AGS 3.*. Then again, it's not a established fact.

Disabling "Attach game data to exe" option helps sometimes, but not always.
(Actually, I recommend having it off always, because it makes loading game much faster in the presence of antivirus)

Another thing you can do is to send "false positive" report to the antivirus creators.

Sweet. It would also be grand if custom properties in AGS4 could use extender attribute APIs instead of needing Get[Text]Property(String key) and Set[Text]Property(String key, [int/String] value). Cf. Ben's latest blog post.

That's an interesting idea, I wish someone mentioned this earlier.

Too bad I'm not making games with AGS, if I had, then I'd probably get similar ideas about how to improve workflow.

In theory it's possible for the editor to generate the hidden script which declares and implements extender attributes, or even if just extender get_/set_ functions if extender attributes are not supported. Technically similar to how Global variables script is generated.

they are simply strings for now, so they are not aware of the type in the schema

I don't have spare time to look into the implementation details right now. But if you're dynamically creating a property description, it might be possible to also attach a custom TypeConverter to it, which references custom property schema item. They will still be strings behind the scenes (because custom properties are actually strings), but TypeConverter may limit values that you may assign there.

This is actually very good. I recall there have been a user request for this, but I cannot remember if we had it in github tracker.

I also made the experiment in ags3, but I guess this should actually be in ags4, where arrays are a thing, but I don't know how this would work with arrays.

I did not understand this part, could you elaborate which arrays you are refering to?

Yes, this was a quick workaround in a room where object positions are calculated according to some variables, I'm going to replace the Wait() with a call to that function now. I just wanted to stop the strange jump to new positions after the fade in quickly while I was having people test it. 

Normally anything you do in "room load" is happening prior to the first room's draw. There should not be any visible "jumps".

But I also notice that the warnings.log file is warning about things like a Wait(1); call in a room_Load() function

Well, you should not have any waits or blocking actions in room load, such things should go into "room after fade-in".

This message is mostly useless even for its purpose, because user cannot tell what the sprite exactly is. The engine reports a internal number, which is assigned automatically, and depends on the order in which the sprite was created.

If there are many of dynamic sprite in the game, they would have to experiment step by step in order to find out whether this is an actual problem, and where.

This must be replaced with something more meaningful.

Please always tell what did you try, and how that did not work. That might let us better understand your current progress and give more precise answer.

In general, the solution would depend on whether you want this "wiping" be fully random, like player using eraser on a canvas, or if you want always same fixed areas to be cleaned one by one.

In the first case you'd need to use Dynamic Sprite and Drawing Surface.
https://adventuregamestudio.github.io/ags-manual/DynamicSprite.html
https://adventuregamestudio.github.io/ags-manual/DrawingSurface.html

In the second, it's possible to do this using several separate objects (overlays or whatnot), representing each area, which are removed by selecting them.

Aha, I found the acsetup.cfg file in the Shell:SavedGames folder. But if that is the one that is used, why is there a acsetup.cfg file in the Compiled folder, do I need to distribute that file at all?

There are 3 config files read by AGS games. Both engine and winsetup programs combine them together, using override rule.
This is explained in this article in the manual:
https://adventuregamestudio.github.io/ags-manual/EngineConfigFile.html

Default Setup in the Editor defines the contents of the config file generated along with the game files.
Winsetup writes the personal user config file.

I compile my program and run the exe file, it runs in a window. I run the Winsetup.exe program, it has the "Fullscreen as borderless window" checked, I click the Save and Run button and my game still opens up in Windowed mode instead of in full-screen.

You may be confusing 2 options: "Start in windowed mode" and "Fullscreen as borderless window".
The second option defines what happens when you run the game in fullscreen: either it is a "real" fullscreen (aka exclusive fullscreen mode) or a borderless window that covers whole desktop.

- Support reordering folders in Project Explorer with drag & drop.

Does this feature include reordering folders in the Sprites tab?

I just installed the latest beta, and I'm not able to reorder my sprites folders.

I wrote a variant that supports dragging Sprite folders, copying the Project Tree functionality;
if you like you may download this temporary build and test it out:
https://cirrus-ci.com/task/5728199629340672

Sorry, but how do I download and install this? The download link took to me a page which contained a bunch of code which I assume i the code for the theme. I copied and saved that into a notepad, but when I open the folder for themes in the AGS Preferences tab, it just opens that notepad file, it doesn't change the theme.

1. Copy the theme file into the Themes folder.
2. Restart the editor, the new theme should be in the "Color Theme" selection.

1. AGS doesn't auto update. Yes, you go to the forum topic with latest release and get it from there.

Worth mentioning, there's a webpage for the latest downloads:
https://www.adventuregamestudio.co.uk/site/ags/

- Is there a harm to updating to the new version? Could it corrupt my project file in some way?

AGS is made to run the project upgrade process, where it may adjust the game settings if necessary.
If it's a patch, or next minor version (3.6.0 -> 3.6.1), then there should not be any harm.
If the versions are too far apart, it may be necessary to go in steps (2.6 -> 2.72 -> 3.6.1 -> 4.0)

In any case, the safest approach is to make backups before upgrading.

If the new version changes certain behavior, or deprecates a script function, that usually is controlled by a switch in General Settings. Backwards Compatibility section is the first to check out, but some options may be in other categories.
https://adventuregamestudio.github.io/ags-manual/GeneralSettings.html#backwards-compatibility

All the changes are explicitly mentioned in the changelog, so that's the first thing to check out if you experience a difference in behavior.
There is also a dedicated section in the manual that explains major changes between versions:
https://adventuregamestudio.github.io/ags-manual/UpgradingFromPreviousVersion.html

Understood, thanks!  In anticipation of 4.0 (which will remove GetGlobalInt, SetGlobalInt, and all deprecated functionality), I've switched over to my version rather than relying on deprecated functions.

Is there a reason why are you not using normal global variables instead?

Thanks, CW!  I'd been using the built-in, long-deprecated GetGlobalInt and SetGlobalInt in 3.6.1.  When I upgraded to 3.6.2, I got a compilation error of "undefined symbol 'GetGlobalInt'", so I'd assumed the functions had finally been removed and created my own version with the same name.

No, they were not removed, they were hidden under the script API setting, as they should have long time ago.

If you are actually using these in your games, they may be enabled again, if you wish, like any of the deprecated functions:
https://adventuregamestudio.github.io/ags-manual/GeneralSettings.html#backwards-compatibility
These two in particular require "Enforce post-2.62 scripting" set to false.

Alternatively, you may explicitly declare these functions in any of your headers. This will allow scripts to use them, since they are still available in the engine.

Since AGS 3.6.0 there's a special property for this called Game.BlockingWaitSkipped:

- Added Game.BlockingWaitSkipped which tells the result of the last blocking wait skipping,
   also including result of the blocking speech skipping.

https://adventuregamestudio.github.io/ags-manual/Game.html#gameblockingwaitskipped

Unfortunately the manual does not mention the meaning of return values.
IIRC they are matching the new return values of Wait():
https://adventuregamestudio.github.io/ags-manual/Globalfunctions_Wait.html#global-functions-wait

0 - ended by timeout or SkipWait();
 != 0 - an integer value which combines InputType, KeyMod, and a key or button code corresponding to that input type; that describes what input device and which button on that device were used to skip this "wait".

On a side note, there are no race conditions in AGS script, because whole game logic is performed on a single thread. Only one script may be run at a time. "repeatedly_execute_always" is run only when there's "waiting" in another script (either call to Wait, or a blocking action). But it's never called in the middle of executing that script itself.