====== Custom Data/Resource pack Resources ======
===== A Foreword =====
While not necessary to any extent, it is suggested that you first get accustomed to the basics of Fabric modding and Json parsing before diving into this. Going through the process of [[tutorial:recipes|adding a recipe type]] is a good way to get a handle on Json.

Furthermore, most examples in this list will be dealing with data packs. However, the process for processing resource packs is effectively the same.

===== Reload Listeners =====

Data packs and resource packs are wonderful things, they are the heart of data-driven design in the Minecraft modding scene, and as such, learning to wrangle them is an important step of one's growth as a modder.

Vanilla comes with a variety of data and resource pack resources, commonly known are things like textures, models, and after the havoc that was wrecked in 1.16.2, biomes, but this can be extended to be able to load custom data from data and resource packs. The process of extending the resource system to suit your mod's needs will be the main topic of this article, and it all starts with the use of ''ResourceManagerHelper''.

Any and all work you plan to do regarding resources must be registered through ''ResourceManagerHelper'', or more specifically, one of its resource types through ''ResourceManagerHelper#get(ResourceType)'', where the type can be either ''SERVER_DATA'' or ''CLIENT_RESOURCES'', corresponding to data pack and resource pack resources respectively, or in a dev env, the files found within the ''resources/assets'' and ''resources/data'' folders respectively (in case you are wondering, yes, your mod's assets and data are handled in the same way as resource packs and data packs internally, so this will also handle anything put in there).

From here on forward, examples will be centered around data packs and use ''ResourceType.SERVER_DATA'' were applicable, do remember that the process for data and resource pack resource processing is the same except for what ''ResourceType'' is being called.

Continuing on the registry side of things, registering your reload listener is done through ''ResourceManagerHelper#get(ResourceType).registerReloadListener(IdentifiableResourceReloadListener)''. Notably once registered your listener will not only handle the initial load of resources, but any subsequent reloads such as those done by ''/reload'' and ''F3 + T''


<code java [enable_line_numbers="true"]>
public class ExampleMod implements ModInitializer {
    @Override
    public void onInitialize() {
        ResourceManagerHelper.get(ResourceType.SERVER_DATA).registerReloadListener([...]);
    }
    
    [...]
}
</code>

===== Reload Listeners 2: The Listener =====

We have now gone over the process of registering your reload listeners, but you have yet to actually //write// one, that part of the process will be discussed in this section (specifically, this section will discuss ''SimpleSynchronousResourceReloadListener'', async listeners will be discussed later).

To simplify this tutorial, an [[https://docs.oracle.com/javase/tutorial/java/javaOO/anonymousclasses.html|anonymous class]] is used to implement ''SimpleSynchronousResourceReloadListener'', so that all the code is in one file.

<code java [enable_line_numbers="true"]>
new SimpleSynchronousResourceReloadListener() {
    @Override
    public Identifier getFabricId() {
        return new Identifier("tutorial", "my_resources");
    }

    @Override
    public void reload(ResourceManager manager) {
        [...]
    }
}
</code>

What you are seeing here is the actual resource reload listener (technically //a// resource reload listener, this is not the only type, however, it is the easiest to implement) that will be registered through ''ResourceManagerHelper''. As you can see there are two notable parts to it, the ''getFabricId'' and ''reload'' methods. The ''getFabricId'' method is the simpler of the two, you just need to return a unique identifier for the reload listener in question, nothing fancy. The reload method, however, is the main area of interest.

The ''reload'' method supplies you with a ''ResourceManager'' with which to load whatever data you want within the constraints of the registry's resource type (remember, SERVER_DATA can only access data packs, CLIENT_RESOURCES can only access resource packs). What you do with the resource manager is really mostly up to you, however, if you are making some aspect of your mod data-driven then you probably want to read from a specific //folder//. Doing this is simple but it does require a bit of care.

Firstly, you are going to want to clear or otherwise prepare to update anything that is storing the info you are going to be fetching through the manager, otherwise, your mod will likely break whenever ''/reload'' or ''F3 + T'' is used. Once that is done you should then use ''ResourceManager#findResources(String path, Predicate<String> pathPredicate)'' to fetch the files contained within whatever folder you want (note: the path is implicitly rooted in either the data or assets folder of all processed data/resource packs). This will get you a collection with the paths to all files that are within the specified path and meet the predicate's criteria (you are probably want to filter for a specific file type, ''.json'' in this tutorial), you will then need to iterate through the collection and do stuff with those paths.

<code java [enable_line_numbers="true"]>
    @Override
    public void reload(ResourceManager manager) {
        // Clear caches here

        for(Identifier id : manager.findResources("my_resource_folder", path -> path.endsWith(".json"))) {
            try(InputStream stream = manager.getResource(id).getInputStream()) {
                // Consume the stream however you want, medium, rare, or well done.
            } catch(Exception e) {
                TUTORIAL_LOG.error("Error occurred while loading resource json " + id.toString(), e);
            }
             
         }
        [...]
    }
</code>

There are two important things to note here. Firstly, all resources must be processed as ''InputStreams''. This means that you need to //close// the resources after processing. Forgetting to do this is an easy way to end up with a resource leak. The easiest way to make sure any and all processed resources get closed is to process them using a try-with-resources block as shown above. The second thing to note is that there is no way to obtain the raw file you are processing, so the "must" part of "all files must be processed as ''InputStreams''" is an actual must and not a should. Conveniently, there are many parsers that will get you your desired file from an input stream, particularly for Json.

Once you are done, your code should look somewhat like this. At this point, you can pat yourself in the back as you have successfully made a part of your mod data-driven! Any files contained in the "my_resource_folder" folder of data packs or mod data will get picked up by the this.

<code java [enable_line_numbers="true"]>
public class ExampleMod implements ModInitializer {
    @Override
    public void onInitialize() {
        ResourceManagerHelper.get(ResourceType.SERVER_DATA).registerReloadListener(new SimpleSynchronousResourceReloadListener() {
            @Override
            public Identifier getFabricId() {
                return new Identifier("tutorial", "my_resources");
            }

            @Override
            public void reload(ResourceManager manager) {
                // Clear Caches Here

                for(Identifier id : manager.findResources("my_resource_folder", path -> path.endsWith(".json"))) {
                    try(InputStream stream = manager.getResource(id).getInputStream()) {
                        // Consume the stream however you want, medium, rare, or well done.
                    } catch(Exception e) {
                        TUTORIAL_LOG.error("Error occurred while loading resource json" + id.toString(), e);
                    }
                }
            }
        });
    }
    [...]
}
</code>