Allow flexible teleporter frames up to 10x10. #6938

martica · 2021-01-30T08:16:02Z

This change extends the frame detection of teleporters to allow frames up to 10x10, in all directions and orientations.

Additionally, it prevents teleporters with solid blocks inside the frame from working.

https://i.imgur.com/gkoKxQJ.jpg

The teleport block must be on a side, not a corner and all blocks inside the frame must not be solid, opaque or impaasible.

SirEndii · 2021-01-30T09:21:13Z

And does the teleportation work?

thiakil · 2021-01-30T13:26:50Z

Please switch to using our multiblock system rather than have it implement its own structure validation.

Sidenote: your usage of Optional seems quite redundant, null would probably suffice in many cases.

pupnewfster · 2021-01-30T14:54:23Z

Also by solid blocks I presume things like rails still work fine inside of them?

martica · 2021-01-30T15:12:00Z

Please switch to using our multiblock system rather than have it implement its own structure validation.

I'll check it out. I hadn't looked that far into the codebase. I was just working from what was already there.

And does the teleportation work?

Yep. For entities, mobs, players, players on horses, players in minecarts, minecarts alone. I don't think my changes affected that.

Also by solid blocks I presume things like rails still work fine inside of them?

Yes. It's possible I've made it too restrictive, the old code didn't care what was inside the frame at all. It may be better to allow users more flexibility and if there are blocks in the frame, don't worry about it. I'd still avoid allowing teleporter blocks/teleporter frame blocks inside as nested teleporters could lead to confusing behaviour.

Sidenote: your usage of Optional seems quite redundant, null would probably suffice in many cases.

Ah. My day job's code style is to generally avoid nulls.

pupnewfster · 2021-01-30T19:43:44Z

I'll check it out. I hadn't looked that far into the codebase. I was just working from what was already there.

Ya we have a pretty decent/expansive system for multiblocks now in V10 (though I am not sure if some parts of it will need to be adjusted for purposes of the teleporter as it is 2D and in theory the frames can be shared). The main reason we didn't bother switching to it before when we wrote the new system and unified the rest of our multiblock formation code is it seemed a bit overkill for such a small and rigid structure as the teleporter used to be confined to.

I'd still avoid allowing teleporter blocks/teleporter frame blocks inside as nested teleporters could lead to confusing behaviour.

That is a good point and one thing I didn't realize would be an issue with this, though I think in theory it shouldn't be that hard in the multiblock validation to blacklist frames/teleporter blocks as invalid "inner" multiblocks.

…mes. Outstanding issues: - Teleporters work now, but at least in my save, existing ones need to be completely rebuilt. - The frame blocks consistently trigger the multiblock validation, but the teleporter blocks themselves don't always trigger it... - Shared frames do not work. - Multiple teleporter blocks can be added to one frame. - I needed to change Structure.java to allow for flat structures and I'm not sure what affect that will have on other uses of the class. It seems like it will just mean that precheck in CuboidStructureValidator will get called more often.

martica · 2021-01-31T01:29:06Z

I got to the point where teleporters do something again. There are a bunch of rough edges still.

pupnewfster

I didn't review everything but left a couple comments. I also am not sure if the way this changes it breaks frames being able to be part of multiple teleporters (for example two both vertical ones that are perpendicular to each other). It may be worth making the frames structural and adjusting or just making another multiblock validator that isn't Cuboid not sure. A good part of me also hopes aidan will be able to way in here some as he knows the logic behind the multiblock system a lot better so may be able to better say when parts are breaking things entirely.

pupnewfster · 2021-01-31T01:30:14Z

src/main/java/mekanism/common/content/teleporter/TeleporterValidator.java

+    public static final VoxelCuboid MIN_BOUNDS = new VoxelCuboid(1, 1, 1);
+    public static final VoxelCuboid MAX_BOUNDS = new VoxelCuboid(17, 18, 17);


I don't think these bounds are correct at all.

Yeah. Those were just copied from elsewhere and adjusted while I tried to get it to work.

Honestly I think a new validator that isn't directly a CuboidValidator may be better/more effective as realistically we want a VoxelPlane not a VoxelCuboid and there isn't a great way to specify the min bounds for a flat thing using the cuboid system.

Yeah. I found the structure concept a bit hard to follow so I was trying to get to a working state again before I explored something more fit to purpose.

pupnewfster · 2021-01-31T01:31:56Z

src/main/java/mekanism/common/lib/multiblock/Structure.java

-                        if (minorPlane.hasFrame() && minorPlane.length() >= 2 && minorPlane.height() >= 2) {
-                            // the plane has a frame and is at least two by two
+                        if (minorPlane.hasFrame()) {
+                            // the plane has a frame


I think this change has some significant implications and in fact may entirely invalidate the point/efficiency of having two plane types (though @aidancbrady can probably say better).

Edit: Thinking about it further, this definitely probably should not be done, as a complete teleporter frame I believe should just be a "single" major plane if I remember how the planes system works (though I only have a rudimentary understanding of it).

pupnewfster · 2021-01-31T01:33:42Z

src/main/java/mekanism/common/lib/multiblock/StructureHelper.java

     *
     * @return found cuboid, or null if it doesn't exist
     */
-    public static VoxelCuboid fetchCuboid(Structure structure, VoxelCuboid minBounds, VoxelCuboid maxBounds) {
+    public static VoxelCuboid fetchCuboid(Structure structure, VoxelCuboid minBounds, VoxelCuboid maxBounds, boolean filled) {


This method should not be used for this purpose as the entire point is for if it is all sides, the other method is much better used for cases like this as it was designed for non rectangular multiblocks and multiblocks that are not "full" for each face such as the thermal evaporation tower.

pupnewfster · 2021-01-31T01:34:28Z

src/main/java/mekanism/common/registries/MekanismBlockTypes.java

@@ -317,8 +318,14 @@ private MekanismBlockTypes() {
          .withEnergyConfig(() -> TELEPORTER_USAGE, MekanismConfig.storage.teleporter)
          .withSupportedUpgrades(EnumSet.of(Upgrade.ANCHOR))
          .without(AttributeStateActive.class, AttributeStateFacing.class, AttributeParticleFX.class)
+          .with(new AttributeSecurity(), new AttributeMultiblock())


You should rebase as I made changes earlier today to make it so that these attributes only ever have one instance created to save slightly on memory.

pupnewfster · 2021-01-31T01:39:16Z

src/main/java/mekanism/common/tile/TileEntityTeleporter.java

-import java.util.Set;
-import java.util.UUID;
+
+import java.util.*;


We don't use wildcard imports

Ugh. IntelliJ kept doing that. :(

This is the code formatting stylesheet we use with IntelliJ https://github.com/mekanism/Mekanism/blob/1.16.x/docs/GoogleStyle4Indents.xml

That will certainly help.

martica · 2021-01-31T01:53:29Z

making the frames structural

Is there an example of this already? A regression in the shared teleporter frames would be ugly.

pupnewfster · 2021-01-31T01:58:56Z

Structural glass is one, take a look at TileEntityStructuralMultiblock. I am not sure if there would be side effects such as it allowing using them in other multiblocks that are not the teleporter in place of structural glass. Though I think there may be a flag for what types of multiblocks structural ones are valid in as some only support reactor glass and not structural glass. In general I believe the validator also doesn't allow structural multiblocks as the frame but as we likely will need a custom validator that ends up consisting of a structure check of a single major plane we can have that validator allow it if necessary.

…rter frames." This reverts commit c68c432

martica · 2021-01-31T05:46:38Z

I've rolled by the WIP commit. If I get time to look at it again, I'll make another attempt.

SimonLyke · 2023-05-05T12:38:11Z

any updates on this?

create-juicey-app · 2024-02-18T22:11:36Z

Welp, we didn't got what we wanted, i dont know why

Allow flexible teleporter frames up to 10x10.

bbe6a26

The teleport block must be on a side, not a corner and all blocks inside the frame must not be solid, opaque or impaasible.

pupnewfster added 1.16 enhancement labels Jan 30, 2021

pupnewfster requested changes Jan 31, 2021

View reviewed changes

Revert "Partial work on using shared multiblock validation for telepo…

3514e11

…rter frames." This reverts commit c68c432

pupnewfster mentioned this pull request Feb 17, 2021

Larger Teleporter Frames #7001

Closed

pupnewfster mentioned this pull request Feb 10, 2022

Bigger Teleporter Frames mekanism/Mekanism-Feature-Requests#49

Open

pupnewfster marked this pull request as draft May 21, 2022 16:28

thiakil closed this Sep 21, 2023

PollyThePancake mentioned this pull request Nov 10, 2024

ATM10 1.21 Mod Suggestions List AllTheMods/ATM-10#2

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Allow flexible teleporter frames up to 10x10. #6938

Allow flexible teleporter frames up to 10x10. #6938

martica commented Jan 30, 2021

SirEndii commented Jan 30, 2021

thiakil commented Jan 30, 2021

pupnewfster commented Jan 30, 2021

martica commented Jan 30, 2021 •

edited

Loading

pupnewfster commented Jan 30, 2021

martica commented Jan 31, 2021

pupnewfster left a comment

pupnewfster Jan 31, 2021

martica Jan 31, 2021

pupnewfster Jan 31, 2021

martica Jan 31, 2021

pupnewfster Jan 31, 2021 •

edited

Loading

pupnewfster Jan 31, 2021

pupnewfster Jan 31, 2021

pupnewfster Jan 31, 2021

martica Jan 31, 2021

pupnewfster Jan 31, 2021

martica Jan 31, 2021

martica commented Jan 31, 2021

pupnewfster commented Jan 31, 2021

martica commented Jan 31, 2021

SimonLyke commented May 5, 2023

create-juicey-app commented Feb 18, 2024

		public static final VoxelCuboid MIN_BOUNDS = new VoxelCuboid(1, 1, 1);
		public static final VoxelCuboid MAX_BOUNDS = new VoxelCuboid(17, 18, 17);

Allow flexible teleporter frames up to 10x10. #6938

Allow flexible teleporter frames up to 10x10. #6938

Conversation

martica commented Jan 30, 2021

SirEndii commented Jan 30, 2021

thiakil commented Jan 30, 2021

pupnewfster commented Jan 30, 2021

martica commented Jan 30, 2021 • edited Loading

pupnewfster commented Jan 30, 2021

martica commented Jan 31, 2021

pupnewfster left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

pupnewfster Jan 31, 2021 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

martica commented Jan 31, 2021

pupnewfster commented Jan 31, 2021

martica commented Jan 31, 2021

SimonLyke commented May 5, 2023

create-juicey-app commented Feb 18, 2024

martica commented Jan 30, 2021 •

edited

Loading

pupnewfster Jan 31, 2021 •

edited

Loading