SpriteTile, a fast dynamic tile system (RELEASED)

No, you're not logging anything. Log all variables and data, so you know what numbers you're actually working with.

--Eric

 

Oh I think I see what you mean. Using this I can see which coordinates the objects are being placed, but there's still something missing, and so it's not working out the way I expect. The trigger position and the object position doesn't match, even though they are place on the same spot in the editor.

Code (csharp):

1.  
2. Tile.GetTriggerPositions (6, 2, reftriggerPositions);
3. for (vari = 0; i < triggerPositions.Count; i++)
4.  {
5. //varcheckPntPos = Tile.WorldToMapPosition (transform.position, 2);
6. //Debug.Log ("Checkpoint " + checkPntPos);
7. Debug.Log ("Checkpoint Spawns at " + triggerPositions[i]);
8. Debug.Log ("With ID Spawns at " + Tile.WorldToMapPosition (checkPointSpawnPos.transform.position, 1));
9. if(Tile.WorldToMapPosition (checkPointSpawnPos.transform.position, 1) == triggerPositions[i])
10.  {
11. Debug.Log ("CheckPoint 1");
12. checkPointSpawnPos.GetComponent<CheckPoint>().checkpointID = 1;
13.  }
14. }
15.  

 

The trigger positions are from layer 2 but the WorldToMapPosition is using layer 1.

--Eric

 

Right, I did that because I have the actual object being spawned using a trigger on layer 1. I'll match all the layers and see how that goes

 

OK, so this is the entire script I'm using for placing/spawning the game objects. I hope this gives you a clearer idea of what I'm working with. Nothing seems to be working as expected.

Code (csharp):

1.  
2. usingUnityEngine;
3. usingSystem.Collections;
4. usingSystem.Collections.Generic;
5. usingSpriteTile;
6.  
7. publicclassLevelManager : MonoBehaviour
8. {
9.  
10. publicGameObjectcurrentCheckPoint;
11. publicboolcheckPointActivated;
12. publicboolswitchActivated;
13.  
14. publicTransformswitchSpawnPos;
15. publicTransformplayerSpawnPos;
16. publicTransformredHazardSpawnPos;
17. publicTransformcheckPointSpawnPos;
18.  
19. privateController2Dplayer;
20. publicTextAssettestLevel;
21. publicintactivatedSwitch;
22.  
23. Vector3playerOrigin;
24.  
25. List<Int2> triggerPositions;
26.  
27. voidAwake ()
28.  {
29. Tile.SetCamera ();
30. Tile.LoadLevel (testLevel);
31.  }
32. //Usethisforinitialization
33. voidStart ()
34.  {
35. player = FindObjectOfType<Controller2D>();
36. GameObjectPlacement();
37. switchActivated = false;
38. checkPointActivated = false;
39. //Debug.Log (currentCheckPoint);
40. playerOrigin = player.transform.position;
41. //Debug.Log (playerOrigin);
42.  }
43.  
44. voidGameObjectPlacement ()
45.  {
46. Tile.GetTriggerPositions (4, 1, reftriggerPositions);
47. for (vari = 0; i < triggerPositions.Count; i++)
48.  {
49. //player.transform.position = newVector2(triggerPositions[i].x, triggerPositions[i].y);
50. playerSpawnPos.position = Tile.MapToWorldPosition (triggerPositions[i]);
51.  }
52.  
53. Tile.GetTriggerPositions (2, 1, reftriggerPositions);
54. for (vari = 0; i < triggerPositions.Count; i++)
55.  {
56. //switchSpawnPos.position = newVector3(triggerPositions[i].x, triggerPositions[i].y, transform.position.z);
57. Instantiate (switchSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), switchSpawnPos.rotation);
58.  }
59.  
60. Tile.GetTriggerPositions (3, 1, reftriggerPositions);
61. for (vari = 0; i < triggerPositions.Count; i++)
62.  {
63. //switchSpawnPos.position = newVector3(triggerPositions[i].x, triggerPositions[i].y, transform.position.z);
64. Instantiate (redHazardSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), redHazardSpawnPos.rotation);
65.  }
66.  
67. Tile.GetTriggerPositions (1, 1, reftriggerPositions);
68. for (vari = 0; i < triggerPositions.Count; i++)
69.  {
70. //switchSpawnPos.position = newVector3(triggerPositions[i].x, triggerPositions[i].y, transform.position.z);
71. Instantiate (checkPointSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), checkPointSpawnPos.rotation);
72.  }
73.  
74. Tile.GetTriggerPositions (6, 2, reftriggerPositions);
75. for (vari = 0; i < triggerPositions.Count; i++)
76.  {
77. //varcheckPntPos = Tile.WorldToMapPosition (transform.position, 2);
78. //Debug.Log ("Checkpoint " + checkPntPos);
79. Debug.Log ("Checkpoint Spawns at " + triggerPositions[i]);
80. Debug.Log ("With ID Spawns at " + Tile.WorldToMapPosition (checkPointSpawnPos.transform.position, 2));
81. if(Tile.WorldToMapPosition (checkPointSpawnPos.transform.position, 2) == triggerPositions[i])
82.  {
83. Debug.Log ("CheckPoint 1");
84. checkPointSpawnPos.GetComponent<CheckPoint>().checkpointID = 1;
85.  }
86. //if(switchSpawnPos.position == Tile.MapToWorldPosition (triggerPositions[i]))
87. // {
88. //switchSpawnPos.GetComponent<Switch>().switchID = 1;
89. // }
90.  }
91.  
92. //Tile.GetTriggerPositions (7, 2, reftriggerPositions);
93. //for (vari = 0; i < triggerPositions.Count; i++)
94. // {
95. //if(checkPointSpawnPos.position == Tile.MapToWorldPosition (triggerPositions[i]))
96. // {
97. //checkPointSpawnPos.GetComponent<CheckPoint>().checkpointID = 2;
98. // }
99. //
100. //if(switchSpawnPos.position == Tile.MapToWorldPosition (triggerPositions[i]))
101. // {
102. //switchSpawnPos.GetComponent<Switch>().switchID = 2;
103. // }
104. // }
105.  
106. Tile.SetLayerActive (1, false);
107. Tile.SetLayerActive (2, false);
108.  
109.  
110.  
111.  }
112.  
113. publicvoidRespawnPlayer()
114.  {
115. //Debug.Log ("RespawnedPlayer");
116. if(checkPointActivated == true)
117.  {
118. player.transform.position = currentCheckPoint.transform.position;
119.  }
120. else
121.  {
122. player.transform.position = playerOrigin;
123.  }
124.  }
125. }
126.  

 

I can't see any case where WorldToMapPosition wouldn't work. Unfortunately I still can't really follow what you're trying to do, or what results you're getting vs. you expect to happen.

--Eric

 

So what I'm trying to so is spawn checkpoints and place them where l've placed triggers in layer 1. Then using a new layer (layer 2) like you suggested, place triggers where the checkpoints would be. These layer 2 triggers would determine what ID number the checkpoints would be assigned. Right now from Tile.WorldToMapPosition (checkPointSpawnPos.position, 2) is telling me that the checkpoint is at (2,8) and I don't have any objects placed there.

Does that make more sense?

 

Wouldn't you just use one loop, and assign the ID to the instantiated checkpoints right after instantiating them? I'm unclear what code like this is supposed to do:

Code (csharp):

1. Tile.GetTriggerPositions (4, 1, reftriggerPositions);
2. for (vari = 0; i < triggerPositions.Count; i++)
3.  {
4. //player.transform.position = newVector2(triggerPositions[i].x, triggerPositions[i].y);
5. playerSpawnPos.position = Tile.MapToWorldPosition (triggerPositions[i]);
6.  }

Since you're just repeatedly assigning different positions to the same object (playerSpawnPos).

--Eric

 

I'm using that piece of code to place the player at a designated starting point in the scene. I don't instantiate the player, I just have him in the hierarchy from the start. At least at this point.

 

You could use the same loop to spawn your objects and assign IDs. After you spawn first object you look at the same cell position but layer2 to to get its ID and then assign that ID to the object you just spawned (all this is at same cell position/coordinates). Then the loop continues to the second object and so on.

In other words : At layer 1 is a trigger with number that tells you code to spawn checkpoint, then at the same cell at layer 2 is its ID.

 

So this works, but I don't I implemented it properly. I don't know how to make sure that the correct IDs are being set to the right checkpoints. Right now the IDs are being set but I'm having spawning issues where the switches and checkpoints are being all jumbled up.

Code (csharp):

1.  
2. Tile.GetTriggerPositions (2, 1, reftriggerPositions);
3. for (vari = 0; i < triggerPositions.Count; i++)
4.  {
5. Instantiate (switchSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), switchSpawnPos.rotation);
6.  
7. Tile.GetTriggerPositions (6, 2, reftriggerPositions);
8. for (vary = 0; y < triggerPositions.Count; y++)
9.  {
10. Debug.Log ("Switch 1");
11. switchSpawnPos.GetComponent<Switch>().switchID = 1;
12.  }
13.  
14. Tile.GetTriggerPositions (7, 2, reftriggerPositions);
15. for (vary = 0; y < triggerPositions.Count; y++)
16.  {
17. Debug.Log ("Switch 2");
18. switchSpawnPos.GetComponent<Switch>().switchID = 19;
19.  }
20.  }
21.  
22. Tile.GetTriggerPositions (3, 1, reftriggerPositions);
23. for (vari = 0; i < triggerPositions.Count; i++)
24.  {
25. //switchSpawnPos.position = newVector3(triggerPositions[i].x, triggerPositions[i].y, transform.position.z);
26. Instantiate (redHazardSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), redHazardSpawnPos.rotation);
27.  }
28.  
29. Tile.GetTriggerPositions (1, 1, reftriggerPositions);
30. for (vari = 0; i < triggerPositions.Count; i++)
31.  {
32. Instantiate (checkPointSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), checkPointSpawnPos.rotation);
33.  
34. Tile.GetTriggerPositions (6, 2, reftriggerPositions);
35. for (vary = 0; y < triggerPositions.Count; y++)
36.  {
37. Debug.Log ("CheckPoint 1");
38. checkPointSpawnPos.GetComponent<CheckPoint>().checkpointID = 1;
39.  }
40. Tile.GetTriggerPositions (7, 2, reftriggerPositions);
41. for (vary = 0; y < triggerPositions.Count; y++)
42.  {
43. Debug.Log ("CheckPoint 2");
44. checkPointSpawnPos.GetComponent<CheckPoint>().checkpointID = 19;
45.  }
46.  }
47.  

 

Forget everything you done so far. Sorry to tell you but its non sense. You should never have almost identical code repeated over and over. If you end up with something like that you done it wrong.

Again forget everything you done so far and look at my and Eric's last posts , think it out (one single loop) and do it that way.

 

Alright, I'll scrap everything. Now, I'm pretty new to this plug in and I'm trying to figure out how it works, I was expecting a straight forward level editor, but that's clearly not the case. My issue here is that I'm not sure how to pick and chose the layers individually, within the same loop without causing conflicts.

Please give me an example of how you'd set up the code. I'm so confused.

 

Code (csharp):

1. // Get trigger positions for trigger #2 on layer 1
2. Tile.GetTriggerPositions (2, 1, ref triggerPositions);
3. // Loop through all trigger positions
4. for (var i = 0; i < triggerPositions.Count; i++) {
5.     // Spawn a check point in world space for each trigger position
6.     var spawnPos = Instantiate (checkPointSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), checkPointSpawnPos.rotation);
7.     // Get the id from the trigger at this position on layer 2
8.     var id = Tile.GetTrigger (triggerPositions[i], 2);
9.     // Assign it to the instantiated check point
10.     spawnPos.GetComponent<CheckPoint>().checkpointID = id;
11. }

--Eric

 

Hi,

Can I create the sprites in runtime from scripts/code instead of using the SpriteTile editor?

 

No, the tiles need to exist in the editor.

--Eric

 

Ok, thanks, but can I at least generate the sprites at runtime and "feed" them to the editor? I don't see a specific function to this so maybe working directly with the source is the way?

 

The sprites have to exist in the project.

--Eric

 

So I'm getting this error message, but I don't know why it would be happening. I debug logged "spawnPos" and it returned null. I'm not sure why this is happening.

NullReferenceException: Object reference not set to an instance of an object
LevelManager.GameObjectPlacement () (at Assets/Scripts/LevelManager.cs:98)
LevelManager.Start () (at Assets/Scripts/LevelManager.cs:35)

 

It means you're trying to refer to an object that doesn't exist; I can't say any more without more info. Your best bet is to use basic debugging tactics such as logging every step, or use the debugger.

--Eric

 

It's strange that I'm receiving this error message, because I have checkPointSpawnPos set to a reference of an object in the inspector. I was casting the spawnPos variable as a GameObject because if I didn't I got an error from using the line

spawnPos.GetComponent<CheckPoint>().checkpointID = id;

The error message being:

Assets/Scripts/LevelManager.cs(98,34): error CS1061: Type `UnityEngine.Object' does not contain a definition for `GetComponent' and no extension method `GetComponent' of type `UnityEngine.Object' could be found (are you missing a using directive or an assembly reference?)

So I tried commenting out that line and the objects are spawned as expected. What needs to change with that last line to make it work?

 

Use the correct type rather than Object.

--Eric

 

So I've tried

Code (csharp):

1.  
2. varspawnPos = Instantiate (switchSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), switchSpawnPos.rotation) asGameObject;
3.  

but get this error
NullReferenceException: Object reference not set to an instance of an object
LevelManager.GameObjectPlacement () (at Assets/Scripts/LevelManager.cs:98)
LevelManager.Start () (at Assets/Scripts/LevelManager.cs:35)


and

Code (csharp):

1.  
2. GameObjectspawnPos = Instantiate (switchSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), switchSpawnPos.rotation);
3.  

but get this error
Assets/Scripts/LevelManager.cs(92,36): error CS0266: Cannot implicitly convert type `UnityEngine.Object' to `UnityEngine.GameObject'. An explicit conversion exists (are you missing a cast?)

 

Assuming switchSpawnPos is a Transform, you can't cast from a Transform to a GameObject.

--Eric

 

Okay, that solved all those issues and gave me what I expected. I changed switchSpawnPos from a Transform to a GameObject type. Then cast var spawnPos as a GameObject, and that seemed to solve it.

Thanks so much, I appreciate all the help. Thanks for being patient with my lack of knowledge.

Code (csharp):

1.  
2. //Gettriggerpositionsfortrigger #2onlayer1
3. Tile.GetTriggerPositions (2, 1, reftriggerPositions);
4. //Loopthroughalltriggerpositions
5. for (vari = 0; i < triggerPositions.Count; i++)
6.  {
7. //Spawnacheckpointinworldspaceforeachtriggerposition
8. varspawnPos = Instantiate (switchSpawnPos, Tile.MapToWorldPosition (triggerPositions[i]), switchSpawnPos.transform.rotation) asGameObject;
9. Debug.Log (spawnPos);
10. //Gettheidfromthetriggeratthispositiononlayer2
11. varid = Tile.GetTrigger (triggerPositions[i], 2);
12. Debug.Log (id);
13. //Assignittotheinstantiatedcheckpoint
14. spawnPos.GetComponent<Switch>().switchID = id;
15.  }
16.  

 

A new update is out now:

SpriteTile 2.12

Changes:
• Tile.LoadMapBlock can specify which layer to use, for files that have multiple layers.

Fixes:
• Tile.CopyGroupToPosition works if no tiles have a physics collider default.
• If a group used with Tile.CopyGroupToPosition has any tiles with a tile number of 0, they're displayed correctly.
• Sprites made with individual textures rather than atlases batch properly when using Unity's sprite packer.

--Eric

 

When will it be on asset store?

 

As soon as they approve it; I don't have any control over that.

--Eric

 

Someone earlier in the thread posted about implementing Normal maps with SpriteTile sprites. Eric you mentioned using a third-party sprite-packer intead of the Unity one. Blizzy said he'd managed it. Would anyone mind elaborating on that process a little bit?

SpriteTile is absolutely awesome by the way.

[EDIT: extra detail] I would assume that Tile.SetMaterial() would come in handy here. I would set the mat to one that uses a bumped shader, but I have no idea how to programatically tell that shader what the sprite's normal map is.

 

I eventually wrote my own sprite import mechanism which packs the sprites into one atlas and also tells Unity that the atlas consists of multiple sprites. In addition to that, it generates a second atlas with an identical layout that contains the normal maps. The shader does nothing special other than providing the normals from the normal maps using the UV coordinates from the SpriteRenderer.

 

Thanks blizzy, I'll go hunting for some more info on custom importing of sprites, thanks for the answer.

 

Thanks for the awesome tool! I just have a quick question is there away I can load the tiles in the editor? I need to be able to create colliders around groups of the objects for our navigation system to see them. I see how I can have each tile have it's own collider but I really just need to group a large section of them together and add some scripts to it. It's becoming a pain to do this as a guessing game. Hopefully there is an easy way I can make this happen.

 

You can use the preview function to show tiles in the scene view. However you can't attach scripts to tiles, since they exist as just a few bytes of memory. The GameObjects that are created to display tiles don't "belong" to any particular tile, but are recycled as needed, which is what allows very large levels to be fast and use a modest amount of memory. So any objects with scripts attached will need to be separate from the tilemap (but as I mentioned you can use the preview function to make it easy to position objects relative to tiles).

--Eric

 

Awesome thanks, yeah we will be making separate game objects with the collider and scripts on them. So that's not an issue just didn't know how to preview them in the editor to make the collider. Thanks for the help!

 

Version 2.12 is on the asset store now.

--Eric

 

So I've run into a curious issue. Here's the code in question:

Code (CSharp):

1. List<Int2> triggerPositions = new List<Int2>();
2. Tile.GetTriggerPositions(x * 10 + y, ref triggerPositions);
3.                  
4. Int2 pos = triggerPositions[0];
5. Debug.Log("pos = " + pos);
6. character.transform.position = Tile.MapToWorldPosition(pos, 1);

This logs

First of all, how could GetTriggerPositions possibly return a position that's out of bounds? Secondly, that should not be out of bounds by any means; the trigger I'm looking for is at that position & the map bounds are 42x31.

 

The trigger position is from a different layer than you're using with MapToWorldPosition, so presumably the two layers have different map sizes.

--Eric

 

Oh! You're right. I didn't even know you could do that. Thanks!

 

I've just improved "out of bounds" error messages for future versions so they mention the layer you're operating on, which should help clarify things.

--Eric

 

Thanks for this, helps a ton.

Also, I found a bug (I think) when calling SetMapTileset on a layer which contains empty tiles. In my case, the two tilesets have exactly the same number of tiles, but if any empty tiles exist in the layer, the call throws an error.

 

You're right, sorry about that. If you want to fix it right now, use the source and in the MapData.cs file, in the SetMapTileset function, change these lines:

Code (csharp):

1.         for (int i = 0; i < end; i++) {
2.             int tile = map[i] & _tileShiftI;

To this:

Code (csharp):

1.         for (int i = 0; i < end; i++) {
2.             if (map[i] < 0) {
3.                 continue;
4.             }
5.             int tile = map[i] & _tileShiftI;

--Eric

 

Did this function make it into the release? I haven't had time to check yet.

 

No, sorry; it will be in the next version.

--Eric

 

Thanks!

 

Thanks for the interim fix!

 

Great asset! Saved crazy amount of work and integrates seemlessly. One thing, how do sets work? As in where are they 'stored'/referenced? I'm wanting to use SpriteTile for creating procedural dungeons, but for testing I want to use edited maps. For this I am creating a simple map and procedurally generating the tiles to display. So in the map editor I have a tileset with numbers, and place 1's for floor and 2's for walls. In the game, this map is read and another map created that replaces all the 2's with suitable wall tiles based on neighbours.

That's the plan, but at the moment it looks like I need two tilesets with identical numbers of tiles. I don't particularly want the editing tileset available in the game as I just want the map data. In theory I could use another Unity project to edit maps and export them to my game (copy level.bytes across), but that's a bit awkward.

So is it possible to load a set from a bunch of sprites when creating a new level? Where is set data kept?

 

The data is kept in SpriteTile/Resources/TileManager. Each set is a list of references to sprites that exist in the project; there aren't any functions for adding sprites at runtime. I'm not sure I'm 100% following what you're doing, but you can procedurally generate levels and then save them for re-use if you want (i.e., Tile.GetLevelBytes and then save the bytes using WriteAllBytes or similar).

--Eric

 

I'm wanting to use the level editor with a different tileset with different tile count to the one used in game.
While editing, I'll use a simplified tileset to design the level. I'll then just read the tile values and dynamically create a new level using SpriteTile and a different tileset.

 

I'm setting up my camera so that as it follows the player if it reaches the limits of the level, it should stop. I'm using Tile.GetMapSize() for this, but I'm wondering if there are other tools I can use for this?

The goal is to clamp the camera to the level bounds.

 

I have been doing something similar with my camera controller except it's driven by player input, not a character. Here is my method for moving the camera by clicking and dragging the middle mouse button, where the user will never see the map edges. You can probably edit it so that instead of a mouse input it uses the character positions.

I get the limit by looking at the camera's orthographic size. I'm using a TileSize of 1.0f so this maps to tile units 1:1. I'm currently overestimating the borders for simplicity, but it could easily be mathed up a bit to get the exact border size for your camera. Either way it will be some function of the camera's ortho size.

FYI, the variables cameraPos (Vector3), orthSize (float), borderX (float) and borderY(float) are declared as class member variables in the class that calls this function so they're not re-created every frame. Remember to do this or you'll get "variable doesn't exist" type errors.

The mapSizeX and mapSizeY could be set earlier on by Tile.GetMapSize().x and Tile.GetMapSize().y. For my game these are statics in the calling class that this method comes from (since I know my map size at the beginning).

Code (csharp):

1.  
2.  
3.      // camera controls (call this in Update)
4.     void MapScroll () {
5.  
6.         // update camera positon and ortho size
7.         cameraPos = Camera.main.transform.position;
8.         orthSize = Camera.main.orthographicSize;
9.         borderX = orthSize * 2.5f;
10.         borderY = orthSize * 1.1f;
11.  
12.         // map move (click and drag)
13.         if(Input.GetMouseButton(2))
14.         {
15.             // get input
16.             float x = Input.GetAxis("Mouse X") * -1f;
17.             float y = Input.GetAxis("Mouse Y") * -1f;
18.  
19.             // set x input to 0 if outside of x borders
20.             x = (cameraPos.x < borderX && x < 0) ? 0 : x;  // if its on the left and you're trying to move left  
21.             x = (cameraPos.x > mapSizeX - borderX && x > 0) ? 0 : x; // if its on the right and you're trying to move right
22.  
23.             // set y input to 0 if outside y borders
24.             y = (cameraPos.y < borderY && y < 0) ? 0 : y;   // if it's on the bottom and you're trying to move down
25.             y = (cameraPos.y > mapSizeY - borderY && y > 0) ? 0 : y; // if its on the top and you're trying to move up
26.  
27.             // move the camera
28.             Camera.main.transform.Translate(x,y,0);
29.         }
30.  
31.     }
32.  
33.  

[Edit: some typos in the code comments]