[1.12.2][SOLVED] Json recipe enchanted book ingredient

[Ipsissimus418]

I've had a bug report raised to me where my crafting recipes using enchanted books were not matching anymore.

The issue came down to me testing with cheated in or enchantment table books and the user creating them with the anvil.

 

The anvil added an extra nbt tag "RepairCost" that caused the book to no longer match in the recipe.

 

So I'm guessing that the ingredient type "minecraft:item_nbt" has to match exactly and have no other nbt tags present.

 

Does anyone have any pointers as to how to handle this situation of extra, unwanted nbt tags being present on the item that item_nbt matches against?

I've looked at IIngredientFactory but I'm not sure that is going to help me.

 

Thanks

Ipsis

 

This is what I'm using to match against.

    {
      "type": "minecraft:item_nbt",
      "item": "minecraft:enchanted_book",
      "nbt": "{ StoredEnchantments: [ { id: 21s, lvl: 1s } ] }"
    }

 

The user showed a screenshot with the nbt as

{RepairCost:1,StoredEnchantments:[{lvl:2s,id:48s}]}

 

Edited January 1, 2018 by Ipsissimus418

[jabelar]

I'm not entirely sure but you might be able to qualify the NBT by treating the ingredient at a conditional ingredient. If you want conditional ingredients, you'll need to create an IIngredientFactory and specify it in _factories.json. In the IIngredientFactory, you can use CraftingHelper.processConditions to check if the conditions are met. If they are, return an Ingredient instance for the specified item; if they aren't, return Ingredient.EMPTY (an Ingredient that never matches any ItemStack). You can see an example implementation by Choonster here.

 

Alternatively, maybe you need to make a custom IRecipe.

Edited December 28, 2017 by jabelar

[Draco18s]

Please don't color your posts. I can't read that shit.

[jabelar]

1 minute ago, Draco18s said:

Please don't color your posts. I can't read that shit.

I didn't do that on purpose. I cut and pasted from a different forum. 

[Draco18s]

21 minutes ago, jabelar said:

I didn't do that on purpose. I cut and pasted from a different forum. 

Then click the convenient "remove formatting" button that shows up when you paste stuff.

Or the one in the tool bar.

Cyan on white is impossible to read.

[jabelar]

1 minute ago, Draco18s said:

Then click the convenient "remove formatting" button that shows up when you paste stuff.

Or the one in the tool bar.

Cyan on white is impossible to read.

I didn't know it was cyan because I use black and white on my monitor for eye health reasons. But yeah, I can try to remember to remove formatting/color.

Edited December 28, 2017 by jabelar

[Choonster]

You'll need to create a copy of IngredientNBT and override Ingredient#apply to perform a partial NBT match instead of a full NBT match.

 

You'll then need to create an implementation of IIngredientFactory that creates an instance of the Ingredient class from the provided JSON object; you can use CraftingHelper.getItemStack to parse an ItemStack from JSON.

 

Register this factory by adding the fully-qualified name to your _factories.json file; you can see an example here.

 

The name you specify for the factory in _factories.json is the name you'll use in the type property of the ingredients in your recipe files.

 

Edit: You might want to use an NBTPredicate for the NBT matching rather than storing a full ItemStack.

 

 

2 hours ago, jabelar said:

I'm not entirely sure but you might be able to qualify the NBT by treating the ingredient at a conditional ingredient. If you want conditional ingredients, you'll need to create an IIngredientFactory and specify it in _factories.json. In the IIngredientFactory, you can use CraftingHelper.processConditions to check if the conditions are met. If they are, return an Ingredient instance for the specified item; if they aren't, return Ingredient.EMPTY (an Ingredient that never matches any ItemStack). You can see an example implementation by Choonster here.

 

Alternatively, maybe you need to make a custom IRecipe.

 

This isn't really what the OP is looking for, since the conditions are only evaluated once when the recipe is loaded.

Edited December 28, 2017 by Choonster

[Ipsissimus418]

So I gave Choonster's pointers a go and I think I have something working.

 

I added a new factory with my new type that allowed me to specify an enchant id and enchant level.

The factory then parsed those json entries and applies them as the enchantment to the itemstack so it shows up in the recipe tooltips.

The factory then created a new IngredientEnchantedBook object with the itemstack, enchantment id and enchantment level.

The IngredientEnchantedBook apply method then uses EnchantmentHelper to pull the enchantments off the input stack and looks for the stored enchantment id and level in that map.

The recipes were then updated to do the following:

 

{
"type": "woot:enchanted_book",
"item": "minecraft:enchanted_book",
"enchant_id", 16,
"enchant_lvl": 1
}

 

This moves away from matching the NBT tags for now, but does mean that as long as a valid enchantment is present it will match.

I am a little concerned with this method processing all the enchantments on the itemstack to do this match, as it seems a lot of processing compared to a quicker(?) partial-nbt match.

 

Below is a link to the commit with the factory and EnchantmentHelper version for future reference.

Github commit

 

I'll investigate the NBTPredicate method next.

 

Edited December 29, 2017 by Ipsissimus418

[Choonster]

Enchantments are managed by a Forge registry, so the numeric IDs are automatically assigned per-save and shouldn't be used. Use the registry name in the JSON instead and store the Enchantment instance in the Ingredient for matching.

[Ipsissimus418]

That was me using vanilla enchantments and thinking it would always work

Thanks, I'll swap over to using the registry name instead.

Github commit

Edited January 1, 2018 by Ipsissimus418

[Ipsissimus418]

I'm going to stick with the enchantment name solution, rather than trying a NBTPredicate solution.

I'll mark this as solved and update the title to something more appropriate. Mainly so nobody thinks that the solution shows partial NBT matching.

[Juice_lmao]

So can I get a straight answer because I have no idea whts going on. Can someone maybe explain this a bit better or something?

[Animefan8888]

17 hours ago, Juice_lmao said:

So can I get a straight answer because I have no idea whts going on. Can someone maybe explain this a bit better or something?

Don't necro posts. Make your own.

[Juice_lmao]

Look Im new and i just want a straight forward answer.

[Animefan8888]

9 hours ago, Juice_lmao said:

Look Im new and i just want a straight forward answer.

Get it on your own post.