[1.19.x] Enable placing fluids into waterlogged blocks with FluidUtil::tryPlaceFluid #9519
Conversation
…eFluid. Enhance consistency to 1.19.3 fix MC-127110 (You can't empty water buckets into waterlogged blocks)
autoforge
bot
added
Triage
ChampionAsh5357
added
Feature
cech12
changed the title
| //If we are executing try to actually fill the container, if it failed return that we failed | ||
| if (action.simulate() || liquidContainer.placeLiquid(world, blockPos, state, resource.getFluid().getFluidType().getStateForPlacement(world, blockPos, resource))) | ||
| if (action.execute()) | ||
| { | ||
| return FluidType.BUCKET_VOLUME; | ||
| liquidContainer.placeLiquid(world, blockPos, state, resource.getFluid().getFluidType().getStateForPlacement(world, blockPos, resource)); | ||
| } | ||
| return FluidType.BUCKET_VOLUME; |
There was a problem hiding this comment.
Why did this need to be changed? I don't understand what this is doing differently other than saying that the value should always be the bucket volume when it asks whether it could place the liquid, which doesn't make sense if it can't actually be placed.
There was a problem hiding this comment.
The liquidContainer is an implementation of the vanilla interface LiquidBlockContainer, which is implemeted by waterloggable blocks. If such a block is waterlogged the placeLiquid method returns false, because no liquid is added to this block. But when you look into the BucketItem implementation, how LiquidBlockContainer is handled there (since the bugfix MC-127110), the result of the placeLiquid method is ignored:
} else if (block instanceof LiquidBlockContainer && ((LiquidBlockContainer)block).canPlaceLiquid(p_150717_,p_150718_,blockstate,content)) {
((LiquidBlockContainer)block).placeLiquid(p_150717_, p_150718_, blockstate, ((FlowingFluid)this.content).getSource(false));
this.playEmptySound(p_150716_, p_150717_, p_150718_);
return true;
} else {
There was a problem hiding this comment.
I think I understand the reasoning now because of a quirk of vanilla. I'll double check the solution and test when I awake in half a day since I currently don't have the headspace to properly review this. However, at a general glance, the solution seems fine.
There was a problem hiding this comment.
So, I don't have an issue with the solution. The main issue is that it is a breaking change in logic, regardless of expected behavior, so I'm not sure whether this should be merged into 1.19 or just held off until 1.20 in roughly 24-36 hours.
There was a problem hiding this comment.
Nice to hear that! :) I don't know, who can answer your question.^^ I can understand, that a breaking change shouldn't be merged into 1.19, but in my opinion it is a behaviour that should be there since vanilla 1.19.3. But how you decide is entirely up to you, :)
|
@ChampionAsh5357 |
This PR adds the possibility to place fluids into waterlogged blocks with the helper method
FluidUtil::tryPlaceFluid.It enhances the consistency to the bugfix MC-127110 (You can't empty water buckets into waterlogged blocks) of Minecraft 1.19.3.
I added a simple fluid container test mod, which adds a fluid container which uses the
FluidUtil::tryPlaceFluidandFluidUtil::tryPickUpFluidmethods to interact with the world.Right clicking a waterlogged block (waterlogged Slab or something else) will empty a filled fluid container by not placing any other water source (like the vanilla bucket).
Without the fix of the
BlockWrapperclass, the placement of the fluid into the waterlogged block fails and it tries to place the fluid at the block of the hit side. Failing on a waterlogged block is not consistent to vanilla.