xmod problem - Spawner Agitator not working

[xenoflot]

For feature requests, just erase this template and clearly describe the feature you'd like to see

Minecraft Version

1.12.2

Forge Version

1.12.2-14.23.5.2768

Mod Version

1.12.2-0.8.3-299

Describe your problem, including steps to reproduce it

Spawner Agitators aren't working in my SSP. Seems to be a cross-mod conflict as I did a test with just JEI and PNC-Re and they're working fine there. Not too many mods to check, I hope. XU2's Resturbed Spawner is the only similar item in my mod set. Details in the linked gist. https://gist.github.com/xenoflot/1f080198c19998ec0e4422ceacebc00e

Any other comments?

If you have any debug configs I can add that might help track it down, please let me know.

[desht]

When you say "not working", do you mean they just don't do anything? No stacktraces or other messages in the log?

[xenoflot]

Sorry, I should've been clearer. They appear to apply correctly; the frame appears around the spawner, and shift-right-clicking removed them again. However, there appears to be no changes to the spawning behaviour. That is, if I'm not within 16 blocks, no spawning occurred. I didn't notice anything obvious from a quick scan of the log. I've linked the log from today's session which includes my testing.

[desht]

Could prove tricky to track down, but I'll go through your mod list and see if anything stands out. Resturbed spawners are a separate block, so shouldn't break anything, but you never know...

[xenoflot]

Found it. Foamfix is the culprit.

[desht]

Ah, interesting, and thanks for tracking it down! Almost certainly related to this log entry:

[12:18:51] [main/INFO] [STDOUT]: [pl.asie.foamfix.coremod.FoamFixTransformer:spliceClasses:125]: Spliced in METHOD: net.minecraft.tileentity.MobSpawnerBaseLogic.func_98279_f

As for whether there's anything I can do about it... probably not. But, looks like there's a foamfix config setting that might help:

    # The amount of ticks per player presence check for mob spawners. Set to 1 to disable the patch and match vanilla behaviour. [default: 10, range: 1-200]
    I:mobSpawnerCheckSpeed=10

Worth trying out.

[xenoflot]

No worries. I use MultiMC and have a test instance set up. It doesn't take long to track down conflicts using a "remove half and see if that fixed it, rinse, repeat" technique :)

Just tested that setting and it does fix the problem. Perhaps you could have a chat with Asie and see if he can offer any insight. https://github.com/asiekierka/FoamFix/issues

Lemme know if you put in a fix. I'll test with the new version.

[desht]

Yeah. It's almost certainly because Foamfix injects its own optimised TileEntityMobSpawner#isActivated() method, and PNC's SemiBlockSpawnerAgitator#update() method calls the vanilla version of that via reflection (it's not a public method by default). I suspect there isn't much to be done here, but I'll mention it in the Foamfix issues anyway.

[desht]

Not much else I can do here - Foamfix issue created & there's a workaround (foamfix settings to disable mob spawner checking). Closing this issue.

[asiekierka]

Yes. The issue is that FoamFix's isActivated() will only return true the first time in a given tick - so if you call updateSpawner(), that will fail to execute. no that's wrong I implemented it right why does PCR not work...

[asiekierka]

Ah, I see.

                //Only tick the logic if it wasn't ticked already by the TE itself, to prevent double ticking.
if(!Reflections.isActivated(spawnerLogic)){

isActivated only tells you if the spawner is within range for activation, which FoamFix caches as the lookup call involved is fairly expensive. As such, adding a fake player would be completely ineffective, as the check had already executed.

[asiekierka]

Here's my proposal: https://git.asie.pl/asie-minecraft/FoamFix/src/branch/0.9.x-1.12.2/src/main/java/pl/asie/foamfix/api/IFoamFixMobSpawnerLogic.java

This should also let you get rid of the FakePlayer, if FoamFix is detected. ;-)

[desht]

Yeah, I think that would do the trick, thanks. I'll try & get round to testing that out soon, will let you know.

[asiekierka]

I've also found a way to fix your particular case of the bug regardless - I just cache the size of playerEntities in addition to the world time, now.