Lily/ClonesPlus: re-re fix performance for "when I start as w/ set to ..." block

[SharkPool-SP]

Now it should work. It waits for variables to initialize

[CubesterYT]

😔

I will wait a bit longer, until enough time has passed

[GarboMuffin]

Fails on this project

https://static.muffin.ink/extensions-1763.sb3

production:

this pull request:

Event does not get a chance to run at all

[GarboMuffin]

Anything that isn't an edge-activated hat will differ at least a little bit. That could break projects. But maybe worth doing since this is probably rare and that it's not running right after the vanilla block is arguably a bug. (if you follow this logic expected output would be when i start as clone then when i started as clone with then when timer then forever then end).

Here is where the vanilla event gets fired. You could do it somewhere around. Might not be able to do this spot exactly since the variable won't be set yet.

https://github.com/TurboWarp/scratch-vm/blob/b900a375fa1c4a9574835c308baf5f8f16fd6cd9/src/sprites/rendered-target.js#L177

[SharkPool-SP]

Anything that isn't an edge-activated hat will differ at least a little bit. That could break projects. But maybe worth doing since this is probably rare and that it's not running right after the vanilla block is arguably a bug. (if you follow this logic expected output would be when i start as clone then when i started as clone with then when timer then forever then end).

Here is where the vanilla event gets fired. You could do it somewhere around. Might not be able to do this spot exactly since the variable won't be set yet.

https://github.com/TurboWarp/scratch-vm/blob/b900a375fa1c4a9574835c308baf5f8f16fd6cd9/src/sprites/rendered-target.js#L177

fixed

[GarboMuffin]

btw the unrelated changes like Scratch.vm.runtime -> runtime probably didn't really need to be included

[GarboMuffin]

!format

[SharkPool-SP]

btw the unrelated changes like Scratch.vm.runtime -> runtime probably didn't really need to be included

eh, ever so slightly faster + simpler

[GarboMuffin]

at the cost of making the diffs bigger for no reason and increasing change of a typo causing a regression

[GarboMuffin]

ok will use this pr to test more things related to the formatting workflow being broken

[GarboMuffin]

!format