Closed JonnyOThan closed 7 months ago
i feel like it has something to do with this part having a subtype that effectively moves the external hatch. Freeiva seems to not realise this and my be checking the original postion of the hatch, which is now occlude by the part itself. this is also supported by the correctly positioned hatch being visible through the window of the incorrect hatch.
FreeIva mostly uses the stock code to figure out if the hatch is obstructed: https://github.com/pizzaoverhead/FreeIva/blob/5488d6e1e4fb0abfe5be67955cf16cd676a0564c/FreeIva/InternalModules/Hatches/Hatch.cs#L1015
There's an adjustment there to the hatchObstructionCheckOutwardDistance which defaults to 1.0f (and I don't see anything in stock code which would be changing it - this is intended to make it LESS likely that the hatch is obstructed). But maybe B9PS or some other piece of code is modifying that behavior to allow the EVA when stock wouldn't have.
Yeah that's why i said, probably something to do with B9PS and freeIVA not realising the hatch position has moved
If it runs the stock check for obstruction at the original position of the hatch then it should be obstructed
If it runs the stock check for obstruction at the original position of the hatch then it should be obstructed
Shouldn't be, airlockTransform
is a reference to the actual transform where the eva kerbal will be created. But if there's more than one it might be referencing the wrong one. Like maybe if B9PS switched airlocks as you say.