* Sideports a couple of init unit tests from Neb.
* Trying to unfuck initialize logic.
* Removing del()s.
* Adjusting return values to Initialize().
* Moving some dangerous object logic from obj onto the two types that use it.
* Rolling back some init changes and commenting out initialized atom unit test.
* this comment formatting is a minor war crime
* Removed sleep() from signaler circuit Initialize().
* Additional Init fixes post-rebase.
* Uncomments subsystem test since that's passing.
Currently the tesla ball only loses energy until it runs out of orbiting mini-balls, then sticks around forever. This makes it keep losing energy until it dissipates.
* Make intercom power change stateful
* Clean up air tank processing and initialize
* Optimize flashlight processing
* Optimize suit cooler processing
* Optimize geiger counter processing
* Optimize lighting processing
* Only process rigs on a person
* One more rig cleanup
- Instead of using auto_use_power to re-tally up machinery's power usage every cycle, track the steady "static" load separately from the transient "oneoff" usage. Machines then only need to inform the area when they use oneoff power or *change* their steady usage.
- Remove auto_use_power and stop SSmachines from calling it.
- Add vars to track "static" usage for each of the three power channels to /area
- Rename the existing three vars to "oneoff" so its clear what they mean (and to catch people accidentally updating them directly)
- Update area power procs and APCs to use the new variables.
- Rename /area/proc/use_power() to use_power_oneoff() to make it clear what it is doing.
- Deprecate /obj/machinery/use_power() in favor of use_power_oneoff() but don't delete yet. Can transition gradually.
- Add logic to the update_power procs on machines to calculate the deltas and update static area power whenever their usage changes.
- Add logic to machines to update area power when they are created, destroyed, or move.
- Moved /obj/machinery procs related to area power usage into machinery_power.dm to make them easier to find.
- Added or updated comments in several places to explain what is going on and how to use it.
- Add additional can_atmos_pass value ATMOS_PASS_PROC which indicates custom behavior requiring calling the CanZASPass proc.
- The benefit being for the other three values we DON'T need to call CanZASPass at all! We already know the behavior without the overhead of a proc call.
- Obviously any atom with can_atmos_pass = ATMOS_PASS_PROC cannot now call ..() in CanZASPass() since the default behavior would be to (recursively) call CanZASPass()
- This required re-numbering the constants, so I also fixed all code that assumed particular values for the constants.
- Switched all types which overrode CanZASPass with custom logic to be can_atmos_pass = ATMOS_PASS_PROC
- Changed /turf/c_airblock() to skip calling /atom/movable/c_airblock() for the three can_atmos_pass values that don't require calling the proc.
- CanZASPass is supposed to return boolean. Nobody noticed this bug because ATMOS_PASS_YES and ATMOS_PASS_NO happen to be defined as 1 and 0. But thats not a good assumption to make, so lets fix it!
