Bug in the init_dirs_cache was not setting direction of pipes, making all cached dirs be for south.
Because mapped in pipes didn't have piping layer set, mapped in pipes conflicted.
The full-tile check for unary machines forgot to exclude *itself*
- Moved pipe construction defines into __defines/construction.dm
- Unified pipe *unwrenching* by creating a standard proc along with the `construction_type` var.
- Eliminated the pipe fitting name & icon_state lookup tables by adding `pipe_state` var on atmos machinery and referencing that.
- Each pipe which can be made from a fitting object should override `pipe_state` with the icon state to be used on the pipe fitting object.
- Eliminated the giant switch statement of doom in pipe construction by delegating that work to `on_construction` proc.
- To make this work, every pipe must implement `get_neighbor_nodes_for_init` which returns a list of nodes which should be re-initialized on that pipe's construction.
- Combined the SCRUBBERS, SUPPLY and REGULAR pipe fitting classes together by storing the `piping_layer` variable and using the `setPipingLayer` procs
- Standardized the code for searching for node neighbors into the `can_be_node` proc.
- This proc is also improved in that is a mutual check, `check_connectable` is called on BOTH objects, so they have to mutually agree to connect as nodes. Eliminates lots of special edge case logic.
- Updated all the `amos_init` procs to use `can_be_node`. In the most common cases, even that boilerplate code is consolidated into the `STANDARD_ATMOS_CHOOSE_NODE` macro.
- Implemented `pipe_flags` which lets pipes declare (or override) certain requirements.
- Adds a "pipe_recipe" datum to help out things that construct pipes. By taking it out of the dispenser, we open the road for multiple dispenser types. No, no RPD yet. Soon.
- Enhances the pipe dispenser to operate on pipe recipe datums instead of hard coded lists of pipes it can construct. These datums are also (partially) initialized from the pipe machine types themselves, reducing having to define stuff in multiple places.
- Switched pipe dispenser UI to use browse(). Not a NanoUI, but makes it a bit prettier with low effort.
- Changed pipe dispenser to use a button selector to switch between Regular/Scrubbers/Supply instead of having separate list items.
- Added icon states to HE pipes to support the "connected on neither side" state.
Bug in the init_dirs_cache was not setting direction of pipes, making all cached dirs be for south.
Because mapped in pipes didn't have piping layer set, mapped in pipes conflicted.
The full-tile check for unary machines forgot to exclude *itself*
- Unified pipe *unwrenching* by creating a standard proc along with the `construction_type` var.
- Eliminated the pipe fitting name & icon_state lookup tables by adding `pipe_state` var on atmos machinery and referencing that.
- Each pipe which can be made from a fitting object should override `pipe_state` with the icon state to be used on the pipe fitting object.
- Eliminated the giant switch statement of doom in pipe construction by delegating that work to `on_construction` proc.
- To make this work, every pipe must implement `get_neighbor_nodes_for_init` which returns a list of nodes which should be re-initialized on that pipe's construction.
- Combined the SCRUBBERS, SUPPLY and REGULAR pipe fitting classes together by storing the `piping_layer` variable and using the `setPipingLayer` procs
- Standardized the code for searching for node neighbors into the `can_be_node` proc.
- This proc is also improved in that is a mutual check, `check_connectable` is called on BOTH objects, so they have to mutually agree to connect as nodes. Eliminates lots of special edge case logic.
- Updated all the `amos_init` procs to use `can_be_node`. In the most common cases, even that boilerplate code is consolidated into the `STANDARD_ATMOS_CHOOSE_NODE` macro.
- Implemented `pipe_flags` which lets pipes declare (or override) certain requirements.
- Adds a "pipe_recipe" datum to help out things that construct pipes. By taking it out of the dispenser, we open the road for multiple dispenser types. No, no RPD yet. Soon.
- Enhances the pipe dispenser to operate on pipe recipe datums instead of hard coded lists of pipes it can construct. These datums are also (partially) initialized from the pipe machine types themselves, reducing having to define stuff in multiple places.
- Switched pipe dispenser UI to use browse(). Not a NanoUI, but makes it a bit prettier with low effort.
- Changed pipe dispenser to use a button selector to switch between Regular/Scrubbers/Supply instead of having separate list items.
- Added icon states to HE pipes to support the "connected on neither side" state.
- Because mirroring is now *much* easier to setup than digital remote control, we reverse the order of inheritance to de-duplicate all the radio code from digital t-valves!
- This requires changing the following type paths:
/obj/machinery/atmospherics/tvalve/mirrored/digital -> /obj/machinery/atmospherics/tvalve/digital/mirrored
/obj/machinery/atmospherics/tvalve/mirrored/digital/bypass -> /obj/machinery/atmospherics/tvalve/digital/mirrored/bypass
- Change made on all maps.
- We can eliminate any overriden methods at all on flipped trinary devices by making a `mirrored` and `tee` variable! This allows us to do things like flip them without delete/spawning.
- T-Valves are also basically trinary. Sadly they are not a subtype, so to share code with them I have to put it into an outside proc. But still worth it.
- We can eliminate any overridden methods at all on flipped trinary devices by making a `mirrored` and `tee` variable! This allows us to do things like flip them without delete/spawning.
- T-Valves are also basically trinary. Sadly they are not a subtype, so to share code with them I have to put it into an outside proc. But still worth it.
- Consolidate common functions into trinary_base
- The only real difference in atmos_init was which direction each node would connect in. Delegated that to an overridable proc, and the rest can be consolidated. Same story for update_underlays()
- Used similar patterns on tvalves which also are trinary even tho they are not subtyes of trinary (for some reason)
- Because mirroring is now *much* easier to setup than digital remote control, we reverse the order of inheritance to de-duplicate all the radio code from digital t-valves!
- This requires changing the following type paths:
/obj/machinery/atmospherics/tvalve/mirrored/digital -> /obj/machinery/atmospherics/tvalve/digital/mirrored
/obj/machinery/atmospherics/tvalve/mirrored/digital/bypass -> /obj/machinery/atmospherics/tvalve/digital/mirrored/bypass
- Change made on all maps.
- We can eliminate any overriden methods at all on flipped trinary devices by making a `mirrored` and `tee` variable! This allows us to do things like flip them without delete/spawning.
- T-Valves are also basically trinary. Sadly they are not a subtype, so to share code with them I have to put it into an outside proc. But still worth it.
- We can eliminate any overridden methods at all on flipped trinary devices by making a `mirrored` and `tee` variable! This allows us to do things like flip them without delete/spawning.
- T-Valves are also basically trinary. Sadly they are not a subtype, so to share code with them I have to put it into an outside proc. But still worth it.
- Consolidate common functions into trinary_base
- The only real difference in atmos_init was which direction each node would connect in. Delegated that to an overridable proc, and the rest can be consolidated. Same story for update_underlays()
- Used similar patterns on tvalves which also are trinary even tho they are not subtyes of trinary (for some reason)
* Yes, all of them.
* Also did a few corrections to redundant New() and broken Destroy() along the way
* Renamed the turf_initializer.initialize() proc to InitializeTurf to avoid confusion.
* Subsumed /area/proc/initialize into /atom/proc/initialize() - Made /area's LateInitialize to get same behavior as before.
* Yes, all of them.
* Also did a few corrections to redundant New() and broken Destroy() along the way
* Renamed the turf_initializer.initialize() proc to InitializeTurf to avoid confusion.
* Subsumed /area/proc/initialize into /atom/proc/initialize() - Made /area's LateInitialize to get same behavior as before.
Specific situation happens when power_change() is called before atmos_init is called for the first time()
update_icons has no business modifying variables like use_power. Just update the icon!
Specific situation happens when power_change() is called before atmos_init is called for the first time()
update_icons has no business modifying variables like use_power. Just update the icon!
The initialize() routines of HE pipes, vents an scrubbers need to be atmos_init() after all.
If you register yourself with radio controller you need to de-register yourself too.
The initialize() routines of HE pipes, vents an scrubbers need to be atmos_init() after all.
If you register yourself with radio controller you need to de-register yourself too.
* to_chat() replacement.
* Revert calling target.init_dir() before connecting.
* This change was added in https://github.com/PolarisSS13/Polaris/pull/3775 to counteract `dir` not being set prior to New() for dynamically loaded maps. The root cause was /atom/New() not calling _preloader.load(). Undoing the change now that /atom/New() is fixed.
* The addition of the init_dir() proc itself however, is useful, because there ARE other times some atmos machinery will want to re-initialize its dir, specifically whenever it is rotated.
* init_dir() must be called in the constructor of all atmospherics machines capable of connecting to another. Since it has to happen for ALL machines, lets move that call to /obj/machinery/atmospherics/New()
* Rename /obj/machinery/atmospherics initialize() to atmos_init()
* These days `initialize()` is used to handle general object initialization that is moved outside of New(). The node connection discovery of atmos machinery needs to happen after all that, and so needs to be in a separate proc.
* Make sure to actually call atmos_init during system startup.
* This is PHASE 1 of a multi-phase conversion. In this first phase we implement the subsystem, but leave it processing the existing global list variables. In the next phase we will switch to use datum variables in the subsystem. The main reason for splitting into two phases is ease of code review; change the meaningful code without the hundreds of machines -> SSmachines.machinery substitutions.
* We did declare macros for adding/removing things to the processing lists, and convert everywhere to use the macros.
* Added var/is_processing to /datum to keep track of whether an instance is already in a processing list (prevents it being in the list twice!) and also debugging, making sure its not in two lists etc.
* NOTE: The global machines list is **no longer sorted** for performance reasons. As far as I know, the only module that actually ever cared was cameras. Our camera system already handles its own sorting in the cameranets anyway, so it should no longer be needed.
