## About The Pull Request
1. Fixes the modsuit pathfinder module's pathfinding for the second
time. This time AI idling broke it, we just make it not idle.
2. Changes the heuristic used by JPS nodes from Chebyshev distance to
Euclidean distance. I have no real logical explanation, it just appeared
to produce a more optimal path. CC @LemonInTheDark
3. Renames `get_dist_euclidian()` to `get_dist_euclidean()`.
Red line: Euclidean dist JPS path (roughly)
Green line: Chebyshev dist JPS path (roughly)
## Changelog
🆑
fix: MODsuit pathfinder module works. Again.
code: AI pathfinding should produce slightly better paths.
/🆑
## About The Pull Request
The Pathfinder module sucks cock because it doesn't work. And the
reasons it doesn't work are as follows:
1. It uses the default JPS pathfinding datum, which has a hard distance
limit of 30, instead of the intended 200.
2. JPS pathfinding as a whole will fail if you encounter more than 3
doors. This is because every door wastes about 5 movement opportunities,
and the default pathfinder only has a limit of 20 before it considers
the entire pathfinding attempt moot and bails out.
Here's how I fixed it:
1. Created a new jps child that has a range of MOD_AI_RANGE
2. Instead of counting all failures during the entire pathfinding
attempt, it will only consider consecutive failures. Every successful
move will reset the pathfinding failure count. This should make JPS
pathfinding more reliable overall?
## Changelog
🆑
fix: Modsuit Pathfinder module is significantly better at finding it's
destination.
/🆑
## About The Pull Request
Basically, this means players holding cardboard cutouts will now assume
their appearance, just like for potted plants. Good for pranking.
I've had to tweak the tactical component and the waddling element a bit
to get them to work as intended while dealing with the multiple sources
of the waddling element.
## Why It's Good For The Game
## Changelog
🆑
add: Players holding cardboard cutouts will now assume their appearance,
just like for potted plants.
/🆑
## About The Pull Request
Implements /datum/pathfind/sssp, which generates /datum/path_map
/datum/path_maps allow us to very efficently generate paths to any turf
they contain from their central point.
We're effectively running the single source shortest paths algorithm.
We expand from the center turf, adding turfs as they're found, and then
processing them in order of addition.
As we go, we remember what turf "found" us first. Reversing this chain
gives us the shortest possible path from the center turf to any turf in
its range (or the inverse).
This isn't all that useful on its own, outside of a few niche cases
(Like if we wanted to get the farthest reachable turf from the center)
but if we could reuse the map more then once, we'd be able to swarm
to/from a point very easily.
Reuse is a bit troublesome, reqiures a timeout system and a way to
compare different movables trying to get paths.
I've implemented it tho. I've refactored CanAStarPass to take a datum,
/datum/can_pass_info. This is built from a movable and a list of access,
and copies all the properties that would impact pathfinding over onto
itself.
There is one case where we don't do this, pathing over openspace
requires checking if we'd fall through the openspace, and the proc for
that takes an atom.
So instead we use the weakref to the owner that we hold onto, and hold
copies of all the values that would impact the check on the datum.
When someone requests a swarmed path their pass info is compared with
the pass info of all other path_maps centered on their target turf. If
it matches and their requested timeout isn't too short, we just reuse
the map.
Timeout is a tricky thing because the longer a map exists the more out
of date it gets.
I've added a few age defines that let you modulate your level of risk
here. We default to only allowing maps that are currently
being generated, or finished generating in our tick.
Hopefully this prevents falling into trouble, but consumers will need to
allow "failed" movements.
As a part of this datumized pass info, I've refactored pathfinding to
use access lists, rather then id cards directly. This also avoids some
dumbass harddel oppertunities, and prevents an idcard from changing mid
path.
Did a few things to the zPass procs, they took args that they did NOT
need, and I thought it'd be better to yeet em.
If you'd all like I could undo the caching/can_pass_info stuff if you'd
all like. I think it's useful generally because it avoids stuff changing
mid pathfind attempt, but if it's too clunky I could nuke it.
Oh also I added optional args to jps that constricts how it handles
diagonals. I've used this to fix bot paths.
## Why It's Good For The Game
Much of this is redundant currently. I'm adding it because it could have
saved hugglebippers, and because I get the feeling it'll be useful for
"grouping" mobs like bees and such.
We're doing more basic mob work currently and I want to provide extra
tools for that work.
https://github.com/tgstation/tgstation/assets/58055496/66aca1f9-c6e7-4173-9c38-c40516d6d853
## Changelog
🆑
add: Adds swarmed pathfinding, trading accuracy for potential
optimization of used correctly
fix: Bots will no longer take diagonal paths, preventing weirdo looking
path visuals
refactor: Refactored bits of pathfinding code, hopefully easier to add
new pathfinding strategies now
/🆑
## About The Pull Request
Title.
Vendor tipping code is now on /atom/movable, and any movable can fall
over like a vendor does. Things like crits have been moved to
type-specific availability tables, their effects are now held in their
own proc, are now random per crushed item, have probability weights,
etc.
In the process of making this PR I also had to fix another issue, where
a bunch of take_damage() overrides had incorrect args, so that explains
the take_damage changes I made.
Tipping now also attacks any atoms on the target, given they use
integrity.
Adds 2 new malf modules.
1. REMOTE VENDOR TIPPING: A mid-cost and mid-supply module allows malf
AIs to remotely tip a vendor in any of the 8 directions. After 0.5
seconds of delay and a visual indicator (along with other warnings), the
vendor falls over.
1.1. In the process of making this I had to expand a arrow sprite to
have orthogonal directions, which is why you may see the testing dmi
being changed.
2. CORE ROLLING: A mid-cost but low-supply ability that allows the AI to
roll around and crush anything it falls on, including mobs. This has a
5% chance to have a critical hit so it isnt THAT terrible - plus it's
guaranteed to never stunlock. It's real utility lies in the fact the AI
now has limited movement without borgs. Also, the psychological factor.
As a bonus, vendor tipping now uses animate and transforms instead of
replacing matrices.
## Why It's Good For The Game
1. Generifying vendor tipping code is just good, period. It's a very
wacky and silly little piece of code that really doesn't need to be
isolated to vendors exclusively. ANY big and heavy object can fall over
and do a ton of damage.
1.1. Also, adding weights to critical hits is really good, because it
lets things like the headgib finally be a lot less terrifying, as
they're a lot less likely to happen.
2. Remote vendor tipping is a bit of a goofy ability that isn't really
THAT practical but has a chance of catching someone unaware and doing
some serious damage to that person alone.
2.1. Atop of this, vendor tipping isn't that loud of an action as say,
blowing things up, or doing a plasma flood. Even overrides aren't this
silent or a non-giveaway. A vendor falling on someone, though, is a
mundane thing that happens a lot. This is a decent way to assassinate
people before going loud (or at least, damage people) that isn't offered
yet.
4.
3.1. For real though, AIs rolling around is just fucking hilarious. The
ability to move isn't offered right now (which isn't that much of a bad
things), but with sufficiently limited charges (or limits to how many
times you can buy the ability), this can be a funny little t hing that
lets the AI potentially hide somewhere on the sat (or just relatively
close to the sat, such as engineering [it can't go through the
teleporter with this but it can go through transit tubes]) without the
need for borgs.
3.2. Also, it lets the AI sacrifically execute people by blowing up
their brains.
## About The Pull Request
Replaces weakref usage in AI blackboards with deleting signals
All blackboard var setting must go through setters rather than directly
## Why It's Good For The Game
This both makes it a ton easier to develop AI for, and also makes it
harder for hard deletes to sneak in, as has been seen with recent 515
prs showing hard deletes in AI blackboards
(To quantify "making it easier to develop AI", I found multiple bugs in
existing AI code due to the usage of weakrefs.)
I'm looking for `@Jacquerel` `@tralezab` 's opinions on the matter, also
maybe `@LemonInTheDark` if they're interested
## Changelog
🆑 Melbert
refactor: Mob ai refactored once again
/🆑
This tracks the seconds per tick of a subsystem, however note that it is
not completely accurate, as subsystems can be delayed, however it's
useful to have this number as a multiplier or ratio, so that if in
future someone changes the subsystem wait time code correctly adjusts
how fast it applies effects
regexes used
git grep --files-with-matches --name-only 'DT_PROB' | xargs -l sed -i
's/DT_PROB/SPT_PROB/g'
git grep --files-with-matches --name-only 'delta_time' | xargs -l sed -i
's/delta_time/seconds_per_tick/g'
## About The Pull Request
Fixes#72116
I've had a persistent issue with basic mob actions reporting this error
and think I finally cracked it
When replanning with `AI_BEHAVIOR_CAN_PLAN_DURING_EXECUTION` it can run
`Setup` on one action leading to the plan changing, meaning that it runs
`finishCommand` to cancel all other existing commands
If you triggered a replan by setting up a movement action in the middle
of another movement action, cancelling the existing action would remove
the target already set by the current one.
We want actions to be able to remove _their own_ movement target but not
if it has been changed by something else in the intervening time.
I fixed this by passing a source every time you set a movement target
and adding a proc which only clears it if you are the source... but this
feels kind of ugly. I couldn't think of anything but if you have a
better idea let me know.
Also while I was doing this I turned it into a feature because I'm
crazy.
If you feed a mouse cheese by hand it will stop being scared of humans
and so will any other mice it attracts from eating more cheese. This is
mostly because I think industrial mouse farming to pass cargo bounties
is funny.
Mice controlled by a Regal Rat lose this behaviour and forget any past
loyalties they may have had.
https://user-images.githubusercontent.com/7483112/208779368-3bd1da0f-4191-4405-86e5-b55a58c2cd00.mp4
Oh also I removed a block about cancelling if you have another target
from the "hunt" behaviour, everywhere using this already achieves that
simply by ordering the actions in expected priority order and it was
messing with how I expected mice to work.
Now if they happen to stop by some cheese they will correctly stop
fleeing in order to eat it before continuing to run away.
## Why It's Good For The Game
Fixes a bug I kept running into.
Makes it possible to set up a mouse farm without them screaming
constantly.
Lets people more easily domesticate mice to support Ratatouille
gameplay.
## Changelog
🆑
add: Mice who are fed cheese by hand will accept humans as friends, at
least until reminded otherwise by their rightful lord.
fix: Fixed a runtime preventing mice from acting correctly when trying
to flee and also eat cheese at the same time.
/🆑
## About The Pull Request
Allows basic mobs / datum AI to switch between movement datums. Useful
if you need JPS in some moments, but simple obstacle avoidance in others
This isn't used anywhere yet, but is a building block for basic bots.
## Why It's Good For The Game
Allows us to pick the right tool for the job!
## Changelog
🆑 Capybara Holly
refactor: Datum AI can now switch to different movement datums in their
behavior.
/🆑
Co-authored-by: Capybara <Capybara@CapybaraMailingServices.com>
Makes the code compatible with 515.1594+
Few simple changes and one very painful one.
Let's start with the easy:
* puts call behind `LIBCALL` define, so call_ext is properly used in 515
* Adds `NAMEOF_STATIC(_,X)` macro for nameof in static definitions since
src is now invalid there.
* Fixes tgui and devserver. From 515 onward the tmp3333{procid} cache
directory is not appened to base path in browser controls so we don't
check for it in base js and put the dev server dummy window file in
actual directory not the byond root.
* Renames the few things that had /final/ in typepath to ultimate since
final is a new keyword
And the very painful change:
`.proc/whatever` format is no longer valid, so we're replacing it with
new nameof() function. All this wrapped in three new macros.
`PROC_REF(X)`,`TYPE_PROC_REF(TYPE,X)`,`GLOBAL_PROC_REF(X)`. Global is
not actually necessary but if we get nameof that does not allow globals
it would be nice validation.
This is pretty unwieldy but there's no real alternative.
If you notice anything weird in the commits let me know because majority
was done with regex replace.
@tgstation/commit-access Since the .proc/stuff is pretty big change.
Co-authored-by: san7890 <the@san7890.com>
Co-authored-by: Mothblocks <35135081+Mothblocks@users.noreply.github.com>
About The Pull Request
Alphabetized several long lists of strings so its easier for us to look through them, just code polish, nothing the players would see.
Fixed some minor spelling errors as well.
Clarified door bolt state to be less ambiguous in the door wiring gui.
Originally it would say the door bolts have fallen, and the door bolts "Look up". i dont know about you but that was very not clear for me to read. Like where are the bolts? In the door or the frame? Arnt there bolts on top and bottom? Just didn't make sense to me.
Now it says "Have engaged!" & "Have disengaged"
hopefully that makes the state clearer at a glance.
I also added a small handful of funny texts to some string files. See changelog
Why It's Good For The Game
Well, who doesn't like a bit of polish? Just makes the game a little easier for people.
Also funny text funny text.
Changelog
spelling: improves spelling and adds more flavortext
Splitting up monkey ai into subtrees allows me to make a punpun ai after this pr is merged, makes stopping planning matter for the AI subtrees, and more generic subtrees that can be used by most ais
It also gets rid of bad practices like setting blackboards in the ai controller.
#60479 changed the attach argument "vary" into "sound_vary" but didnt change the two associative arguments for vary to sound_vary. this does that. now Attach() wont runtime for the footstep element
Converts most spans into span procs. Mostly used regex for this and sorted out any compile time errors afterwards so there could be some bugs.
Was initially going to do defines, but ninja said to make it into a proc, and if there's any overhead, they can easily be changed to defines.
Makes it easier to control the formatting and prevents typos when creating spans as it'll runtime if you misspell instead of silently failing.
Reduces the code you need to write when writing spans, as you don't need to close the span as that's automatically handled by the proc.
(Note from Lemon: This should be converted to defines once we update the minimum version to 514. Didn't do it now because byond pain and such)
About The Pull Request
Renames a directory such that it does not have a .dm prefix as that is linted against and made it onto master.
Why It's Good For The Game
The linter can stop yelling at everyone over a filepath.
These felt like another good test-case for ai controllers that someone could easily grasp, this makes it so that we can use the actual vending machine instead of a fake mimic mob. In theory you could even still use the vending machine, if you like living on the edge ofcourse.
The vendors now try to move towards you, and try to dive on top of you to squish you.
