* Enable whitelisting specific DNM prototypes per map
* Enable whitelisting directories
* Rename fields
* Use a HashSet instead of an array
* Add check for unused whitelist entries
* Remove whitelisting for meta (warden's rubber stamp was removed)
* Add glob support courtesy of @IProduceWidgets
* Update xmldoc
* Fix usages of TryIndex()
Most usages of TryIndex() were using it incorrectly. Checking whether prototype IDs specified in prototypes actually existed before using them. This is not appropriate as it's just hiding bugs that should be getting caught by the YAML linter and other tools. (#39115)
This then resulted in TryIndex() getting modified to log errors (94f98073b0), which is incorrect as it causes false-positive errors in proper uses of the API: external data validation. (#39098)
This commit goes through and checks every call site of TryIndex() to see whether they were correct. Most call sites were replaced with the new Resolve(), which is suitable for these "defensive programming" use cases.
Fixes#39115
Breaking change: while doing this I noticed IdCardComponent and related systems were erroneously using ProtoId<AccessLevelPrototype> for job prototypes. This has been corrected.
* fix tests
---------
Co-authored-by: slarticodefast <161409025+slarticodefast@users.noreply.github.com>
* Add altInteract option to interaction test helper methods
* Add water cooler interaction test
* Oops, that's not a NetEntity
* Is.Not.Empty
* SPlayer
* Assert.Multiple
* Document parameters
* Make yaml gas serialization cleaner
* fix exception
* fix validation code
* rudimentary test & permissive loading
* change it a bit
* Test fixes and adjustments
This PR adds delta-pressure damage. In short, airtight structures can now take damage proportional to the difference in pressures between the sides of the structure.
* Init Commit
* Typos
* Commit 2
* Save Interaction Test Mob from failing
* ssss
* Confident I've gotten all the correct prototypes
* Whoops forgot to edit those
* aaaaa
* Better solution
* Test fail fixes
* Yaml fix
* THE FINAL TEST FIX
* Final fix(?)
* whoops
* Added a WeightlessnessChangedEvent
* Check out this diff
* Wait I'm dumb
* Final optimization and don't duplicate code
* Death to IsWeightless
* File scoped namespaces
* REVIEW
* Fix test fails
* FIX TEST FAILS REAL
* A
* Commit of doom
* borgar
* We don't need to specify on map init apparently
* Fuck it
* LOAD BEARING COMMENT
---------
Co-authored-by: Princess Cheeseballs <66055347+Pronana@users.noreply.github.com>
* Improve and expand `TakeRoleAndReturn` to fail on bug #38292
* fix#38292 and expanded test cases
* use validated EntProtoIds for tests
remove unusued using declarations
* use const strings that match the TestPrototypes
* refactor: move puddle evaporation + absorbents to shared
* refactor: move SolutionRegeneration to shared
* refactor: make AbsorbentSystem visuals clientside
* style: general formatting/cleanup on touched files
- Few logical simplifications
- Add field for hard-coded sparkle effect ent
- Switch stuff to Entity<T>
No actual prediction fixes in this commit (though in
retrospect I should've done this commit last).
* fix: use predicted variants for predicted code
* fix: average out evaporation rates in mixtures
* refactor: move SolutionPurge to shared
* style: Basic SolutionPurgeComponent field cleanup
* fix: general prediction + timing + networking fixes
- Moves client side visuals back to shared because other
players exist
- Don't accumulate CurTime in Purge/RegenerationSystem
- Network the next update field in Purge/RegenerationSystem to
deal with UI mispredictions???
* fix: add udder bug workaround
Not needed for SolutionPurgeSystem which doesn't resolve
solutions (probably fine that SolutionPurgeSystem doesn't
cache since it's much rarer, though it probably should), and
likely not needed for AbsorbentSystem since it only resolves
against puddles which, I don't think can be in containers.
* fix: don't divide by zero for evaporation speed = 0.
* refactor: revert evaporation changes
Will cherry-pick these out in another PR.
Also reverting the evaporation speed bugfix since it's easier
to revert all at once. :)
* fix: component cleanup; autopause fields, use ProtoID
* fix: remove unused AbsorbentComponentState
* fix: ProtoId is not string
* refactor: move PuddleSystem.UpdateAppearance to shared
* style: general PuddleSystem.UpdateAppearance tweaks
- Switch to Entity<T>
- Use ProtoIds
- Minor simplifications
* fix: add udderly silly PVS workaround
* cleanup
* fix
---------
Co-authored-by: slarticodefast <161409025+slarticodefast@users.noreply.github.com>
* Convert string literals to protoids in Content.Tests
* Convert string literals to protoids or consts in Content.IntegrationTests
* Fix linter failures
Tricksy static using misled me
* Fix MapRenderer integration test usage to properly show output.
Added an ITestContextLike interface that can be used to properly run the integration test infrastructure OUTSIDE A TEST.
* Use System.Test.Json instead of Newtonsoft.Json for MapRenderer
* Fix map renderer JSON output being broken
I love not testing or even reading the surrounding code.
* Fix un-reusable integration instances getting leaked.
The pair state was always getting set to Ready even if the instance was killed, meaning it was getting put back into the pool even if killed.
* Mark map renderer integration instances as destructive to avoid memory leak.
* Fix file specification handling.
Map file specification is now backwards compatible again (loose filename match to search prototypes). It also supports proper direct OS filename arguments. The former is the fallback scenario is extremely important for the map server still.
Cleaned up the way that target map files are passed through the application, so mixed file/prototype specifications are now handled properly (which can be caused by the fallback behavior).
Fixes JSON data export to use the proper user-facing map name. This only works if a prototype ID is specified *or* the legacy file behavior is used.
Restructured MapPainter into an instance that has multiple functions called on it, so not all data has to be passed through a single Paint() call.
Clean up the godawful map/grid detection code. Now we just load both in a single call, because yes you can do that. This relies on LogOrphanedGrids = false in the map loader options, which I think is fine for our purposes.
Improved error handling in much of the program.
* Fix duplicate map names in map renderer output
I'm not sure *what* this output is used for, but I'm sure having it duplicated per grid isn't intentional.
* Make maprenderer command line parsing bail on unknown - options
* Fix incorrect docs for --viewer maprenderer argument
It doesn't change directory layout
* Fix parallax layer specification to not use imgur as a fucking CDN
Files are now copied to a separate folder _parallax, and these files are referenced by the parallax configuration.
Parallax data is only output when instructed to via --parallax.
This will break parallax on current map server builds, but it should be graceful. Also, that's fucking good considering we shouldn't be using imgur links. Purge it.
* Fix incorrect assert in test pair clean return
* Restore other map viewer parallax layers, fix attribution.
* This isn't a valid copyright statement but the validator forces me to enter something here.
* checkpoint
* pt 2
* pt... i forgot
* pt 4
* patch
* More test fixes
* optimization!!!
* the REAL hand system
* fix RetractableItemActionSystem.cs oversight
* the review
* test
* remove test usage of body prototype
* Update Content.IntegrationTests/Tests/Interaction/InteractionTest.cs
Co-authored-by: Tayrtahn <tayrtahn@gmail.com>
* hellcode
* hellcode 2
* Minor cleanup
* test
* Chasing the last of the bugs
* changes
---------
Co-authored-by: Tayrtahn <tayrtahn@gmail.com>
* Add interaction test for retractable arm blade
* Update for HandsSystem refactor
* Revert "Update for HandsSystem refactor"
This reverts commit e01bb9a7e318dd916240d57b30f48af9594bff1f.
* Combine WaitAssertion blocks
* Add basic test of station initial power supply
* Add info about stored power vs needed amount
* Update map list
* Get it compiling again
* Get it working again
* Only run if explicitly requested
* Fix merge
* We call 'em batteries
* Mark the test as dirty
* Add the chameleon controller implant
* address the issues (Git please dont kill me)
* Address the review and fix some merge conflicts!
* Cleanup
* Add use delay
* Silly mistakes
* Making a PR at 2 am: Gone wrong
* Predict use delay and disable the buttons until you can choose another
* First phase custom clothing
* Better system, now relays to agent id and mindshield. Chameleon loadouts are a lot better to work with as well
* Address the review! No more evil goto
* Slams way is better I should have read more closely xD
* Some of the jobs
* Add to Cargo, CentComm, Service, Passenger, Ninja, Cluwne, Wizard + Minor changes to existing; Add chameleon to bandanas, medals, jugsuits and HUDs
* Add everything else
* Fix test
* Job name
* This looks better
* Add department organization
* Minor cleanup
* Added some mindshields
* Remove redudent comment and change funcion name to be clearer
* Fix cluwne outfit
* fix merge conflicts
---------
Co-authored-by: SlamBamActionman <slambamactionman@gmail.com>
* feat: now vacuum cleaner can suck solutions from floor
* refactor using AbsorbentSystem instead of separate vacuum cleaner
* refactor: remove unused vacuum cleaner files
* refactor: renamed ConnectedContainerComponent to SlotBasedConnectedContainerComponent (and system)
* fix: fix invalid comp name
* fix: no more spray nozzle messaging about water inside bottles etc.
* refactor: minor refactor in SlotBasedConnectedContainerSystem and adjustments after merge
* refactor: cleanups
* refactor: renaming
* refactor: update to use _puddleSystem.GetAbsorbentReagents
* refactor: changed interactions with SlotBasedConnectedContainerSystem into events
* refactor: new sound and action delay adjusted to sound (amount tweaked a bit accordingly, almost)
* refactor: added networking for SlotBasedConnectedContainerComponent
* fix attribution for vacuum-cleaner-fast.ogg
* trying to fix multi-license for mix sound file
* remove empty line
* refactor: remove trailing whitespace
* by ref struct, brother
---------
Co-authored-by: pa.pecherskij <pa.pecherskij@interfax.ru>
Co-authored-by: EmoGarbage404 <retron404@gmail.com>
