## About The Pull Request
Ports https://github.com/Monkestation/Monkestation2.0/pull/5744,
alongside the metadata stuff from
https://github.com/Monkestation/Monkestation2.0/pull/11617 and
https://github.com/Monkestation/Monkestation2.0/pull/11644
> This PR adds a websocket client to the chatbox, that redirects all
incoming payloads from the server, to the specified server, which allows
for external integrations with the game. All of which is parsable JSON.
>
> Some ideas include RGB feedback, triggering an event in VTubing
software, or automating a physical device based on what's sent to the
chat. (For example, a haptic suit, and have a regex for "in the right
leg" or any body part)
The websocket can also be used to request basic "metadata" about the
current round - server name, station name, round id, map name, round
duration, and game state (all this information can be obtained via the
`status` world topic without being logged in anyways, to be clear) - if
the server has webroot set up, it will also send the webroot base url
and the links to the chat assets (i added this just so i could embed the
chat styling and icons in my chat log html)
**To be clear, the websocket is entirely client-side, if it's not
obvious.**
<details>
<summary>Example of metadata JSON</summary>
```json
{
"type": "metadata",
"payload": {
"game_version": "/tg/Station 13",
"server_name": "kbity station 13 (tg)",
"round_id": "152",
"map_name": "Minimal Runtime Station",
"round_duration": 6,
"gamestate": 3
}
}
```
</details>
<sub>if y'all make a freaky joke i will beat you with a rock, I do not
care about _that_ kind of thing in the slightest.</sub>
## Why It's Good For The Game
Lots of interesting things. I've made a doohickey that allows me to
automatically save my chat logs instead of having to manually click
"Save Chat Log" in settings repeatedly (which is also limited by the
chat scrollback limit)
"i have had a recurring thought of adding websocket support so I can
swap my spotify playlists whenever I get into combat on lavaland" -
smartkar
## Screenshots
<img width="676" height="158" alt="2026-05-27 (1779927348) ~
dreamseeker"
src="https://github.com/user-attachments/assets/28185ddd-b6b6-423e-9b46-cfaa519972ac"
/>
<img width="2298" height="608" alt="2026-05-27 (1779927386) ~
WindowsTerminal"
src="https://github.com/user-attachments/assets/1e41b783-6c43-4399-9ab5-a9c8bf67d117"
/>
<img width="1651" height="391" alt="2026-05-27 (1779927395) ~
dreamseeker"
src="https://github.com/user-attachments/assets/5e905670-edd6-4efc-8d0c-3d19e39799a6"
/>
## Changelog
🆑 Absolucy, Flleeppyy
add: Added a websocket client, you can now forward incoming chat
messages to an external program, i.e allowing you to change music
playlists based on events, or such.
/🆑
## About The Pull Request
This defaults every tgui to fancy mode and removes the preference
entirely.
## Why It's Good For The Game
If you look through the code you'll find comments like these
> //some browsers (IE8) have trouble with unsupported css3 elements that
break the panel's functionality, so we won't load those if a user is in
no frills tgui mode since that's for similar compatability support
We're far and away from IE8, therefore our need for compatibility
support, so I think it's valid to remove this so called 'no frills tgui
mode'. It's tied into the event message system with every backend
update, so there's a tad less overhead.
## Changelog
🆑
del: TGUI now defaults to fancy mode, there is only fancy mode. Welcome
to the future.
/🆑
## About The Pull Request
Fixes#90694
I believe the heart of this issue is that the audio player was trying to
log an error object, but logs only take strings. This should prevent the
blue screen and provide admin feedback
## Why It's Good For The Game
- Everyone survives your fatal attempt to play king gizzard the lizard
wizard
- Better admin feedback why you cannot play aforementioned song
## Changelog
🆑
fix: We're no strangers to bugs: Playing protected audio shouldn't crash
the server anymore!
/🆑
## About The Pull Request
What it was doing was by and large fine, HOW it was doing it SUCKED
I've cleaned it up and the nearby code some, notable hits include:
- random if check in secrets ui that was totally unused
- proc called add that actually set
- lists not defined as such
- stupid var names
- proc args which did nothing
- code which did nothing
- oververbose code
- proc/var names with no spacing at all
Note: This might have changed behavior accidentally, I've done my best
to test but we'll need to look out for issue reports in coming days.
## Why It's Good For The Game
I was working on bitflag code and saw red, now it's 2 hours later.
## Changelog
🆑
code: Brought browser code up to standard with the rest of the codebase
admin: Hey lads, I cleaned up how non TGUI windows work on the backend,
please let me know if anything is broken! PING ME MOTHERFUCKER
/🆑
In rare cases, we could lose the telemetry or ready message. We'll now
have a check to re-request once if we did not receive any telemetry.
Also the chat was able to init before the setting were loaded, even
though a rare condition, this could lead to the setting wiped.
Recommending to TM, as our chat differs heavily from the original and I
only gave it a short test.
## About The Pull Request
## Why It's Good For The Game
## Changelog
🆑
fix: TG chat can no longer initialize before the settings loading was
called
qol: chat settings can be exported and imported
/🆑
## About The Pull Request
Replaces the asset subsystem's spritesheet generator with a rust-based
implementation (https://github.com/tgstation/rust-g/pull/160).
This is a rough port of
https://github.com/BeeStation/BeeStation-Hornet/pull/10404, but it
includes fixes for some cases I didn't catch that apply on TG.
(FWIW we've been using this system on prod for over a year and
encountered no major issues.)
### TG MAINTAINER NOTE
### Batched Spritesheets
`/datum/asset/spritesheet_batched`: A version of the spritesheet system
that collects a list of `/datum/universal_icon`s and sends them off to
rustg asynchronously, and the generation also runs on another thread, so
the game doesn't block during realize_spritesheet. The rust generation
is about 10x faster when it comes to actual icon generation, but the
biggest perk of the batched spritesheets is the caching system.
This PR notably does not convert a few things to the new spritesheet
generator.
- Species and antagonist icons in the preferences view because they use
getFlatIcon ~~which can't be converted to universal icons~~.
- Yes, this is still a *massive* cost to init, unfortunately. On Bee, I
actually enabled the 'legacy' cache on prod and development, which you
can see in my PR. That's why I added the 'clear cache' verb and the
`unregister()` procs, because it can force a regeneration at runtime. I
decided not to port this, since I think it would be detrimental to the
large amount of contributors here.
- It is *technically* possible to port parts of this to the uni_icon
system by making a uni_icon version of getFlatIcon. However, some
overlays use runtime-generated icons which are ~~completely unparseable
to IconForge, since they're stored in the RSC and don't exist as files
anywhere~~. This is most noticeable with things like hair (which blend
additively with the hair mask on the server, thus making them invisible
to `get_flat_uni_icon`). It also doesn't help that species and antag
icons will still need to generate a bunch of dummies and delete them to
even verify cache validity.
- It is actually possible to write the RSC icons to the filesystem
(using fcopy) and reference them in IconForge. However, I'm going to
wait on doing this until I port my GAGS implementation because it
requires GAGS to exist on the filesystem as well.
#### Caching
IconForge generates a cache based on the set of icons used, all
transform operations applied, and the source DMIs of each icon used
within the spritesheet. It can compare the hashes and invalidate the
cache automatically if any of these change. This means we can enable
caching on development, and have absolutely no downsides, because if
anything changes, the cache invalidates itself.
The caching has a mean cost of ~5ms and saves a lot of time compared to
generating the spritesheet, even with rust's faster generation. The main
downside is that the cache still requires building the list of icons and
their transforms, then json encoding it to send to rustg.
Here's an abbreviated example of a cache JSON. All of these need to
match for the cache to be valid. `input_hash` contains the transform
definitions for all the sprites in the spritesheet, so if the input to
iconforge changes, that hash catches it. The `sizes` and `sprites` are
loaded into DM.
```json
{
"input_hash": "99f1bc67d590e000",
"dmi_hashes": {
"icons/ui/achievements/achievements.dmi": "771200c75da11c62"
},
"sizes": [
"76x76"
],
"sprites": {
"achievement-rustascend": {
"size_id": "76x76",
"position": 1
}
},
"rustg_version": "3.6.0",
"dm_version": 1
}
```
### Universal Icons
Universal icons are just a collection of DMI, Icon State, and any icon
transformation procs you apply (blends, crops, scales). They can be
convered to DM icons via `to_icon()`. I've included an implementation of
GAGS that produces universal icons, allowing GAGS items to be converted
into them. IconForge can read universal icons and add them to
spritesheets. It's basically just a wrapper that reimplements BYOND icon
procs.
### Other Stuff
Converts some uses of md5asfile within legacy spritesheets to use
rustg_hash_file instead, improving the performance of their generation.
Fixes lizard body markings not showing in previews, and re-adds eyes to
the ethereal color preview. This is a side effect of IconForge having
*much* better error handling than DM icon procs. Invalid stuff that gets
passed around will error instead of silently doing nothing.
Changes the CSS used in legacy spritesheet generation to split
`background: url(...) no-repeat` into separate props. This is necessary
for WebView2, as IE treats these properties differently - adding
`background-color` to an icon object (as seen in the R&D console) won't
work if you don't split these out.
Deletes unused spritesheets and their associated icons (condiments
spritesheet, old PDA spritesheet)
## Why It's Good For The Game
If you press "Character Setup", the 10-13sec of lag is now approximately
0.5-2 seconds.
Tracy profile showing the time spent on get_asset_datum. I pressed the
preferences button during init on both branches. Do note that this was
ran with a smart cache HIT, so no generation occurred.
Much lower worst-case for /datum/asset/New (which includes
`create_spritesheets()` and `register()`)
Here's a look at the internal costs from rustg - as you can see
`generate_spritesheet()` is very fast:
### Comparison for a single spritesheet - chat spritesheet:
**Before**
**After**
## Changelog
🆑
fix: Fixed lizard body markings and ethereal feature previews in the
preference menu missing some overlays.
refactor: Optimized spritesheet asset generation greatly using rustg
IconForge, greatly reducing post-initialization lag as well as reducing
init times and saving server computation.
config: Added 'smart' asset caching, for batched rustg IconForge
spritesheets. It is persistent and suitable for use on local, with
automatic invalidation.
add: Added admin verbs - Debug -> Clear Smart/Legacy Asset Cache for
spritesheets.
fix: Fixed R&D console icons breaking on WebView2/516
/🆑
## About The Pull Request
Was just scrolling through the Paradise github since they seem to have
more work done for 516 to see if there's anything I can port over, found
this and thought why not.
Ports parts of https://github.com/ParadiseSS13/Paradise/pull/25105
Specifically, updaing all hrefs to use the internal ``byond://``, and
adding it to grep.
## Why It's Good For The Game
More work towards 516.
## Changelog
Nothing player-facing.
## About The Pull Request
123 changed files and multiple crashes after writing broken regex, I
replaced most remains of direct spans with macros. This cleans up the
code and makes it easier to work with in general, see justification for
the original PR. I also fixed a bunch of broken and/or unclosed spans
here too.
I intentionally avoided replacing spans with multiple classes (in most
cases) and spans in the middle of strings as it would impact readability
(in my opinion at least) and could be done later if required.
## Why It's Good For The Game
Cleaner code, actually using our macros, fixes borked HTML in some
places. See original PR.
## Changelog
Nothing player-facing
## About Why It's Good For The Game The Pull Request
What it says on the tin. A requested feature for people who like to keep
an eye on swaths of new players. A config that is off by default, but a
Discord role ID can be added to have TGS ping that role.
Also sends alerts to Discord if a player with sketchy telemetry joins.
## Changelog
🆑 Tattle
admin: Pings an admin role on discord when a new player joins (when
enabled by config)
/🆑
---------
Co-authored-by: tattle <article.disaster@gmail.com>
# MAINTAINER - USE THE BUTTON THAT SAYS "MERGE MASTER" THEN SET THE PR
TO AUTO-MERGE! IT'S MUCH EASIER FOR ME TO FIX THINGS BEFORE THEY SKEW
RATHER THAN AFTER THE FACT.
## About The Pull Request
Hey there,
This took a while to do, but here's the gist:
Python file now regexes every file in `/code` except for those that have
some valid reason to be tacking on more global defines. Some of those
reasons are simply just that I don't have the time right now (doing what
you see in this PR took a few hours) to refactor and parse what should
belong and what should be thrown out. For the time being though, this PR
will at least _halt_ people making the mistake of not `#undef`ing any
files they `#define` "locally", or within the scope of a file.
Most people forget to do this and this leads to a lot of mess later on
due to how many variables can be unmanaged on the global level. I've
made this mistake, you've made this mistake, it's a common thing. Let's
automatically check for it so it can be fixed no-stress.
Scenarios this PR corrects:
* Forgetting to undef a define but undeffing others.
* Not undeffing any defines in your file.
* Earmarking a define as a "file local" define, but not defining it.
* Having a define be a "file local" define, but having it be used
elsewhere.
* Having a "local" define not even be in the file that it only shows up
in.
* Having a completely unused define*
(* I kept some of these because they seemed important... Others were
junked.)
## Why It's Good For The Game
If you wanna use it across multiple files, no reason to not make it a
global define (maybe there's a few reasons but let's assume that this is
the 95% case).
Let me know if you don't like how I re-arranged some of the defines and
how you'd rather see it be implemented, and I'd be happy to do that.
This was mostly just "eh does it need it or not" sorta stuff.
I used a pretty cool way to detect if we should use the standardized
GitHub "error" output, you can see the results of that here
https://github.com/san7890/bruhstation/actions/runs/4549766579/jobs/8022186846#step:7:792
## Changelog
Nothing that really concerns players.
(I fixed up all this stuff using vscode, no regexes beyond what you see
in the python script. sorry downstreams)
Similar vein to #37116
This is supposed to be standard, yet here we are.
SHOULDN'T change anything, but there's likely something out there that's
bound to behave different because of it.
These were done manually, regex to find things that MIGHT need to be
corrected;
`^#define.+\+((?!\)).)*$`
`^#define.+-((?!\)).)*$`
`^#define.+\*((?!\)).)*$`
`^#define.+\/((?!\)).)*$` (yeah that's a lot of stuff.)
`^#define.+%((?!\)).)*$`
`^#define.+SECONDS((?!\)).)*$`
`^#define.+MINUTES((?!\)).)*$`
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
adds the Refresh TGUI verb to the debug category OOC category, but it isn't locked behind any permissions. i'd put it in OOC, but.. eeeeeeh? there's not real much use for this verb other for debuggers. i put it in the OOC category
Why It's Good For The Game
debug purposes mostly
not player facing lole
refactors the status panel to utilize the tgui/byond communication APIs instead of passing along href data, as well as converts the entirety of it into a datum/tgui_window
Co-authored-by: Aleksej Komarov <stylemistake@gmail.com>
About The Pull Request
This pull request improves tgui API in many ways.
Using TGUI for custom HTML popups
This standardizes and simplifies the process of HTML popup creation and DM <-> JS communication.
Makes tgui window API a perfect alternative for old-style browser panels. It will be super useful for @Iamgoofball since he wanted to make a lightweight browser element that plays background music, and this will make his life a lot easier.
It is now possible to create tgui windows with fully inlined JS and CSS, which can be used to make unkillable tgui-based UIs (can't white/blue screen due to network errors). You can split files into JS and CSS, and still serve a single HTML file using this.
Moved sendMessage function to the Byond API object, where it rightfully belongs, and now supports a shorthand form: Byond.sendMessage(type, payload). This shortens and simplifies a lot of code.
Refactored window.update to no longer be public. Now to subscribe to incoming messages, you should use new public APIs: Byond.subscribe(fn) and Byond.subscribeTo(type, fn), and TGUI internally uses these functions as well, which reduces boilerplate in index.js.
Renamed window.__windowId__ to Byond.windowId (old variable is still available for backwards compatibility).
Byond API now supports null id, e.g. Byond.winget(null, 'url'), which makes things like this possible:
// Fetch URL of a currently connected server
Byond.winget(null, 'url').then((serverUrl) => {
// Connect to this server (opens a new dreamseeker instance)
Byond.call(serverUrl);
// Close this client because new instance is connecting
Byond.command('.quit');
});
Certain polyfills are now statically compiled (commited into git) and are baked into tgui.html. The downside is that HTML went 16 kB -> 50 kB. The upside is that you can now use a relatively modern level API with full support for IE8 when writing plain old html UIs using /datum/tgui_window directly. They are committed into git, because polyfills will never need to be updated (unless of course we randomly decide to get rid of ie8.js and html5shiv.js).
Breaking Changes
No breaking changes. This should be tested for regressions. Upgrading is simple if you're on a relatively up-to-date branch - copy paste all affected tgui files and you're good.
As it is right now, we never actually clear the temporary list processing_queries
So if the subsystem is for some reason unable to complete a run, we will just whip right back around to it again
If it's been long enough, this could even cause horrific log spam. There was just now a manuel round with roughly 30k undeleted query errors. not good.
But what was actually not deleting you may ask?
Well
When you create a db request, a 5 minute timer starts. after those 5 minutes are up, the request is qdeleted by the db subsystem
This is to prevent the creation of unused requests, and to handle requests that are never cleaned up
Telemetry code was creating all of its db requests inside a for loop that could check tick, and then later
attempting to call them in series
Since requests by default sleep, this almost always lead to undeleted queries, which harddel'd given long enough periods
I've fixed this by moving the data gathering away from the query creation
Why is it good for the game
I was working on atmos code, happy, safe in my delusion, when suddenly I got a ping from tattle freaking out over 200 undeleted queries a second
This resolves that issue, so I can once again live in peace
Changelog
cl
admin: Telemetry code will spam you with undeleted query logs much less often now!
server: Improved how the db subsystem handles undeleted queries, should never have an incident like that again
/cl
About The Pull Request
Logs tgui telemetry connections into the database. Useful since they are normally capped to 5.
Does not change the fact that the "banned account in connection history" part is still based on your history at that time. I figured it could potentially be very slow to go through your entire database history.
## About The Pull Request
**Upgrades:**
- Yarn 3.0
- TypeScript 4.3
- Sass 1.37
- Required some refactoring of `/` into `math.div()` in CSS
**Dependency removals:**
- Removed ESM package, see: https://github.com/standard-things/esm/pull/902
I initially thought it was impossible to stop relying on this package, but fortunately, ES module support in Node 12+ now comes standard and I only had to convert the very few external module imports to `require()` (because Yarn PnP).
I also moved `logging.js` directly into `tgui-dev-server` package, because that's where it is used. One less internal dependency.
**Sidegrades:**
- Removed creation of a common tgui chunk, because in practice it creates unnecessary complexity (devs sometimes get a white screen due to this chunk being invalid) and doesn't really save that much data on CDN, and **definitely** doesn't make tgui load faster.
I think that is all. I tested it a bit and everything seemingly works.
* Fix finding of dreamseeker instances and not die on cache copy failures
* Upgrade dependencies, resolve webpack dependencies explicitly
* Preload tgui asset to fix reloading of a not yet sent tgui asset
* Rewrite node bootstrap and DM build task
* Redundant array accessor
* Add install-state.gz to tgui build dependencies
* Update build.js
* Use a proper dmlang vscode var
* fix HARDER
* Remove ping case since the hard method already does a better job.
* Update code/modules/tgui_panel/external.dm
Co-authored-by: Fikou <piotrbryla@onet.pl>
Co-authored-by: Aleksej Komarov <stylemistake@gmail.com>
Co-authored-by: Fikou <piotrbryla@onet.pl>
This PR changes the message pipeline a little bit to support list-based messages, which can be annotated with custom data. Function signature of to_chat was slightly changed as well:
// Plain text message
to_chat(client,
type = MESSAGE_TYPE_INFO,
text = "foo")
// HTML message
to_chat(client,
type = MESSAGE_TYPE_INFO,
html = "<span class='notice'>foo</span>")
Old to_chat format is still supported, but handle_whitespace, trailing_newline and confidential flags have no effect. confidential flag could still be revived though, if there is enough merit in it, for example to filter out confidential messages when saving a chat log.
The reason for using /list and not /datum, is because lists are plain faster, and there are minimal data transformations - these lists are fed directly to json_encode and sent to tgchat.
Plain text messages do not need to be HTML-escaped, which makes them safer and more performant than HTML messages. Plain text messages can be made interactive (or formatted with CSS) by custom-handling them in javscript based on message type and annotations.
It would be impossible to annotate every single message in the game (at the moment of writing, there are 9447 to_chat calls in the code), but it could be done selectively, for only those messages that are hard to classify by span classes (and there are still A LOT of them).
Please annotate more messages. Thank you.
Fixes#52943Fixes#52908Fixes#52816
Changelog
cl
add: tgchat: Unread message count is now smarter and won't increase on other tabs if you have already read the same message in the active tab.
add: tgchat: Admin PMs are now properly annotated and can be filtered into separate tabs.
fix: tgchat: Fix: Highlighted message overlay no longer blocks clicks. Clicking a highlighted (F) link should work as it should.
fix: tgui: Fixed NTOS bluescreen due to calling .includes() on a stylesheet href which could be null on certain browsers.
code: tgchat: Chat schema bumped to version 5. All chat-related settings were reset to avoid breakage.
/cl
Replaces goonchat with a tgui based chat panel
Fixes#52898Fixes#52663
It is as fast as goonchat was (if not faster in certain circumstances), and is very extensible. It has all the necessary code for sorting messages into categories, which means that one of the next features will be multiple tab support.
Additional features that you will get with tgchat right now:
Massively faster server-side performance compared to goonchat, especially if batching multiple messages to one client.
Message persistence across rounds and reconnects. (All messages are stored client-side in IndexedDB)
More robust scroll tracking. If you scroll up, it will not change the scroll position on new messages like goonchat did.
Multiple message combining. (Currently set to combine up to 5 messages over last 5 seconds).
If using the highlighting feature, it highlights the whole message as well as the matching word.
"Now playing" widget, with preview of the song title, a knob for adjusting the volume and a stop button.
Architecture is as following:
```
to_chat() -+
|
SSchat
(queue, batching)
|
window.send_message()
|
v
+-------------+
| tgui-panel |
|+-----------+|
|| tgchat ||
|+-----------+|
+-------------+
```
Subsystem is basically goonchat, but without all the garbage that slows the servers down (string concatenation, double urlencoding, sanitizing, etc). Now, instead of all that, it's being slowed down by json_encode in /datum/tgui_window/proc/send_message, which IMO is completely worth it, and allows sending various templates and widgets to tgchat.
/datum/tgui_window abstracts the whole window away from you, establishes a nice message-passing interface between DM and JS, with two message queues on each side, automatically loads js/css assets for you, basically does everything. You as a developer only have to worry about sending/receiving messages and write javascript.
tgui-panel is a slimmed down version of tgui, and functions as a container for various widgets, and tgchat is one of them. It of course can be expanded with more stuff.
It's also a separate entry point and a JS bundle, so it's not bloating the main tgui bundle, and is currently sitting at about 230kB.
Lucy