- Updated the initialization logic to include a check for sharedDirtyBuffer alongside sharedTileRingHeader and sharedTileRingData, ensuring all necessary data is present before creating sharedTileRing views.
- Consolidated defended state logic for tiles into dedicated methods in GameImpl to improve clarity and maintainability.
- Updated CanvasTerritoryRenderer to utilize the new isDefended method for determining tile defense status.
- Removed redundant checks and streamlined the painting logic for territory tiles.
- Extend SharedTileRing to include a shared dirtyFlags buffer alongside header and data
- Pass shared dirty buffer through WorkerClient/WorkerMessages and initialize views in Worker.worker
- In SAB mode, mark tiles dirty via Atomics.compareExchange before enqueuing to ensure each tile is queued at most once until processed
- On the main thread, clear dirty flags when draining the ring and build packedTileUpdates from distinct tile refs
- Keep non-SAB behaviour unchanged while reducing ring pressure and making overflows reflect true backlog, not duplicate updates
- Share GameMapImpl tile state between worker and main via SharedArrayBuffer
- Add SAB-backed tile update ring buffer to stream tile changes instead of postMessage payloads
- Wire shared state/ring through WorkerClient, Worker.worker, GameRunner, and ClientGameRunner
- Update GameView to skip updateTile when shared state is enabled and consume tile refs from the ring
Added src/core/worker/SharedTileRing.ts, which defines a SharedArrayBuffer-backed ring buffer (SharedTileRingBuffers/SharedTileRingViews) and helpers pushTileUpdate (worker-side writer) and drainTileUpdates (main-thread reader) using Atomics.
Extended GameRunner (src/core/GameRunner.ts) with an optional tileUpdateSink?: (update: bigint) => void; when provided, tile updates are sent to the sink instead of being packed into GameUpdateViewData.packedTileUpdates (those become an empty BigUint64Array in this mode).
Extended the worker protocol (src/core/worker/WorkerMessages.ts) so the init message can optionally carry sharedTileRingHeader and sharedTileRingData (the two SABs for the ring).
Updated WorkerClient (src/core/worker/WorkerClient.ts) to accept optional SharedTileRingBuffers in its constructor and, during initialize(), include those SABs in the init message sent to Worker.worker.ts.
Updated the worker implementation (src/core/worker/Worker.worker.ts) to build sharedTileRing views from the SABs on "init" and pass a (update: bigint) => pushTileUpdate(...) sink into createGameRunner, so every tile update from the simulation goes into the shared ring instead of the message payload.
Updated the client side (src/client/ClientGameRunner.ts):
In createClientGame, when SharedArrayBuffer/Atomics are available and crossOriginIsolated is true, it allocates the ring (createSharedTileRingBuffers with capacity 262144), creates views, passes buffers into WorkerClient, and keeps the views in ClientGameRunner.
ClientGameRunner now has a tileRingViews field and, in mergeGameUpdates, if the ring is present it ignores gu.packedTileUpdates and instead drains up to MAX_TILE_UPDATES_PER_RENDER (100 000) tile updates from the SAB ring into combinedPackedTileUpdates before constructing the merged GameUpdateViewData passed to GameView.update.
If SABs are unavailable, it falls back to the original behaviour (packing gu.packedTileUpdates from messages), so the feature is opt-in and backward compatible.
Net effect: when the page is cross-origin isolated, the worker no longer sends per-tick BigUint64Array tile payloads via postMessage; it streams tile updates into a shared ring buffer, and the main thread pulls and applies them at its own pace while still batching non-tile updates and rendering exactly as before. If you’d like, I can next run the existing tests/build in your environment to sanity-check everything.
removed:
- catchUpMode and its CATCH_UP_ENTER/EXIT thresholds in ClientGameRunner
- tick metrics fields and overlay UI for inCatchUpMode and beatsPerFrame
- leftover worker heartbeat plumbing (message type + WorkerClient.sendHeartbeat) that was no longer used after self-clocking
changed:
- backlog tracking: keep serverTurnHighWater / lastProcessedTick / backlogTurns, but simplify it to just compute backlog and a backlogGrowing flag instead of driving a dedicated catch-up mode
- frame skip: adaptRenderFrequency now only increases renderEveryN when backlog > 0 and still growing; when backlog is stable/shrinking or zero, it decays renderEveryN back toward 1
- render loop: uses the backlog-aware renderEveryN unconditionally (no catch-up flag), and resets skipping completely when backlog reaches 0
- metrics/overlay: TickMetricsEvent now carries backlogTurns and renderEveryN; the performance overlay displays backlog and current “render every N frames” but no longer mentions catch-up or heartbeats
Learnings during branch development leading to this
Once the worker self-clocks, a separate “catch-up mode” and beats-per-frame knob don’t add real control; they just complicate the model.
Backlog is still a valuable signal, but it’s more effective as a quantitative input (backlog size and whether it’s growing) than as a boolean mode toggle.
Frame skipping should be driven by actual backlog pressure plus frame cost: throttle only while backlog is growing and frames are heavy, and automatically relax back to full-rate rendering once the simulation catches up.
GameRunner exposes pending work via a new hasPendingTurns() so the worker can check whether more ticks need to be processed.
Worker auto-runs ticks: as soon as it initializes or receives a new turn, it calls processPendingTurns() and loops executeNextTick() while hasPendingTurns() is true. No more "heartbeat" message type; the worker no longer depends on the main thread’s RAF loop to advance the simulation.
Client main thread simplified:
Removed CATCH_UP_HEARTBEATS_PER_FRAME, the heartbeat loop, and the lastBeatsPerFrame tracking.
keepWorkerAlive now just manages frame skipping + draining. When it decides to render (based on renderEveryN), it drains pendingUpdates, merges them, updates GameView, and runs renderer.tick().
Because rendering a batch always implies draining, we restored the invariant that every GameView.update is paired with a layer tick() (no more lost incremental updates).
MAX_RENDER_EVERY_N is now 5 to keep the queue from growing too large while the worker sprints.
## Description:
Fix for v28.
After PR #2378 changes, unit type Construction doesn't exist anymore.
Resulting in totalUnitLevels counting in construction buildings towards
total units in TeamStats, old and new Build Menu and Player Info
Overlay. To fix this we now need to filter out the construction phase.
totalUnitLevels is only used for displaying total count purposes, it
doesn't affect other logic.
Before PR 2378 changes:
https://github.com/user-attachments/assets/7748092a-8b7b-41bb-bc68-439ba922b479
After PR 2378 changes:
https://github.com/user-attachments/assets/bc77d7c2-be02-487a-b46a-c02367ae5ad8
After fix in this PR it is back to what is was before:
https://github.com/user-attachments/assets/8d7d14b6-7b6b-4ddb-96d1-17a0a45727f3
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
tryout33
When user tries to join either a public or private multiplayer game, the
turnstile callback is triggered, and the turnstile token is passed to
the server when joining a game.
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
evan
Previously, the connection and reconnection logic were identical in
Worker.ts, so clients would need to be re-authorized for cosmetics etc
even when reconnecting. Now, on reconnect, Worker.ts only does
authentication - verifying the jwt is valid.
This will allow clients to require a valid turnstile token when first
connecting, and not when reconnecting after a broken ws connection.
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
evan
## Description:
Overhaul to nuke previews:
- Nuke previews now show when they become invulnerable (if
invulnerability setting enabled).
- Nuke previews now show if a SAM defense could intercept it, and where
the intercept occurs.
- SAMs range is now color-coded and outlined to be able to know if
missile will be intercepted by it.
<img width="1326" height="862" alt="image"
src="https://github.com/user-attachments/assets/4cc43a0b-ebac-4707-85bb-4d422f23da28"
/>
Colors/outlines aren't meant to be final, and there are constants to
play around with at the start of both files.
I also kept the naive implementation since the performance cost isn't
too high from my testing and any improvements I can think of are small
and require a bigger rewrite.
## Please complete the following:
- [X] I have added screenshots for all UI updates
- [X] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [X] I have added relevant tests to the test directory
- [X] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
bibizu
## Description:
Special bot names. If the solution seems convoluted for such an easy
thing, that is because: not all bots find a spawn position, so only
assign a candidate name after finding a spawn. And the first few are
almost always overwritten by Nation spawns so the first 20 just get a
random name. Only then do we assign from the provided lists. For the
random names, some might get the same name but that's not an issue as
no-one will notice and they're off the map quite fast anyway.
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
tryout33
## Description:
Fixes for v28. In https://github.com/openfrontio/OpenFrontIO/pull/2444,
lobby team preview was added. But for players with clan tags, this
doesn't work correctly. They don't get assigned to the same team in the
preview, while they do get assigned to the same team in the actual game.
Also added some small fixes and QoL improvements like showing the number
of Nations next to the number of Players. Since we needed that info
anyway. Did not choose to show and assign Nations to teams (just the
numbers), see why under CONSIDERED OPTIONS THAT I DIDN'T WENT WITH.
**BEFORE:**
https://youtu.be/AV_aDJ4PgOk
<img width="767" height="117" alt="Malformed argument because of double
accolades for remove_player"
src="https://github.com/user-attachments/assets/7de1114e-7ce1-4a8f-95cc-6b0528a61e3b"
/>
**AFTER:**
https://youtu.be/aDCKkwedqes
Cause of bug:
maxTeamSize is a number in assignTeams, only used to assign clan
players. It uses the length of the players array as input. At actual
game start, the Nations are also in the players array. But at lobby
preview the Nations aren't yet fetched. So when 2 players want to do a 3
Teams private lobby with them using clan tags to be in the same team.
maxTeamCount would do Math.ceil(2/3)=1. So only 1 clan player per team
as a result. While actually there could be 10 Nations which would result
in maxTeamCount Math.ceil(12/3)=4 so in the game they would actually be
assigned to the same team.
Fix for bug:
fetch Nations count in HostLobbyModal and pass on to LobbyTeamView. Add
it to the number of players for maxTeamCount that
assignTeamsLobbyPreview uses for its calculation. Also added nation
count to the similar teamMaxSize in LobbyTeamView itself, to display the
same and correct number of max players. For random maps, we now need to
know the random map before the game starts to get its Nation count. So
made some changes for that too in HostLobbyModal.
Also fixed:
- willUpdate ran comptuteTeamPreview every second, now checks if
properties like 'client' actually changed. PollPlayers in HostLobbyModal
'changes' the clients property every second even if there are no actual
changes. Checks if the other properties are actually changed too, to
make it more future proof.
- cache teamsList so it is only fetched once instead of first in
computeTeamPreview and then again for showTeamColors.
- don't show the "Empty Teams" header if there are no empty teams.
- console error ICU format error SyntaxError: MALFORMED ARGUMENT.
Because of double accolades around remove_player in translation value.
- remove fallback for comparing clients on clientID, which used client
name. Players may have the same names so it's not safe to compare based
on name.
Also show number of Nations next to number of Players:
now we now the nationCount since it is needed for the fix, show number
of Nations next to number of Players. It's handy and it prevents
confusion as to why it says 2/32 for two teams if there are only 2
players; it is because there are 61 Nations as well on the World map for
example.
Also determine number of teams based on Players + Nations:
now we now the nationCount since it is needed for the fix, use it to
determine the number of teams. Just like populateTeams in GameImpl does.
This means for Duos on the World map, a minimum of 31 teams will be
shown since there are 61 Nations. This is better than just show two
teams based on 1 or 2 humans in the lobby. Also it makes more clear how
many teams there'll be the game and how the players and nations are
divided over the teams. Choose to not show the Nations' team assignments
though. That could be for another PR. See explanation under CONSIDERED
OPTIONS THAT I DIDN'T WENT WITH.
Also show Nations team as pre-filled for HumansVSNations:
now we now the nationCount since it is needed for the fix, for
HumansVSNations, show the Nations team as fully assigned and non-empty.
For example for World map it shows Nations 61/61. Don't show them as
Empty Team but as Assigned Team. Although i choose not to show the
actual Nations (see CONSIDERED OPTIONS THAT I DIDN'T WENT WITH), this
makes it clear their team is pre-filled and how many Nations you're
actually up against. Whereas for other Team game types like 7 Teams or
Duos, it will display the teams that the Nations will fill up as empty.
CONSIDERED OPTIONS THAT I DIDN'T WENT WITH
- Use an optional param 'nationsCount' to assignTeams with default = 0.
And simply add nationsCount to the players.length count for maxTeamSize.
This would be error prone; 'nationsCount' should then never be assigned
a value except when called from LobbyTeamView. But in the future someone
might assign it a value even when called from somewhere else. Then you
could say, check if there are Nations in the players array and if so, do
not use Nationscount because we know they are already counted from
players.length. But if Disable Nations is enabled, there would be no
Nations in the players array but if nationsCount was somehow given a
value we still should not use it. So again, too developer error prone.
- Not only fetch the number of Nations, but actually get the Nations
themselves to show them in the team assign preview as well. They are
shuffled on team assignment but of course deterministicly (nation player
ID assigned based on Pseudorandom seeded with GameID in GameRunner, then
shuffled in TeamAssignment with Pseudorandom seeded with map's first
Nation's playerID), so we could replicate it. But then, how to
distinguish humans and Nations in the UI? This feels like something for
a follow-up PR.
FOR A FUTURE PR
- change the way Clan team overflow is handled. Now they're "kicked" as
in not assigned to a team without their knowing, are loaded into the
game but cannot spawn. This UX could need some improvement. And the
logic can be improved too, ie. don't "kick" too soon, check the number
of Clans in the lobby and the number of teams to decide if we can assign
the 'overflowing' Clan player to another team that doesn't have
rivalling Clan players. Far out of scope for this PR.
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
tryout33
## Description:
Adds a map based on Lisbon and the surrounding area. Also credits the
ESA's Copernicus Digital Elevation Model, which was used to create this
map and the Gulf of St. Lawrence map.
<img width="1257" height="1257" alt="screenshot of the new Lisbon map"
src="https://github.com/user-attachments/assets/39fa73da-c77d-4d5c-8d00-7ee2794d0298"
/>
## Checklist:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Discord Username:
sehentsin
## Description:
Adds the Gulf of St. Lawrence map and the flags of New Brunswick
(ca_nb), Nova Scotia (ca_ns), and Prince Edward Island (ca_pe).
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
<img width="1380" height="1150" alt="Screenshot 2025-11-26 201008"
src="https://github.com/user-attachments/assets/77531058-b429-4c68-b645-a3e59033458b"
/>
<img width="1434" height="1195" alt="Screenshot 2025-11-26 202128"
src="https://github.com/user-attachments/assets/9c3e2bc2-882f-4662-a32b-16e17db852f2"
/>
## Discord username
sehentsin
Resolves#2517
## Description:
Fixes territory skin option not doing anything.
Also changes some instances of directly accessing
"this.cosmetics.pattern" to use the precalculated "pattern" variable.
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
Lavodan
## Description:
Nations no longer send random boats to their bordering enemies.
Usually it looked a bit stupid when nations did that ("Why don't you
attack your bordering enemy directly?")
## Please complete the following:
- [X] I have added screenshots for all UI updates
- [X] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [X] I have added relevant tests to the test directory
- [X] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
FloPinguin
## Description:
Add 🏭 & 🚂 emojis, which is commonly requested. Also 3 other emojis to
fill the new emoji line: 🫴✋🙏
To be discussed: this adds a new line in the emoji list: is it too much?
Eventually we could use categories to filter the emojis.
<img width="199" height="465" alt="image"
src="https://github.com/user-attachments/assets/cb6a44e9-30cd-48f7-90ab-8a7c32dcd180"
/>
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
IngloriousTom
## Description:
Add 🏭 & 🚂 emojis, which is commonly requested. Also 3 other emojis to
fill the new emoji line: 🫴✋🙏
To be discussed: this adds a new line in the emoji list: is it too much?
Eventually we could use categories to filter the emojis.
<img width="199" height="465" alt="image"
src="https://github.com/user-attachments/assets/cb6a44e9-30cd-48f7-90ab-8a7c32dcd180"
/>
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
IngloriousTom
## Description:
- Removed the temporary UnitType.Construction and embedded construction
state into real units via isUnderConstruction().
- Centralized non-structure spawning to perform a single validation
right before unit creation/launch.
- Updated UI layers to render construction state without relying on the
removed enum.
- Adjusted and created tests to match the new flow and to cover the
no-refundscenarios.
# Tests updated
- tests/economy/ConstructionGold.test.ts: covers structure cost
deduction and income, tolerant of passive income; ensures no refunds
during construction.
- tests/nukes/HydrogenAndMirv.test.ts: accounts for single-check launch
flow; MIRV test targets a player-owned tile; ensures launch after
payment.
- tests/client/graphics/UILayer.test.ts: mocks now provide
isUnderConstruction and real type strings;
## Please complete the following:
- [X] I have added screenshots for all UI updates
- [X] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [X] I have added relevant tests to the test directory
- [X] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
CrackeRR1
---------
Co-authored-by: Evan <evanpelle@gmail.com>
## Description:
Clan tag was removed when overwriting profane username. The local player
still sees the name they put in though, and are assigned to a team based
on the clan tag. Other player's browsers don't assign them to their team
because the clan tag isn't visible to them.
**Fixes:**
- GameRunner.ts > username.ts: fetch clan tag before potentially
overwriting bad username. Prepend non-profane clan tag back to the name
string afterwards.
- Util.ts: added getClanTagOriginalCase; we can't use getClanTag in
censorNameWithClanTag because it returns all caps and we needed to
retain the orginal capitalization. Explained in code comment.
- Game.ts: no changes. By keeping the getClanTag in PlayerInfo
contructor, TeamAssignment still gets clan param to correctly assign
clan teams, other players get to see the clan tag of the "BeNicer"
player, and GameServer archivegame() and LocalServer endGame() can still
do getClanTag to have the same data at the end of the game, and test
files keep working.
**Screencap after fix:**
https://github.com/user-attachments/assets/564c0ffd-577e-4653-ba33-155d2135a9f0
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
tryout33
## Description:
Clan tag was removed when overwriting profane username. The local player
still sees the name they put in though, and are assigned to a team based
on the clan tag. Other player's browsers don't assign them to their team
because the clan tag isn't visible to them.
**Fixes:**
- GameRunner.ts > username.ts: fetch clan tag before potentially
overwriting bad username. Prepend non-profane clan tag back to the name
string afterwards.
- Util.ts: added getClanTagOriginalCase; we can't use getClanTag in
censorNameWithClanTag because it returns all caps and we needed to
retain the orginal capitalization. Explained in code comment.
- Game.ts: no changes. By keeping the getClanTag in PlayerInfo
contructor, TeamAssignment still gets clan param to correctly assign
clan teams, other players get to see the clan tag of the "BeNicer"
player, and GameServer archivegame() and LocalServer endGame() can still
do getClanTag to have the same data at the end of the game, and test
files keep working.
**Screencap after fix:**
https://github.com/user-attachments/assets/564c0ffd-577e-4653-ba33-155d2135a9f0
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
tryout33
## Description:
In commit
https://github.com/openfrontio/OpenFrontIO/commit/bbf72bd14f7f31146c687523aea8fc0aff31bbe1#diff-ee2fcbca50d87cc09d2c7d2667210defe2e3e111239820c89c40283be5385b64
it was added that startManpower in DefaultConfig used
playerInfo.nation.strength to set the starting troops for a Nation. In
ExecutionManager, param nation of PlayerInfo was set during the
instantiation of a new FakeHumanExecution.
However in commit
https://github.com/openfrontio/OpenFrontIO/commit/d6a412aa50dd86d474d80c216fd9ba36e7426ef9#diff-2d0a5d8b171d8b504f934891025e42742e142ef0964d6e17712bfdcd30bf050c
the changes made it so that **param nation of PlayerInfo was never
set**. While startManpower in DefaultConfig still checked for
playerinfo.nation.strength. Since it was always undefined, it would use
a multiplier of 1 instead of the actual nation strength.
This PR fixes it by passing the nation strength to param nationStrength
in PlayerInfo. Removing param strength from class Nation. Strength isn't
used anywhere else so this isn't a problem and it also consolidates
human player info and nation player info even more. We could have also
used the Nation.strength directly, but that would have required more
code in addPlayers and addPlayer in GameImpl, especially for Teams
games. So this PR has the simplest solution.
- I did add a config setting useNationStrengthForStartManpower with a
comment that explains its reason for being. Namely that in the months
that startManpower didn't get to use nation strength because of the bug,
FakeHumans have become much harder to fight. Re-enabling higher starting
troops from this fix would make them even harder to fight, and i think
rebalancing is needed before that.
- Or we could decide to scrap Nation strength altogether, as it is only
ever used to set starting troops anyway. This would make map making a
little easier as a bycatch.
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
tryout33
## Description:
This PR optimizes how the rail network looks up railroads connecting two
stations by introducing an O(1) neighbor→railroad map on `TrainStation`.
It also updates `getOrientedRailroad` and railroad deletion to use this
new API, avoiding repeated linear scans over all railroads attached to a
station.
### What changed
- **TrainStation neighbor→railroad index**
- Added `railroadByNeighbor: Map<TrainStation, Railroad>` to
`TrainStation` for quick edge lookup.
- Kept `railroads: Set<Railroad>` for iteration and existing APIs.
- Updated lifecycle methods to keep both data structures in sync:
- `addRailroad(railRoad: Railroad)` now:
- Adds to `railroads`.
- Computes the neighbor station (`railRoad.from === this ? railRoad.to :
railRoad.from`).
- Stores the mapping in `railroadByNeighbor`.
- `removeRailroad(railRoad: Railroad)` now:
- Removes from `railroads`.
- Removes the corresponding entry from `railroadByNeighbor`.
- `clearRailroads()` now clears both `railroads` and
`railroadByNeighbor`.
- Added `getRailroadTo(station: TrainStation): Railroad | null` to
retrieve the connecting railroad in O(1).
- **Use the new API in `TrainStation` and `Railroad`**
- `TrainStation.removeNeighboringRails(station)` now calls
`removeRailroad(toRemove)` instead of manually deleting from the set,
ensuring the map stays in sync.
- `Railroad.delete(game)` now calls `from.removeRailroad(this)` and
`to.removeRailroad(this)` instead of mutating the sets directly.
- **Refactor `getOrientedRailroad` to use O(1) lookup**
- Replaced a linear scan over `from.getRailroads()` with a direct
lookup:
```ts
export function getOrientedRailroad(
from: TrainStation,
to: TrainStation,
): OrientedRailroad | null {
const railroad = from.getRailroadTo(to);
if (!railroad) return null;
// If tiles are stored from -> to, we go forward when railroad.to === to
const forward = railroad.to === to;
return new OrientedRailroad(railroad, forward);
}
```
- Behavior is preserved:
- `getRailroadTo` returns the same `Railroad` instance that was
previously found by scanning `getRailroads()`.
- Direction (`forward` vs reversed) is still derived from the
`Railroad.from` / `.to` fields in the same way as before.
### Motivation
- `getOrientedRailroad` and upcoming logic both need to resolve “the
railroad between station A and station B” frequently.
- The old pattern (`for (const railroad of from.getRailroads()) { ...
}`) was:
- O(degree) per lookup,
- Repeated in multiple places,
- Harder to maintain as more features (like fare-based costs) touch this
code.
- Centralizing edge lookup in a dedicated `railroadByNeighbor` map makes
this:
- **O(1)** per lookup,
- Less error-prone (one source of truth),
- Easier to reuse from new systems (e.g. train pathfinding, fare-aware
logic).
### Impact / Risk
- **Public behavior:** No functional change in how railroads are
created, deleted, or oriented; only the lookup mechanism changed.
- **Internal invariants:** Correctness relies on:
- All railroad creations using `addRailroad` on both endpoints (already
true via `RailNetworkImpl.connect`).
- All removals (`Railroad.delete`,
`TrainStation.removeNeighboringRails`, `disconnectFromNetwork`) using
`removeRailroad` / `clearRailroads`, which this PR updates.
- **Tests:** Existing `TrainStation` tests still pass; they exercise
`addRailroad`, `removeNeighboringRails`, and `getRailroads()`, which
continue to behave the same from the outside.
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
DISCORD_USERNAME
Resolves#2448
Hi team,
I've implemented and locally tested the alliance-related changes
(including unit tests and some manual simulation with multiple browser
profiles).
Unfortunately I wasn't able to perform full end-to-end testing on the
live game server with two separate machines/accounts.
If someone on the team (or another contributor) can verify the alliance
flow with two real players, that would be greatly appreciated before
merging. Happy to hop on a call or provide any clarification needed.
Thanks!
## Description:
Fixed a race condition bug where donations (troops/gold) between human
players failed after forming an alliance. The issue was caused by a
one-tick delay in `AllianceRequestReplyExecution`: the alliance
acceptance logic ran in `tick()` instead of `init()`, meaning the
alliance wasn't created until the tick after the execution was added. If
a donation execution was added in the same turn as the alliance
acceptance, it would fail the `isFriendly()` check because the alliance
didn't exist yet.
**Root cause:** When human players formed alliances via reply, the
execution model delayed alliance creation by one tick, while bots called
`accept()` directly without this delay.
**Solution:** Moved alliance acceptance logic from `tick()` to `init()`
in `AllianceRequestReplyExecution.ts`, ensuring immediate alliance
creation and eliminating race conditions with donations.
**Changes:**
- Modified
`src/core/execution/alliance/AllianceRequestReplyExecution.ts` to
process alliance replies in `init()` instead of `tick()`
- Added comprehensive test suite `tests/AllianceDonation.test.ts` with 5
test cases covering donation scenarios after alliance formation (reply
and mutual request flows)
- All existing tests pass (323 total)
## Please complete the following:
- [x] I have added screenshots for all UI updates (N/A - backend logic
fix only)
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file (N/A - no user-facing text
changes)
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
Discord: loacky
GitHub: @LoackyBit
---------
Co-authored-by: Evan <evanpelle@gmail.com>
## Description:
Made a type of map we don't already have: Small square map, one Island
in each corner. Could be great for boat gameplay or maybe even nuke
wars.
The special thing is: **All islands are exactly 25% of the territory!**
Was a lot of work drawing that in GIMP...
<img width="1164" height="1168" alt="Screenshot 2025-11-20 000839"
src="https://github.com/user-attachments/assets/ad8345c7-562d-49e8-b367-12be9274f3e4"
/>
<img width="1227" height="1222" alt="Screenshot 2025-11-20 000633"
src="https://github.com/user-attachments/assets/f7e2c58a-fcb3-4e07-91f2-aead5f497fad"
/>
<img width="361" height="288" alt="Screenshot 2025-11-20 000655"
src="https://github.com/user-attachments/assets/120f82ef-2d19-497b-8a31-819e30013c89"
/>
<img width="756" height="282" alt="Screenshot 2025-11-20 005949"
src="https://github.com/user-attachments/assets/8ee04da3-d5fa-4ec9-9e99-9e30ebda2b78"
/>
## Please complete the following:
- [X] I have added screenshots for all UI updates
- [X] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [X] I have added relevant tests to the test directory
- [X] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
FloPinguin
---------
Co-authored-by: Evan <evanpelle@gmail.com>
Resolves#1092
## Description:
Added a team preview to the Host Lobby: players listed on the left,
teams on the right in two scrollable columns with color dots matching
in-game colors. Implemented accurate server-parity team assignment
(including clan grouping).
Screenshots:
<img width="817" height="519" alt="Screenshot 2025-11-13 173721"
src="https://github.com/user-attachments/assets/ec646238-7efa-4c8f-9c0a-171b61fd3f20"
/>
<img width="762" height="425" alt="Screenshot 2025-11-13 175400"
src="https://github.com/user-attachments/assets/ebdccb80-4c07-41d5-8f69-3ea983d4b243"
/>
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
abodcraft1
---------
Co-authored-by: Evan <evanpelle@gmail.com>
Description:
There is a boating exploit where players could repeatedly send and
retreat boats to effectively increase troop regeneration and maintain
almost double the max troop cap. This PR fixes#2388.
Summary
This pr adds a troop penalty to the boat retreats in
src/core/execution/TransportShipExecution.ts of 25 percent (currently
the same as land attacks, but may require fine tuning). this prevents,
to some degree, the ability to stockpile large amounts of troops above
your max cap.
Testing
I tested the change locally and confirmed that the troop penalty is
applied at the correct times and under the correct conditions, i.e. it
is only applied on successful retreat (applying on reteat initiation may
be confusing if you are told you have lost troops if the transport ship
never arrives anyway). The boating exploit is far less effective with
this fix in place.
Notes
- This is a exploit and does not introduce any new features.
- Existing game behavior outside the retreat penalty remains unchanged.
- The attack message is the same as the land attack, so the translation
should already be present.
Checklist
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
---------
Co-authored-by: Evan <evanpelle@gmail.com>
## Description:
Add dynamic alliance icon with time-based fill and extension request
indicator
- Implement bottom-up green fill on alliance icon proportional to
remaining time
- Use AllianceIconFaded.svg as base layer with green overlay clipped
from top
- Add 20-82.40% clip range to account for icon vertical offset
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
<img width="1132" height="631" alt="Screenshot 2025-11-18 205205"
src="https://github.com/user-attachments/assets/4af71ddc-f847-4460-9046-167275efc773"
/>
<img width="1387" height="792" alt="Screenshot 2025-11-18 205532"
src="https://github.com/user-attachments/assets/9dd0e018-323f-4de1-bae8-2633c09fe867"
/>
## Please put your Discord username so you can be contacted if a bug or
regression is found:
hauke4707
---------
Co-authored-by: Evan <evanpelle@gmail.com>
Resolves#2452
## Description:
Created a Clan Stats PR to show top clans. In another PR we can show the
player leaderboard to show top players.
Based on PR from https://github.com/Geekyhobo
<img width="659" height="792" alt="Screenshot 2025-11-19 at 10 00 40 AM"
src="https://github.com/user-attachments/assets/9333b7e2-2357-47a6-a7c8-788cf81e9be3"
/>
## Please complete the following:
- [x] I have added screenshots for all UI updates
- [x] I process any text displayed to the user through translateText()
and I've added it to the en.json file
- [x] I have added relevant tests to the test directory
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
evan
Co-authored-by: Geekyhobo <geekyhobo@users.noreply.github.com>
## Description:
After the v27 playtest, some players experienced instant death on spawn.
The issue was that the human random spawn occasionally coincided with a
bot’s spawn. Previously, bot spawns didn’t account for human spawn
locations and could appear on the same tile, now they don’t.
## Please complete the following:
- [x] I confirm I have thoroughly tested these changes and take full
responsibility for any bugs introduced
## Please put your Discord username so you can be contacted if a bug or
regression is found:
nikolaj_mykola
---------
Co-authored-by: Evan <evanpelle@gmail.com>
scamiv