Commit Graph

1681 Commits

Author SHA1 Message Date
Ludovic Marie 7b11005688 Merge remote-tracking branch 'origin/main' into HEAD
# Conflicts:
#	crates/horizon-client/src/game/remote.rs
2026-10-27 06:37:26 +01:00
levzzz ddfa71a46f Merge the End crystal and dragon selectors into the cross-level selectors
Both pull requests reshape Game::select in game/commands.rs, so the
cross-level branch now carries the End crystal and dragon candidates as
well: the crystals and the dragon are listed by level_candidates, and
level_holds knows them, so an unbounded selector finds them in the level
that holds them.

Co-Authored-By: minimax-m3.1-flash (Claude Code) <noreply@users.noreply.github.com>
2026-10-27 09:33:56 +05:00
Ludovic Marie a2f62b1545 parity_hosts: playhorizon's workers run on CPUs 0-3, 3 harnesses at a time
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 04:42:19 +01:00
Ludovic Marie cb20bb508b remote.sh: runs get CPUs 0-3 by default; benchmarks hold them with --exclusive
playhorizon's CI runner has CPUs 4-7, and remote.sh pinned agents' runs there
too: CPUs 4-7 sat at 100% while 0-3 idled, and window harnesses in pull
requests' parity jobs lost ticks. Runs share a lock per CPU set; an
--exclusive run (a benchmark) holds it alone. world_reopen also waits for
the world to load before its first count, as after each reopen.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 04:31:40 +01:00
Ludovic Marie 4168590221 Merge remote-tracking branch 'origin/main' into fix/window-harnesses-under-load 2026-10-27 04:30:35 +01:00
Ludovic Marie 03cc30ae45 Merge origin/main into refactor/horizon-items
entity_nbt keeps the name_text re-export (main's body is the same as
horizon-client-core's) and takes main's new spawn_bonus. Main's edits to
blocks, bundle, enchant, item_data and items land in horizon-items.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 04:28:06 +01:00
Ludovic Marie 5a6ae76ec3 Merge pull request #2486 from ludovic111/fix/parity-reap-orphans
🤖 Agent: parity.py reaps the harness processes a failed or timed-out run leaves behind; parity passed on playhorizon.
2026-10-27 04:16:36 +01:00
Ludovic Marie 8b87d128e6 parity: world_reopen and server_player_sculk count ticks, not request round trips
On playhorizon (frames drawn on the CPU, control requests answered slowly)
world_reopen read the minecarts before the mineshaft's last one arrived (3
then 4 on every reopen), and the sculk walk polled the sensor with separate
requests that missed its 30-tick active window. The minecart list is read
once it stops changing; both clients sample the sensor every tick from one
ctl script, alongside the walk.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 04:08:32 +01:00
Ludovic Marie 4ba8105ecc Merge pull request #2104 from rdbtCVS/fix/1482-evoker-fangs
🤖 Agent: Gate check by Ludovic's AI agent (Claude). Checked against 26.3: the server sends entity event 4 once its warmup ends (`EvokerFangs.tick`), `handleEntityEvent(4)` starts the client's attack and plays the sound locally, and the client only counts `lifeTicks` down for the animation, while the server deals the damage and discards. Moving the host's sound to `play_unshared` matches that, since each client plays its own. It conflicted with #2094 (both hook entity events) once that merged, so I merged main into it, keeping both calls. `lan_e2e.py --check fangs` passes in both modes. It conflicted again with main in `tools/lan_e2e.py` (new checks on both sides), so I merged main in once more, keeping every check; clippy is clean. Fixes #1482. Thanks @rdbtCVS! Gate: this batch merged with wave 1 in one tree: clippy `-D warnings` clean, 1164 horizon-client lib tests pass, and of the 88 harnesses `parity.py --changed-files` picked, 87 pass (`block_entity_render` fails from #2099, which is held).
2026-10-27 03:56:41 +01:00
Ludovic Marie 226b42a046 Merge pull request #2449 from RoyalPixelsDE/fix/end-portal-without-players
🤖 Agent: Gate check by Ludovic's AI agent (Claude). Checked against 26.3: `EnderDragonFight.tick` scans the state and places the exit portal only when `dragonEvent.getPlayers()` is not empty, so a force-loaded End with nobody in it gets no portal, as here; the new `forced_end` harness compares both servers. Gate: this wave's 50 PRs merged on main in one tree (union 9033a9c3b): fmt and clippy `-D warnings` clean, the horizon-client (1136 lib plus the integration tests), horizon-client-core (37), horizon-world (165) and horizon-protocol (100) tests pass, the fast build links, and the 140 harnesses `parity.py` picks for the tree's changed files pass on playhorizon (remote_fireworks, a known load flake there, passed on a rerun on a quiet Mac mini). `lan_e2e.py` passes 14 checks with the host on its server thread and in-frame (chest, xp, damage, kills, projectiles, mobs, drops, body, stopsound, digging, instant_digging, books, knockback, barter); `hurt` fails on main too when run after `damage` (check-order dependency, not this wave). Thanks @FusselTV!
2026-10-27 03:54:13 +01:00
Ludovic Marie 80e28dfe63 creative_mining harness: items.rs now lives in horizon-items
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 03:49:17 +01:00
Ludovic Marie 27f73e1db3 Merge origin/main into refactor/horizon-items
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 03:49:17 +01:00
Ludovic Marie e91678b994 Merge pull request #2589 from ludovic111/fix/remote-fireworks-slow-screens
🤖 Agent: remote_fireworks passes 3 of 3 on playhorizon (frames drawn on the CPU) with the earlier shots and the newly-red count.
2026-10-27 03:26:40 +01:00
Ludovic Marie 75bcbce441 Merge origin/main into refactor/horizon-items
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 02:27:09 +01:00
Ludovic Marie fde2dfffef Merge pull request #2598 from ludovic111/fix/friends-guest-spot
🤖 Agent: fixes parity-main's friends_integration (red since 4be5c8eb5). Cause: #2501 spawns new players at a random spot near the spawn, as vanilla does, and test_friends.py assumed the guest stood on the spawn itself. Main failed 4/4 on the Mac mini, main with #2501 reverted passed 2/2, and this branch passed 4/4 on the server thread and 1/1 in-frame. Python only; harness-only, skip-release.
2026-10-27 02:14:59 +01:00
Ludovic Marie a0ef14f1e2 Merge pull request #2582 from ludovic111/perf/parity-one-client-profile
🤖 Agent: harnesses build horizon-client in one profile; 6 harnesses checked fresh on playhorizon and the Mac mini.
2026-10-27 01:58:38 +01:00
Ludovic Marie 0a75323caa Merge pull request #2584 from ludovic111/perf/parity-manifest-aware
🤖 Agent: crate splits' workspace-only manifest changes no longer select every harness (354 to 209 on #2160 and #2215); other PRs select as before.
2026-10-27 01:49:27 +01:00
Ludovic Marie 2f489f54c5 Merge origin/main (with #2160) into refactor/horizon-items
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 01:38:11 +01:00
Ludovic Marie f5b7bdba1d Merge pull request #2160 from ludovic111/refactor/horizon-command
🤖 Agent: merging the first crate split (horizon-command + horizon-client-core, pure move, re-exported under the old paths). CI parity passed at 17f8419ac (head merged into main), verify and launcher pass, approved by rdbt's review agent; landed by hand while the REST pool refills.
2026-10-27 01:36:58 +01:00
Ludovic Marie 6087b01f66 test_friends: put the guest on cleared, flat ground before it walks
Since #2501 a player new to the world spawns at a random free spot near the
world spawn, as vanilla's PlayerSpawnFinder does. The guest often stood against
a hill, so walking forward moved it less than a block, or terrain hid the block
it had to aim at: friends_integration failed on every run on the Mac mini.

The host now clears a floor beside itself and teleports the guest there before
the movement checks, and the block check waits for the guest to stand still.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 01:21:06 +01:00
Ludovic Marie b6dd6c35ae Merge remote-tracking branch 'origin/main' into perf/parity-manifest-aware 2026-10-27 01:10:36 +01:00
Ludovic Marie 1af4d0e432 parity: a workspace crate's features count through the sources they gate
A manifest line defining or turning on a workspace crate's feature no longer
makes the manifests count, if every source naming the feature (or one it turns
on) changed too, as in a split, where the new crate's files are all new.
#2215 (horizon-items with a test-support feature) now selects 209 harnesses
instead of 354.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 01:10:36 +01:00
Ludovic Marie 68c9597237 Merge remote-tracking branch 'origin/main' into perf/parity-one-client-profile 2026-10-27 01:09:18 +01:00
Ludovic Marie bedba4a3d9 lan_e2e fireworks: count newly red pixels, take the later shots 4 ticks after the burst
On playhorizon (frames drawn on the CPU) a screenshot takes ~0.6 s: with a
16-tick wait the later shots came after the sparks' lifetime, and the setup's
own red chat lines, gone by then, made the host's count negative.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-27 01:08:50 +01:00
levzzz 9ab0e2b6a8 Cover the slime block's push and bounce in the LAN harness
A piston pushing a slime block moves the body in front of it and bounces
it: the block's swept collision shape pushes the body out of the way, and
the slime block sets the body's velocity along the push axis
(`PistonMovingBlockEntity.moveCollidedEntities`, the `causeBounce` branch).
The new `piston-slime` LAN case checks both for the host and the guest, with
a plain block as the control: a plain block only pushes the body 0.81
blocks, the slime block pushes and bounces it 2.5.

Co-Authored-By: minimax-m3.1-flash (Claude Code) <noreply@users.noreply.github.com>
2026-10-27 05:08:40 +05:00
levzzz 04672c9613 Merge remote-tracking branch 'upstream/main' into fix/selector-end-crystals 2026-10-27 05:04:30 +05:00
levzzz fa4f38f824 commands: an unbounded selector reaches the other loaded levels
`Game::select` gathered its candidates from the level in place only, so
`@e[tag=...]` never found an entity in another dimension the server
still holds: vanilla's `EntitySelector` searches every loaded level
unless the selector carries a spatial constraint.

The selector now searches each loaded level in turn, and the commands
that act on the entity it found (`/data get entity`, `/kill`, `/tp`)
act on it in the level that holds it. The experience orbs and the
projectiles are named by their entity id, like the other entities, so
the entity a command acts on is the one the selector found.

Fixes #1524

Co-Authored-By: minimax-m3.1-flash (Claude Code) <noreply@users.noreply.github.com>
2026-10-27 05:01:07 +05:00
levzzz 14dac749e3 Merge remote-tracking branch 'upstream/main' into fix/thrown-projectile-flight 2026-10-27 04:50:53 +05:00
levzzz 10d0132a53 Merge remote-tracking branch 'upstream/main' into fix/mining-dig-stop-check 2026-10-27 04:50:40 +05:00
minimax-m3.1-flash 5be7c02d0c Cover the mining dig stop in the LAN harness (#1340, #1908)
A block the client breaks at once (a creative break, a dead bush) sends no STOP or ABORT, so the server has to forget the dig itself. Without it the server keeps mining the emptied cell: the block put back there is destroyed again (#1340) and the mining sound keeps playing at it (#1908, the cell's sound group is stone).

The new dig_stop check breaks a stone block in creative and a dead bush in survival for the host and the guest, then puts the block back and waits: it must stay, and no mining sound may play.

Co-Authored-By: minimax-m3.1-flash (Claude Code) <noreply@users.noreply.github.com>
2026-10-27 04:38:22 +05:00
minimax-m3.1-flash e11471cda0 Run the LAN harness headless
HORIZON_E2E_HEADLESS launches the games with --offscreen --no-pause, so the suite runs on a machine with no focused window: without it the games pause on the focus loss and wait_ready never completes. macOS keeps its offscreen launch.

Co-Authored-By: minimax-m3.1-flash (Claude Code) <noreply@users.noreply.github.com>
2026-10-27 04:38:22 +05:00
levzzz 6384234e37 Stop a piston push at the world border
`move_pushed` resolved the pushed movement with `SceneCollisions`, which
supplies no world-border collision, so a piston next to the border could push
a player through it even though ordinary movement stops at the border.
`Entity.move` collects the border's walls in
`collectCollidersIgnoringWorldBorder` whenever
`WorldBorder.isInsideCloseToBorder` holds, and `CollisionGetter.borderCollision`
adds `WorldBorder.getCollisionShape` to them. The push now resolves through
`EntityCollisions::with_border`, which applies that same rule to the pushed
body's own x and z, so a body pushed into the border's wall stops there.

Co-Authored-By: minimax-m3.1-flash (Claude Code) <noreply@users.noreply.github.com>
2026-10-27 04:32:16 +05:00
Carlos Nunez 3caf28c4d6 Merge remote-tracking branch 'upstream/main' into tmp/merge-2298
# Conflicts:
#	tools/lan_e2e.py
2026-10-26 19:27:09 -04:00
levzzz 4c9075a072 Keep the thrown-item reproduction for #1762
A server's client ticks a projectile itself between the entity tracker's
updates, as vanilla's Entity no-op interpolation handler does; before #2086 it
only eased towards the server's position, so a thrown egg, ender pearl or wind
charge hung in the air for half a second. The client tick is now a named
function with unit tests for that flight, and the parity harness keeps the
reproduction the issue asked for: a connected client throws each of them and
samples their position twice as often as the tracker's interval. The LAN
harness can also run headless now (HORIZON_E2E_HEADLESS), so the games render
offscreen and never pause on the focus loss on a machine with no focused
window.

Co-Authored-By: minimax-m3.1-flash (Claude Code) <noreply@users.noreply.github.com>
2026-10-27 04:08:31 +05:00
Louis Nordbø 64cd55cf74 Finish a use on the server's USE_ITEM_COMPLETE, and eat on the client too
On a server client (single player's server thread, LAN guests, dedicated
servers) the window ended a consumable's use on its own timer and never ate
the food: its FoodData waited for the server's next ClientboundSetHealth.
With the button still held, the next startUseItem found the food bar below
20 and began another bite, so the eating animation went on after the bar
was full (#2471).

Vanilla's client never completes a use by itself: LivingEntity.updateUsingItem
completes it only on the server, and ServerPlayer.completeUsingItem first
sends the player entity event 9 (USE_ITEM_COMPLETE). Player.handleEntityEvent(9)
then runs completeUsingItem on the client, where Consumable.onConsume calls
the FoodProperties listener on both sides: the client's FoodData eats at once.

- The server sends event 9 to the acting guest when a consumable, a potion or
  a spyglass use completes. Like the guest's packets, the events are queued
  during their turn (kept across nested turns) and sent with their entity id.
- The client keeps using until that event, then shows the last crumbs, eats
  the food (the next ClientboundSetHealth replaces it) and starts the item's
  use_cooldown, which ItemStack.use only starts at once for an item with no
  use time.
- What vanilla's client checks so that a use can't outlive the server's:
  MultiPlayerGameMode.useItem passes on an item cooling down (chorus fruit
  held through its cooldown), LivingEntity.updatingUsingItem stops when the
  hand no longer holds the item the use began with (ItemStack.isSameItem),
  a respawn is a new player with no use, and Consumable.startConsuming
  starts none for a consumable with no use time.
- ItemCooldowns tick where Player.tick ticks them: after LivingEntity.tick's
  use on the server and after Minecraft.handleKeybinds on the client. Both
  sides ticked them first, so a guest's entry ended a tick later on the
  server than on their client, whose held retry was refused and never ended.

lan_e2e's new eat_full check eats melon slices up to a full bar with the
button still held: before, a new bite had started (use ticks 26, 2, 9, 15);
now nothing is in use. Held on two chorus fruits, both are eaten across the
cooldown. consume-release holds its completed use a little longer, as the
use now ends with the server's event.

Fixes #2471

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 23:59:03 +01:00
levzzz 8909e0510c commands: selectors find the End crystals and the Ender Dragon
Horizon's command targets enumerated the ordinary entity collections
only, so `@e[type=end_crystal]` and `@e[type=ender_dragon]` matched
nothing: the End crystals and the dragon live in their own lists, and
neither carried an entity id a command could name.

The crystals and the dragon now take an id from the entity counter when
they are created, as EndCrystal's and EnderDragon's constructors do; the
selector offers them as candidates; `/data get entity` reads them and
`/kill` removes them the way `Entity.kill` does. `/summon
minecraft:end_crystal` creates one, as vanilla does.

Fixes #1531

Co-Authored-By: minimax-m3.1-flash (Claude Code) <noreply@users.noreply.github.com>
2026-10-27 03:55:39 +05:00
Ludovic Marie 17f8419acd Merge origin/main into refactor/horizon-command
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 23:40:59 +01:00
Ludovic Marie 9555f375d4 parity: remote_fireball is no longer known flaky (#2499)
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 23:31:29 +01:00
Ludovic Marie 1b1571ed6a parity: a crate split's manifest changes don't select every harness
Cargo.lock is a shared input and every harness depends on its crates'
Cargo.toml, so a crate split (#2160) selected all 352 harnesses although
its moved code reaches 123. With --changed, Cargo.lock and Cargo.toml
changes that only add, remove or move workspace crates (no registry or git
package, version, feature, dependency or build setting changed across them,
compared with the merge base) are left out; the moved code counts through
its own sources. Any other manifest change still counts, as before.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 23:30:59 +01:00
Ludovic Marie a891037c9f parity: harnesses build horizon-client in the fast profile only
Under parity.py the release and fast profiles have identical settings
(PROFILE: no LTO, 16 codegen units, incremental), yet 38 harnesses built
horizon-client in release and 16 others in fast: every run where the client
changed compiled it twice. They all use fast now. Release stays for
horizon-world's probes, which no fast harness builds.

Measured on playhorizon (4 CPUs), all 13 probe builds after 78 merges:
1578 s, of which the release client (bin + client_probe) took 485 s.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 23:24:54 +01:00
Ludovic Marie 840d11a3b2 Merge pull request #2499 from ludovic111/fix/remote-fireball-ticks
🤖 Agent: test-health-1: remote_fireball flake fix (slow fireball, one ctl script; 4/4 on playhorizon); checks green.
2026-10-26 22:44:46 +01:00
levzzz bafd010865 Merge remote-tracking branch 'upstream/main' into fix/piston-entity-push 2026-10-27 02:22:04 +05:00
Ludovic Marie d090686ffa Merge pull request #2500 from ludovic111/tools/remote-cleanup
🤖 Agent: test-health-1: remote.sh stops the host command with its caller and cleans what it leaves; checks green.
2026-10-26 22:21:54 +01:00
Ludovic Marie 5f73d0d02d Merge refactor/horizon-command (with main) into refactor/horizon-items
Main's items.rs changes (equip_damage_on_hurt, resists_cactus) land in
crates/horizon-items/src/items.rs.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 22:20:35 +01:00
Ludovic Marie 5a1d1c38ed Merge pull request #2501 from ludovic111/fix/lan-guest-stations
🤖 Agent: LAN guests (host on its server thread) and a dedicated server's new players joined inside each other on the exact spawn, so they clicked the other player instead of a table or chest. New players now get PlayerSpawnFinder.findSpawn as in vanilla. lan_e2e spawned/crafting/chest pass in both host modes, a dedicated two-guest check passes, and the 32 affected harnesses pass.
2026-10-26 22:19:19 +01:00
Ludovic Marie aa0eb406d7 Merge pull request #2496 from ludovic111/fix/dedicated-server-steps
🤖 Agent: test-health-1: dedicated_server steps wait for the server's reply (flake fix, passes under injected stalls); checks green.
2026-10-26 22:14:45 +01:00
Ludovic Marie 8f02822a9a remote.sh: stop the command on the host with its caller, and what it leaves running
The command ran on after remote.sh was stopped or its connection dropped,
holding the name's lock, and a game or daemon it left behind outlived the
run. It now runs in a scope named after the run (remote-<name>.scope),
stopped when the command ends, when the ssh session ends first, and before
the next run under that name.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 22:09:44 +01:00
Ludovic Marie 031f1e58a3 remote_fireball: a slow fireball, sampled from one ctl process
The harness sampled the fireball from a new horizon-ctl process each time,
10-16 ticks apart on a loaded runner: the fireball (~2 blocks a tick) left
the cleared corridor or the guest's tracking range before the sixth sample,
and the harness failed. It now flies slowly (acceleration_power 0.02, under
0.4 blocks a tick) down a longer corridor, and the guest takes 16 samples
in one ctl script (lan_e2e's Game.script returns every reply with every=True).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 22:07:12 +01:00
Ludovic Marie d3a662b760 Merge origin/main into refactor/horizon-command
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 22:06:35 +01:00
Ludovic Marie d6a80c5193 Merge pull request #2489 from ludovic111/tools/remote-linux
🤖 Agent: test-health-1: tools/remote.sh runs agents' Linux work on playhorizon (fetch from GitHub, no secrets sent); verify green.
2026-10-26 21:52:49 +01:00