Main's critical-hit change added the same Strike.attacker field; keep
one, then the crit particles before the victim's thorns.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: Gate check by Ludovic's AI agent (Claude). I checked the loot functions against vanilla 26.3 directly. A vanilla server placed 48 chests with fixed LootTableSeed values: 16 tables (end city, stronghold library, ancient city, mansion, dungeon, trial chamber rewards and more) times 3 seeds, then rolled them. Horizon then opened the same world. Every chest got the same items, counts, slots and enchantments as vanilla, enchant_randomly and enchant_with_levels included: 48/48 on the server thread and 48/48 in-frame. The only difference is the order of keys inside the enchantments map, which vanilla doesn't read. EnchantRandomly and EnchantWithLevels match the bytecode, and rdbt's single-level review point is covered by `a_single_level_enchantment_draws_no_level`. It is also built and run in an integration tree with the other open PRs. Thanks @levzzz5154!
VanillaMobAiOracle ran out of its 3g heap in PoiManager.tick on a busy CI
runner. Each scenario forces 9x9 chunks at a new origin; the live heap after
GC swings between 0.3 and 1 GB over a run, with peaks near 2 GB before GC.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Hub::stack resolved every item id of a stack to the outer item's, so a
bundle's contents and a crossbow's charged projectiles reached a server's
clients as more bundles and crossbows. The tooltip then drew a bundle of
bundles, full. Each stack inside now looks its own id up.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A remote client (single player's window on its server thread, a LAN
guest, a dedicated server's client) never ran the client side of a
piston's move: the block event reached block_entity_event, which has no
piston ticker, so moving_piston blocks arrived with no moving block
entity and pushed blocks teleported (#1704).
Vanilla's runBlockEvents broadcasts a handled event to the clients, whose
own level re-runs triggerEvent (PistonBaseBlock.moveBlocks) to build the
PistonMovingBlockEntity it then ticks and renders interpolated. Do the
same:
- run_block_events emits a BlockEntityEvent for a handled event; the LAN
and integrated hubs send it as ClientboundBlockEventPacket, before the
tick's block changes, as vanilla's broadcast precedes
ChunkHolder.broadcastChanges.
- The remote client re-runs triggerEvent for pistons, note blocks and
bells on its own level (dropResources stays server-side, as
popResource checks !isClientSide), and ticks the resulting moving
pistons each client tick so they animate and collide between the
server's states. Block events a client-side move queues run right
away, as ClientLevel.blockEvent does.
Note blocks and bell rings reach the guests through the same packet,
which they did not get before.
Co-Authored-By: Devin <noreply@cognition.ai>
🤖 Agent: merged main (aec52c85e). pr_ready --quick passes (1308 horizon-client tests, clippy, 507 LOC). The welcome server mod e2e passes 35/35 in every mode: in-frame, server thread, LAN host in-frame and on its thread with a guest, dedicated with a guest. Verify is green; parity now runs on main (parity-scope). Queued.
linux-runner.sh takes its packages from apt where there is no Nix, the
registration token from RUNNER_TOKEN (a machine where gh must not be logged
in), and RUNNER_CPUS / RUNNER_MEMORY for a machine that does nothing else.
playhorizon, a partner's 8-thread server, joins parity_hosts.json.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Player.doSweepAttack keeps nearby entities with distanceToSqr < 9.0; it
doesn't use the entity interaction range.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: Gate check by Ludovic's AI agent (Claude). The server fix in this PR landed first in #1779 (levzzz's PR, same change to `hub.rs` and `agent.rs`, with the same `lan_bot` test). I merged main into the branch and kept main's side, so what this PR now adds is its two unit tests: `sprint::tests::sprint_survives_releasing_the_key_through_a_jump` (the client sprint state survives the key coming up on the jump tick) and `living_motion::tests::sprint_jump_adds_look_direction_impulse` (LivingEntity.jumpFromGround's 0.2 impulse along the look, and none for a walk). Both pass, and `lan_e2e --check sprint-fov` passes in-frame and on the server thread in the integration tree. Labelled skip-release, because players see no change. Thanks @mc070309 for tracking the cause down and for the tests.
🤖 Agent: Merging. verify and parity passed; the new unit test covers the merge of both key forms, and save_roundtrip no longer sees duplicate RecipesUsed keys.
bf81ae635 started saving recipes_used, but Horizon's recipe ids lack the
minecraft: namespace, so a furnace could hold both forms of one id and
vanilla refused it with "Duplicate entry for key" (save_roundtrip).
recipe_used_key gives the full id; loading merges the two forms from saves
written in between.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: Merging. verify passed, along with launcher tests (11 registry tests), the horizon-theme icon test and clippy -D warnings. #2173, which it was stacked on, is merged.
🤖 Agent: Merging. verify and mod-registry passed, and pr_ready --quick passed (clippy, horizon-mods and launcher tests). The index's downloads field is optional, so the schema doesn't change.
🤖 Agent: Merging. verify passed and the website tests (44) pass. Ratings appear once #2167's service is live; until then the pages show downloads only.
🤖 Agent: verify and parity pass (109 harnesses affected; parity green). The dedicated-server e2e (tools/quiet_commands_e2e.py) passes 7/7. Queued. #2171 (the mod API, stacked on it) follows once this lands.
🤖 Agent: contrib-gate-17 verified this head (26.3 sources, integration tree build, clippy, tests, parity; details in the PR comment). Thanks @lazy-importer!
🤖 Agent: contrib-gate-17 verified this head (26.3 sources, integration tree build, clippy, tests, parity; details in the PR comment). Thanks @levzzz5154!
Horizon results and project pages show a project's rating from the
registry service (GET playhorizon.dev/api/mods/ratings, fetched with the
results, kept ten minutes): five pixel stars, the mean and how many, or
"No ratings yet". Nothing shows when the service doesn't answer.
The launcher has no GitHub account, so a project's page offers "Sign in
with GitHub to rate": it opens the site's sign-in, which comes back to the
project's page and its stars (one rating per GitHub account).
HORIZON_MOD_RATINGS (a file or https:// address) points a test registry at
its own ratings. The star icon is drawn for Horizon on the icons' grid.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: contrib-gate-17 verified this head (26.3 sources, integration tree build, clippy, tests, parity; details in the PR comment). Thanks @Domanffe!
BookEditScreen's page is a MultiLineEditBox. Its getPreviousWord and
getNextWord split words at any whitespace (Character.isWhitespace, line breaks
included) and start from the cursor clamped to the last character. The
single-line EditBox.getWordPosition keeps looking at spaces only, so the line
break rule moves to the multiline box alone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: contrib-gate-17 verified this head (26.3 sources, integration tree build, clippy, tests, parity; details in the PR comment). Thanks @Domanffe!
Main's #1779 already keeps the sprint through PlayerInput (hub.rs, agent.rs)
and has the same lan_bot test, so this keeps main's side: what the branch adds
is the two unit tests.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The mesher now keeps a face's light emission in its colour's alpha byte. The
terrain shader drops it, but the native material path (merged since) passed
the colour through as the tint, so a light-emitting translucent face such as
a nether portal drew a little more transparent with HDR off.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: contrib-gate-17 verified this head (26.3 sources, integration tree build, clippy, tests, parity; details in the PR comment). Thanks @Domanffe!
Restore integer atlas origins and sample local texel centers before nearest filtering. Cover the packed stage origins with a GPU sampling regression.
Fixes#1882.
Co-Authored-By: Codex GPT-6.1 Sol <noreply@openai.com>
The registry's publish job (ludovic111/horizon-mods#7) writes each
project's `downloads` into index.json, summed from GitHub's counts of its
release assets every hour. `index::Project` reads it (absent: 0, never
written when 0, so older readers and the schema are unchanged).
The launcher's Horizon results show "12.3k downloads" like Modrinth's, and
list the most downloaded first.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Checked against the 26.3 bytecode (tools/rea_vanilla.py):
- Cushion.interact passes (PASS) when the cushion already has a rider;
CONSUME is only for a startRiding that fails.
- CushionItem.useOn calls destroyIfInFire right after addFreshEntity, and
posts GameEvent.ENTITY_PLACE from the player.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: contrib-gate-17 verified this head (26.3 sources, integration tree build, clippy, tests, parity; details in the PR comment). Thanks @Domanffe!
Hoglins now sense the nearest block from the hoglin repellents tag within 8 blocks sideways and 4 up or down. While one is near they become passive for 200 ticks (the attack target is dropped and no new one is picked), walk away from it, and avoid picking random walk spots within 8 blocks of it. Being hit ends the passive state.
CampfireBlock.useItemOn returns SUCCESS_SERVER (a swing) only when
placeFood took the item, and CONSUME (no swing) when all four slots are
taken.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A mod hands `run-quiet` a list of lines (at most 256 a call). In single
player they run as the player typing them would, with the player's source
made silent; on a Horizon server (it announces horizon:quiet_commands) they
go as one payload that the server runs the same way; on any other server
they fall back to chat commands, whose feedback shows. command.run is
unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: contrib-gate-17 verified this head (26.3 sources, integration tree build, clippy, tests, parity; details in the PR comment). Thanks @levzzz5154!
- The list shows each project's rating (stars, mean and count) and its
downloads ("12.3k downloads"), ordered by downloads by default; an Order
menu also sorts by rating or name (?sort= too).
- A project's page shows the same, then a rating control: five stars when
signed in (the account's own rating lit), "Sign in with GitHub to rate"
(coming back to the page) when not, a note for its owners.
- Ratings come from the registry service's GET /api/mods/ratings; a rating
is POSTed with the session's CSRF token. Until the service answers, the
pages show downloads only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: contrib-gate-17 verified this head (26.3 sources, integration tree build, clippy, tests, parity; details in the PR comment). Thanks @Domanffe!