affected_here called parity.affected_harnesses inside the filter, once for
each of the 343 harnesses. Since the selection reads the Rust items each
probe reaches (#1964) one call takes about a minute and a half, so
pr_ready sat for hours before its first step.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: verified: OUTPUT_VERSION = 1 and output-versions.txt carry the digest measured on 10 CI preparations. tools/prepare_output.py check passes on v0.2.64's preparation and fails on v0.2.47's. This head's own parity job prepares with its preparer and checks the digest. Merging.
Disk: clear_large_outputs only looked at .local directories named after the
harness, so the 174 harnesses writing to rust-<name>-parity kept every dump
(86 GB in one runner checkout, 13 GB for state_provider alone). It now
clears those too, keeps vanilla outputs only up to 256 MB (larger ones cost
more disk than rerunning their oracle), and handles shard names.
collision, world and geometry dropped one mkdtemp run directory per run
(11 GB): each run now removes the earlier ones.
Time: harnesses inherited the runner's stdin, a pipe on GitHub's runner, and
parity_shim never reuses an oracle run whose stdin is a pipe, so oracle reuse
never worked on CI. Harnesses now get /dev/null.
CI: parity.yml falls back to the head's own selection when the names main's
`--list` printed don't match its summary count.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
parity-main runs the vanilla harnesses on every push to main, one run at a
time: merges that land during a run wait in one pending run, which checks
every merge since main's last green parity-main (tools/parity_main.py base,
then parity.py --changed). A failure is bisected over the batch's
first-parent merges, running only the failing harnesses, and the breaking
merge's pull request gets a comment.
Game releases ship only a commit both checks and parity-main passed on
(release_train.py; release.yml also starts when parity-main finishes; a
forced release doesn't wait). The per-PR parity job now runs only for a PR
that changes a harness, an oracle, a probe or the runner, or carries
run-parity.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: merged on verify. It changes only which harnesses run, not the game; evidence in the body: all 61 failed parity runs of 10-18..10-25 are still caught, and LLVM coverage on 6 harnesses confirms the selection. Its own parity selects no harness. CI speed is Ludovic's priority tonight.
`--changed` / `--changed-files` now narrow each harness's recorded Rust
sources to those its probes can reach (tools/parity_reach.py): an item-level
reachability walk from the probe modules a harness ran (or the binary it
built) over free items, inherent methods, trait impls and macros, matched by
name so it only over-approximates. Parent modules count through their `mod`
declarations, compared with the merge base. A harness whose probe can't be
placed keeps every source; `--by-crate` turns the narrowing off.
On the last 20 PRs merged on 2026-10-25: 2255 -> 1061 harnesses picked,
2384 -> 1592 harness-minutes. Replaying the week's 61 failed CI parity runs,
every failed harness the old selection picked is still picked; LLVM coverage of
six harnesses' probes executes only kept files.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Launchers copy a player's prepared resources to a new release while its preparer declares the
same OUTPUT_VERSION (the launcher side is its own pull request). The number is bumped by hand;
CI keeps it honest: the parity job already prepares with every change to the preparer, and
tools/prepare_output.py check now fails when that preparation's digest is not the one
crates/horizon-prepare/output-versions.txt gives the current number, or when the file changed
other than by appending a version. The digest canonicalises what differs between two runs of
the same preparer (JSON key order, creative tab tag order, lambda class addresses, logs, the
sounds): 10 preparations by 5 preparer versions on the three Mac runners give one digest, and a Linux preparation by an older preparer gives the Mac runners' digest for it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
In stack_tooltip keep #1806's translate through `display_get` and this
PR's `push` helper for the styled tooltip lines.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
🤖 Agent: thank you! Merged after the contribution gate (checked against vanilla 26.3; one integration branch of main + its batch: clippy -D warnings, tests, the affected parity harnesses (111/112, all pass), in game on the server thread and in-frame, lan_e2e in both host modes) and this head's verify.
🤖 Agent: thank you! Merged after the contribution gate (checked against vanilla 26.3; one integration branch of main + all 14 of your PRs: fmt, clippy -D warnings, horizon-client/world tests, horizon-ctl scenarios on the server thread and in-frame, lan_e2e in both host modes, and the 30 harnesses covering these changes: 29 pass, the 30th is /help listing /pregen from #1805) and this head's verify.
🤖 Agent: thank you! Merged after the contribution gate (checked against vanilla 26.3; one integration branch of main + all 14 of your PRs: fmt, clippy -D warnings, horizon-client/world tests, horizon-ctl scenarios on the server thread and in-frame, lan_e2e in both host modes, and the 30 harnesses covering these changes: 29 pass, the 30th is /help listing /pregen from #1805) and this head's verify.
🤖 Agent: thank you! Merged after the contribution gate (checked against vanilla 26.3; one integration branch of main + all 14 of your PRs: fmt, clippy -D warnings, horizon-client/world tests, horizon-ctl scenarios on the server thread and in-frame, lan_e2e in both host modes, and the 30 harnesses covering these changes: 29 pass, the 30th is /help listing /pregen from #1805) and this head's verify.
🤖 Agent: thank you! Merged after the contribution gate (checked against vanilla 26.3; one integration branch of main + all 14 of your PRs: fmt, clippy -D warnings, horizon-client/world tests, horizon-ctl scenarios on the server thread and in-frame, lan_e2e in both host modes, and the 30 harnesses covering these changes: 29 pass, the 30th is /help listing /pregen from #1805) and this head's verify.
🤖 Agent: thank you! Merged after the contribution gate (checked against vanilla 26.3; one integration branch of main + all 14 of your PRs: fmt, clippy -D warnings, horizon-client/world tests, horizon-ctl scenarios on the server thread and in-frame, lan_e2e in both host modes, and the 30 harnesses covering these changes: 29 pass, the 30th is /help listing /pregen from #1805) and this head's verify.
A cow summoned with max_health and movement_speed bases, and a zombie
with an attack_damage base and a saved Health, against vanilla.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The three harnesses add commands to the same LocatorBarClient lines;
this branch now builds on #1847 (itself on #1843), keeps all three
commands, and test_cursor_death_parity.py runs through
locator.one_player.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Both add the same respawn command to LocatorBarClient; this branch now
builds on #1843's, and test_pearl_death_parity.py runs through
locator.one_player.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
test_hardcore_respawn_parity.py copied test_locator_bar_parity.py's
build and run code; both now share locator.sides, and one-player
scripts run through locator.one_player.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
🤖 Agent: thank you! Merged after the contribution gate (checked against vanilla 26.3, tested in the modes it touches; pr_ready on main + the six gate PRs: fmt, clippy, 1327 tests, 112/112 affected harnesses) and this head's verify.
Each item still to come gets a "Want it" button with its count; the Contribute
tab gets a bug report form (the #horizon-bugs template, which the telemetry
server opens as a GitHub issue and links back) and a suggestion form (to the
private queue Kira relays), and a Contributors list that shows the Discord name
of those who agreed on Kira (/contributors.json). The forms and buttons appear
once the hub's server answers; until then the page points to GitHub and Discord.
roadmap/README.md documents the intake API that Kira shares.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A stack held on the inventory cursor at death stayed there through the
respawn, even with keep_inventory off. In vanilla the dead player is
removed 20 ticks after dying (or when it respawns sooner), and
Player.remove's inventoryMenu.removed drops what the cursor held where
the player died. Do the same for server players and the in-frame
player, and add a harness that dies with a cursor stack on a dedicated
server and compares with vanilla.
With ender_pearls_vanish_on_death on, a pearl thrown just before death
kept flying and pulled the respawned player back to it. Vanilla's
ThrownEnderpearl.tick discards the pearl instead of ticking it while
its owner is dead. Add a harness that throws a pearl, dies and respawns
on a dedicated server and compares the teleports with vanilla's.
One entry per merged pull request (with its roadmap item), game or launcher
release, shipped roadmap item (post and announce) and weekly summary, newest
first, ids stable forever: the feed kira's src/journal.js posts to #updates and
X. The journal's item links are hub#<id>; the board's and the journal's ids are
reserved too, and the page-wide unique-id test covers every tab.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
/ride now looks up a mob's vehicle as well as the player's, so mobs
can mount mobs, get the already-riding and loop errors, and dismount
to the vehicle's dismount spot. execute on vehicle uses the same
lookup. The command_exec corpus gets a mob riding group.
A hardcore death offered another survival life. Vanilla's death screen
says Spectate in hardcore, and the server makes the respawned player a
spectator (ServerGamePacketListenerImpl.handleClientCommand). Do the
same on the server and in the in-frame world, and add a harness that
compares a hardcore respawn on a dedicated server with vanilla's.
Kanban: every open issue and pull request and what closed in the last seven days, placed in To do,
In progress, In review and Done by the rules of Kira's work board, each card with its roadmap item
(its roadmap:<id> label or the item linking it) and a filter by item. Journal: every release with its
changes grouped by roadmap item, the merges no release has yet, and a weekly summary from
roadmap/journal/<year>-<week>.md. The same journal is /hub/journal.json, for Kira's #updates and posts
elsewhere: each entry names the roadmap items it moved and shipped, with their 240-character text.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Kira (kira #10-#14) fetches https://playhorizon.dev/data/roadmap.json and links
items as <hub>#<id>: the JSON moves there (readable from any origin) with the
announce flag, and each item's anchor is its id. Ids the page uses itself (tabs,
sections, headings) are refused by the check; a test keeps every id of the page
unique, which caught the overview's headings sharing ids with the roadmap's.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Pass the crystal damage cause through destruction so explosion damage removes crystals without another blast, including during dragon respawn. Keep fight callbacks and direct-hit blasts.
Co-Authored-By: OpenAI Codex <noreply@openai.com>
playhorizon.dev/hub is the public entry point to what is happening on Horizon. Overview: where
Horizon is, the 3 to 5 things in focus, what is next, what shipped last, how to help. Roadmap: Now /
Next / Later / Help wanted / Shipped, filtered by area, every item with its issues and pull requests
in their GitHub state (baked at build, refreshed in the browser). Contribute: good first tasks,
claiming, connecting an agent to Kira, the rules. A Simple / Technical switch swaps every item's
plain and dev text and is remembered. /roadmap redirects to the Roadmap tab.
tools/roadmap.py build now also writes website/data/roadmap.json, which the site reads and serves at
/hub/roadmap.json.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Kira's X poster (kira #12) posts shipped items flagged announce in journal.json.
Set on the five whose release is news: modpack world generation, the mod
registry, Discord Rich Presence, Microsoft sign-in and CurseForge. Flags must be
booleans.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
One source of truth for what is being done, what comes next and where to help: roadmap/roadmap.toml,
each item said twice (plain words for players, the technical picture for contributors), Now / Next /
Later / Help wanted / Shipped, no dates. tools/roadmap.py builds ROADMAP.md from it, checks its rules
offline (CI, through tests/test_roadmap.py) and against GitHub (links resolve, no stale Now item),
ships items whose pull requests all went out in a release (sync), and keeps a roadmap:<id> label per
item (labels). README links it; roadmap/README.md is the maintenance guide.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Clear remote passenger mount and pose flags when a vehicle disappears, matching Entity.setRemoved. Add a two-passenger boat destruction check for both single-player paths and LAN observers.
Co-Authored-By: Codex (GPT-6) <noreply@openai.com>
Let PlayerCommand control the server sprint flag, matching vanilla, so releasing Ctrl cannot leave the client's speed modifier and FOV stuck. Cover the packet sequence and FOV recovery in regression tests.
Co-Authored-By: Codex (GPT-6) <noreply@openai.com>
Each frame the game hands the presence thread its status (menus; single
player, multiplayer or hosting a world with its party size; the
dimension; the game mode), and only a change wakes it. The switch
"Discord Rich Presence" (on by default, discordPresence in
horizon-extras.txt) sits beside Controller Settings. Dedicated servers
never start it. tools/discord_presence_e2e.py runs real games against a
fake Discord: title, world, Nether and quit on the server thread and
in-frame, a LAN host (both paths) and guest, the switch, and a Discord
started after the game.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
discord_presence.rs speaks Discord's local IPC itself (discord-ipc-N
socket or named pipe; opcode + length + JSON frames; HANDSHAKE, then
SET_ACTIVITY with the pid) from a thread of its own. The game hands it a
status; the thread connects lazily, retries every 15 s, keeps within 5
updates per 20 s, refreshes every minute and clears the activity on
quit. Inert while DISCORD_APP_ID is the placeholder, with
HORIZON_NO_DISCORD=1, and for automated runs unless
HORIZON_DISCORD_IPC_DIR points at a test's fake Discord; lab.game_env
sets HORIZON_NO_DISCORD=1.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Conflict in tools/lan_e2e.py's check list: main added shulker_bullets
beside crystals; both kept, crystals' entry aligned with the others.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ludovic Marie
Joansitoh
Kacpherek
rdbtCVS
levzzz