Profile version 2 with a progression block (v1 migrated with no progression), the core's
projection decision (hold) applied to a working copy, per-realm context storage (schema 8),
key-wise merge of build records with the canonical level and xp frozen, session pin in the
v2 messages, mid-session progression change ends the session under the old pin.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
OwnerService/HostService exchange only SessionOffer, PortableSessionStarted, PortableCheckpoint and
OwnerAck. Every checkpoint is merge3(C0, B0, current B1); sequences are monotonic and idempotent;
stale/superseded/malformed/oversized messages change nothing. Design and login-path audit in
docs/PORTABLE_SESSIONS.md.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Live stages drive real worldservers on both realms (A -> B -> play -> reconcile -> A -> B) with
independent expectations; portable_smoke example for the manual real-client test;
docs/PORTABLE_RECONCILE.md; reset script for the disposable realms.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
merge3(target, base, ours) applies only the base->ours delta; session reconcile is
(C0, B0, B1, lenient), in-place update is (realm now, synced snapshot, new canonical, strict).
Item/pet mappings gain presence (present/filtered/realm_local) so realm-filtered state is
told from player deletion. Updates are journaled like imports (marker + alloc row) and
recovered from the realm. Live tests cover filtered vs deleted items, checkpoints, pets,
conflicts, lost answers, atomic rollback and the A->B->A->B round trip without a server.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Imports a canonical snapshot into a stopped disposable realm through the existing
Db abstraction, in one MySQL transaction: guids allocated inside the transaction
from the realm's own tables and every reference remapped, orphan cleanup, name
and start-position handling, self-asserting result, and a marker row. Untrusted
strings and blobs reach SQL only as hex literals from a typed encoder.
Manager mappings are recorded only after the realm commit, in one local
transaction; an import journal (schema 3) plus the marker recover a crash
between the two commits. Preflight refuses wrong ruleset, running realm,
missing account/content and more before anything is written.
Stock character_talent rows are never written: a real worldserver asserts on a
non-talent row. Item identity no longer includes the crafter (v2).
Production import remains the designed in-core PortableImportService
(docs/PORTABLE_IMPORT_SERVICE.md); nothing of it is implemented.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
character_settings policy v2: the carry list is ten strict key patterns
(core.ascension_active_spec, _starter, _starter_live, _reset_credits,
_slot.active, _slot.#, _slot.#.build.#, _slot.#.bar.#, _build.#, _bar.#),
matched exactly, not by prefix. Spell charges, Destiny Weaver, dynamic XP,
Runemaster, prestige/glory, tutorial flags, Wildcard state and anything unknown
are quarantined in the opaque extension and never applied. The importer's own
marker source (coa.portable.) is dropped.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Reads one offline character out of a realm's characters database through the
existing Db abstraction: fixed SQL with the local guid as the only (numeric)
input, one consistent READ ONLY snapshot per character, hex-encoded text.
Maps the approved minimal v1 scope to the portable model; refuses online,
deleted, bot, active challenge/game mode/Manastorm and unknown-table cases.
Adds the 163-table registry, the settings policy, "Make Portable" in one store
transaction, and item mapping lifecycle with content identity (schema 2) so a
recycled local item guid never inherits an old portable item.
Recorded answers of a real MySQL drive the unit tests; live tests re-verify
them, read-only enforcement and snapshot consistency on a disposable server.
No realm database is written and nothing changes for existing installs.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Adds crates/coa-core/src/portable: UUIDv7 identities, namespaced content ids,
the v1 portable character model (validated, bounded, canonical), compressed
hashed snapshots, a union-only collection set, and a SQLite (WAL) store with
revisions, stale-revision rejection, rollback-as-new-revision, configurable
history, realm/item mappings and forward-only migrations.
No realm database, network or UI is touched; no existing behaviour changes.
45 new tests. Format documented in docs/PORTABLE_FORMAT.md.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Db runs mysql, mysqldump and mysqladmin inside the database container with
docker exec -i when the server folder is a Docker installation, connecting over
TCP on the container's loopback like the repack's commands do on the host. The
password still travels only in MYSQL_PWD. Every caller keeps using Db unchanged
(accounts, backups, population, migrations, the RA service account), and the
host needs no MySQL client. The repack path is untouched: one line before its
tools are looked up. A stopped container or a missing one is reported as a
database that is not running.
Tested: 4 new tests (commands built for the container, no password on a command
line, repack unchanged, error classification), and a real run against the
Docker folder: SQL queries, a quick backup (156 characters tables), restore with
the previous database kept, a damaged backup refused, the RA service account
created by SQL, the world server console reached through its published
loopback port, an account created through it and listed by the Manager.
A server folder with Settings/docker.json is a Docker installation: MySQL, the
auth server and the world server run in containers controlled through the
docker command line. Every other folder is a repack and goes through the
unchanged Windows code. driver::run, process::observe and layout::scan check
for the marker first and hand over to coa_core::docker.
- start: network, runtime image (libraries only, binaries are mounted from
Core/), database healthy, world listening, then auth. A container that stops
while starting reports the cause found in its log (DatabaseNotRunning,
PortInUse, ...). New error code DockerUnavailable (docker missing, stopped,
or not usable by this user).
- stop: world (180 s to save), auth, database; exit code 0 observed.
- observe: container state, health and port readiness mapped to the existing
ServiceState. Readiness is a connection that stays open, because Docker's
proxy accepts connections before the server inside listens.
- passwords go by environment, never on a command line; the remote console is
published on 127.0.0.1 only; the server's own database updater is turned off
(it needs a mysql client the image does not have and stops the server).
Tested: 18 new tests with an in-memory Docker, and a real run through
examples/drive: database, world (ready in 11 s) and auth started from a
package-shaped folder, stopped with exit code 0 and restarted.
Not included: installation, database setup, backups and SQL through the
container, updates, firewall/exposure, the client under Wine/Proton.
A server without a module's active file (an old repack updated by the Manager) made two modules log a
'missing property' line on every world tick.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
- main: set WEBKIT_DISABLE_DMABUF_RENDERER=1 on Linux unless the user did.
WebKitGTK's DMA-BUF renderer ended the process at startup with a Wayland
"Protocol error" (seen with an NVIDIA card on Wayland).
- data_dir: use $XDG_DATA_HOME (absolute only) or ~/.local/share on Linux
instead of the temp folder. Windows keeps %LOCALAPPDATA%.
- open_link: xdg-open on Linux, explorer.exe on Windows as before.
- tailscale_exe: look in the usual Linux locations.
Every Windows branch is the previous code unchanged.
- config snapshots: two snapshots of the same scope in the same millisecond
shared one folder. A restore snapshots the current state just before reading
the snapshot it restores, so on a fast disk the snapshot being restored was
overwritten with the current files and the restore did nothing. Each snapshot
now gets its own folder (numeric suffix when the name is taken). This was the
intermittent failure of snapshot_restore_roundtrip_is_byte_exact_and_undoable.
- package extract: restore the execute bit on ELF binaries and #! scripts (the
archive carries no modes). No effect on Windows.
- fsx::starts_with_ci: case-insensitive on Windows only; /srv/CoA and /srv/coa
are different folders on Linux.
- registry test: use a root path that is a root on the running OS.
cargo test --workspace on Linux: 157 passed, 0 failed.
A server whose module configs lack a key logs the same warning on every tick, which filled the 512 KB log tails of a
bug report and hid the actual failure. Read 8 MB back, fold the repeats, then keep the last part.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>