The die request, the smash request, the loaded notice, chat commands
and the racing pickup notice were handled for whichever object they
named. They are now only handled for the player that sent them, or for
the vehicle that player drives where the client sends them for it. A
racing pickup also has to be an object of a race track that is not a
player, a vehicle or the zone's own object.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
Selling paid for the count in the message while the move stopped at
what the stack held. The count now has to be between 1 and the stack's
count, the item can not be one the vendor already holds for buy back,
and the coins and alternate currency are worked out from the number of
items that left the inventory.
Buying back needs an item in the buy back inventory and a count the
stack holds. Both prices are worked out as 64 bit numbers.
Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
The save is stored as the client sent it in the new ugc_unsplit table, and each model split from it points back to it (ugc.unsplit_id), so the whole build can be looked at or restored later. A save is kept while any model split from it exists: deleting models removes saves no model points to any more, and deleting a character removes its saves.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When a brick by brick save is split into models, each model was centred on
the middle of its bones' origins, on the lowest origin. A brick's origin is at
a corner of it, not its middle, so a single 2x2 came back offset by half a
brick, and the model's floor was taken from the origins rather than from the
bricks.
Now each model is placed as the live server placed it: at the bottom middle
of the box around its bricks. Each brick's own box comes from the client's
brickdb by its design ID and is turned and moved by its bone (all 8 corners),
and the model's box holds every brick's. Its x and z middle and its lowest y
are the new origin, still snapped to the 0.8 grid on x and z, and every bone
moves by that; rigids and rotations are untouched. A design the brickdb has no
box for counts its bones' origins, as before, so no brick is left out.
Lxfml::NormalizePosition and Split take the boxes by design ID, and the brick
by brick save passes the ones read from the client (read on the first save,
then kept).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The dashboard's copy of the terrain reader in dCommon is gone; main's in dZoneManager reads for everyone, and keeps
what the copy had over it: colour and texture maps from before version 32 become RGBA, cut to the resolution the
client keeps. The scene map (ZoneScenes) and the dCommon tests compile against that reader.
A scene ID takes its layer as the zone file numbers it (0 General, 1 Audio) as well as the enum, so main's scene tools
and tests build; /getscene prints the layer as a number.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Main's model behaviors (every rotation, DoDamage, OnTimer, OnStartup, MoveBackToStart, moves along the model's
local axes), raw terrain reading with scene lookups and scene tools, property model and brick by brick fixes and the
chat server's player count fix, merged onto this branch's structures:
- game messages stay structs; the angular velocity messages join them and the handler sets a message's target
- strips own their speed again and the block names, moves and rotations stay in the shared action table
(PropertyBehaviorActions, which the dashboard's behavior player reads; it now lists rotations too)
- the zone keeps reading its file through ZoneFile and takes the boundaries from it, then loads the raw terrain as main
does; the old dNavigation terrain reader is gone
- the show all response stays a struct and logs a player count mismatch as main does
- the property management component tells models who sent a chat message
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* feat: raw terrain parsing for scene data
Replace old dNavigation/dTerrain raw parser with new Raw module in
dZoneManager. Parse heightmaps, color maps, and scene maps from .raw
files to determine which scene a position belongs to. Build scene
adjacency graph from terrain data and scene transitions.
Adds NiColor type, SceneColor lookup table, eSceneType enum, terrain
mesh generation with OBJ export, and debug slash commands for scene
visualization.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor: deduplicate terrain grid lookup logic, remove unnecessary cast
Add IsValidForSceneLookup(), GetSceneIDAtGrid(), and GridToWorldPos()
to Raw::Chunk. Use them in GenerateTerrainMesh, GetSceneIDFromPosition,
SpawnScenePoints, and SpawnAllScenePoints instead of duplicated inline
math. Remove unnecessary static_cast in LoadSceneTransitionInfo.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix: add version < 32 support to raw terrain parser
The parser was only correct for v32 files. For v<32:
- colorMapResolution is width-1, not width
- Color map reads width*width*4 bytes (BGRA per-pixel), not colorMapRes^2*4
- No DDS lightmap
- No texture settings byte or blend DDS
- Scene map: v31 reads (colorMapRes+1)^2 cells with border skip,
v<31 skips 1 byte
- No mesh data
Verified against all 309 .raw files across multiple client versions.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix: log and reject legacy raw terrain format, clean up ReadRaw API
Log a clear message and throw if the zone uses a legacy raw format
(LUZ version < 30) since no known maps use it. Remove ReadLegacyRaw
and the mapVersion parameter from ReadRaw.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix: revert v33 scene field renames to unknown1/unknown2
These fields are skipped by the client in every known build (ptr += 12
and ptr += 4). No client ever reads them into a struct. Keep the
original unknown1/unknown2 names from main.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix: load the zone without scene lookups when its raw terrain is missing or unreadable
Log at debug when it parses; build the terrain mesh only for the OBJ export; read the zone name and description only for version 31 and later, as the client does.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix: look up the scene at a position as the client does, in constant time
Chunk bounds span (width - 1) * scale; the nearest scene map cell is used and the far edge is kept one cell inside. Chunks are indexed by grid cell when they form a grid. Skipping reads instead of seeking, so a cut-off file can't seek past the end. Terrain mesh indices follow the vertex order.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* feat: scene point commands take a step and use brick LOTs from the BrickIDTable
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* chore: rename NiColor::ToXYZ to Luminance
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test: raw terrain reading, scene lookup, scene graph and scene commands
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* test: close the OBJ file before removing it so the test passes on Windows
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix: keep the zone boundary lines and read the luz scene count as the client does
Each boundary line's destination is read as a map ID and an instance ID into a LWOZONEID. The scene count is a byte before version 37 and a u32 from it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix: name the fourth flair colour byte padding
The client never sets or reads it, so it holds leftover memory.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix: keep the scene map of version 31 and older raws
These files have a (resolution + 1)^2 scene map, kept in sceneMapLegacy. Version 31 uses its first resolution rows and columns; older versions are all scene 0, as in the client.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* refactor: simplify the raw heightmap size check
The dimensions are capped at kMaxResolution, so the product can't overflow.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* chore: remove the scene point spawn commands and send /getadjacentscenes as one message
The spawn commands were only for testing. /getscene and /getadjacentscenes stay.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Physics/, Activity/ (activity, scripted activity, shooting gallery,
quickbuild, minigame control, racing control base, racing control, gate rush),
Vendor/, Character/ (what the client makes with a character, plus ghost,
minifig and mission), Sound/ and Property/; the rest stays where it was.
File moves only; DLU_COMPONENT_INCLUDE_DIRS gives every target the folders.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Minigame control (50) is an activity component in the client
(LWOMinigameComponent) and is made from the registry, so the zone control
object (LOT 2365, registry id 0) has one; it sends what an activity sends
(dirty, no players), not a fixed 0x40000000 (which nothing sent: the old
id > 0 check never made it). The zone's script on that object is server
only: the client makes no script component for it, so it sends nothing.
DLU's zone control construction now equals live's byte for byte (209
captures, e.g. GruntMonkey world2); the decoder reads 50 as an activity.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
CaptureProperty::Build reads, per world of a capture, the property it was:
each model (constructed with a model component, or a brick built model) with
its LOT, object, spawner, blueprint and behavior count and where it stood per
time span, from its constructions, serializations and destruction; the
DownloadPropertyData of that map and the GetModelsOnProperty counts; and the
events: placed (a PlaceModelResponse with the model made at that position,
before or after it), moved, removed (with the DeleteModelFromClient reason),
editing, and behavior messages.
CaptureTool property <bundle> --cdserver prints it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A compressed construction config is a u32 uncompressed size, a u32 compressed
size and the zlib bytes: the reader took the first size for the second and lost
its place, so live model constructions never read. It now inflates and reads
the entries.
A model's item component isn't made (Entity::Initialize): its model component
writes the item info. The layout left the registry's item component in, so a
premade model's components read one block twice and didn't match. Live writes
the same bits from its item component, so both now read exactly.
Tested with a live placed model's construction and the server's own model.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every live capture decoded: 83 of 60,659 constructions and 5 of 609,919
serializations now don't read exactly (was about 1 in 10 constructions, and
the decode aborted on text that isn't UTF-8).
- trigger component (header trigger bit): a bit and the trigger ID, last
- BBB component on objects listing component 107 (characters)
- local space info, buff immunities, phantom physics distance, skills in
progress, a choice build's setting
- moving platform: path when dirty, then subcomponents while a 1 bit comes
before one (mover and simple mover)
- pet: names under the dirty bit, on updates too; no item or model component
- models: the plain model block, or the mutable one (behaviors, editing
info) when the config has propertyObjectID or inInventory; the UGC block
is the item's
- zone file variants: markedAsPhantom (phantom physics) and renderDisabled
(no render data)
- compressed LDF: u32 uncompressed and compressed sizes, inflated and shown
as entries
- narrow text that isn't UTF-8 is shown byte by byte, and the capture tool
prints invalid text with replacements instead of aborting
Tests use live byte samples for each.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The other item sources live marked in AddItemToInventoryClientSync's extra
info, with the keys and values live sent:
- mission and achievement rewards: _Metric_Mission_ID_Int and source LOT 1,
the player (225 of 225 live rewards)
- activity rewards: _Metric_Activity_ID_Int and the activity object's LOT
- vendor purchases: the vendor's LOT and _Metric_Currency_Delta_Int, the coins
paid as a negative number (left out when the item costs no coins)
- mail attachments: _Metric_Mail_ID_Int64
- traded items: _Metric_Transaction_ID_Int64, the trade's ID
Loot::GiveLoot gains overloads that pass them on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Live's AddItemToInventoryClientSync for a picked up drop had
_Metric_Souce_LOT_Int in its extra info on all 2,344 live pickups: the LOT
of the DropClientLoot's source object, 1 when a player was the source
(activity rewards and chests, which drop from the player). DLU sent none.
The drop remembers its source object's LOT when it is registered for the
player, and the pickup sends it. A source object already gone when it drops sends
no key. Package contents keep sending none, as live's did.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ReplicaDecoder mirrors Entity::WriteBaseReplicaData and each component's Serialize, keeps what each network ID is
per world instance, and shows the rest as bits when no layout fits. CaptureTool links the game: game messages are
decoded (and compared in replays), and decode --cdserver reads replica packets.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Live filled a NewMail notice with the mail it was about: the mail ID, the
player, the attachment (LOT -1 without one) and a count of 1, and at load sent
one notice per unread mail. DLU sent one notice with every field 0 and the
unread total as the count. The answer to NotificationRequest names the player
and no mail, with LOT -1 and the unread count, as live's did.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Staff arm a packet capture on the dashboard; master passes MESSAGE_CAPTURE_CONTROL ARM to
every world, auth and chat and arms its own. Each server's PacketCapture tap (dServer receive,
and a send hook in RakPeer::Send so replica constructions are seen too) records into one
preallocated chunk per server and ships sealed chunks through master on the main loop when
capture_flush_bytes or capture_flush_interval_ms is reached; past capture_buffer_max_mb the
oldest chunks are dropped and the dashboard records a gap. Nothing is armed: one flag check.
- targets: an account (from its login; packets before the login are kept per connection
and added once auth or the world knows whose they are), a character (from when it is
picked), or everything; up to 8 at once (a bit each in the record mask)
- worlds and auth record their clients' packets and the master link messages of a captured
player (session keys by name, zone transfers by request, player added/removed, migration);
chat finds the player in each packet; master records server traffic for everything
- secrets are never recorded: structs that carry them (login request, login response user
key, world validation session key, session key messages between servers) are read,
blanked and written again before recording; auth keeps only the handshake and login
- PacketDecoder: a registry by service and message id names every packet and decodes the
registered structs; CaptureBundle is the file format (DLUBNDL1, metadata, records);
CaptureTools orders records on one timeline, pulls movement out, makes bundles portable
or anonymous and diffs replays
- the dashboard keeps packet captures in message_capture_sessions (capture_kind 1) and
their packets in a file under capture_dir, one write per batch; arming is audited
- MESSAGE_CAPTURE_CONTROL/DATA only gain appended enum values and trailing fields
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Every server now splits its packet counts by peer: its own connections
(players on auth and worlds), its master link, or other servers (the
worlds on chat, every server on master, a world's chat link, which is
now counted too). HTTP requests carrying X-Darkflame-Server count as
another server's, the dashboard counts its own requests to the UGC
server, and each HTTP client address is counted. The report also gets
each RakNet connection's statistics (worlds name the player on it),
trimmed to the 32 busiest with the rest summed. Counting stays on the
main loop, except the dashboard's UGC fetches, which only touch the
locked recorder.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Make again gets a Made by picker, /api/ugc/options lists the processor
choices and default, the settings page has the toolbox_* settings, and
/reprocessproperty takes native or toolbox-blender.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A quickbuild's HasItem preconditions took their items only when the build
completed and never gave them back. Live took them when the build started
and gave them back when it was cancelled: the FV Stone Warrior pedestal
(LOT 8551, precondition 99: 5 of LOT 6194) took the items at Building 5
times and added them back with the loot source Quickbuild on each of the 3
cancels in the live captures.
Preconditions now report their item costs instead of removing items while
checking. A quickbuild takes them at the start, gives them back on cancel
or a reset during the build, keeps them on completion, and gives them back
when the builder leaves the world mid-build.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Live sent quickbuild completion loot (about 190 samples) and the dragon,
BONS, spider queen and Frakjaw chest and wishing well rewards with the
player who earned them as the DropClientLoot source and owner,
use_position true and the spawn position at the object. DLU used the object
as the source, so use_position was false.
DropActivityLoot now uses the player as the source by default. The growing
flowers and the VE mission console have no live samples and keep the
object as the source.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ActivityRewards coins were always read from CurrencyTable npcminlevel 1.
Live used the reward row's ChallengeRating as the npcminlevel, and level 1
when the currency index has no row for that level:
- FV foot races (ChallengeRating 4, index 1) gave 36 and 48 coins, which
only level 4 (30-50) fits; level 1 is 3-5.
- Frakjaw's chest (activity 58, ChallengeRating 6) gave 250 each to a team
of 2: its currency indices 123-126 only have a level 6 row (500), so DLU
gave nothing.
- Quickbuilds, wishing wells and chests have ChallengeRating 1; survival
and the shooting galleries have ratings with no row and keep level 1.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
DLU sent a DropClientLoot for 0 coins with every drop whose coin range was
0, such as item-only smashables and scripted drops. None of the 3163 live
currency drops was for 0 coins.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Live sent a drop's currency DropClientLoot before its item drops in 1498 of
1502 live drop groups that had both. DLU sent the coins last.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The third ray backend, so every machine has a library for it: Embree on x86
CPUs (embree), HIPRT on AMD and NVIDIA GPUs (hiprt), Embree through SYCL on
Intel Arc and Xe GPUs (embree-gpu).
- CMake option DLU_EMBREE_SYCL (off). dUgcServer/EmbreeSycl is a project of
its own built by a SYCL compiler (DLU_SYCL_CXX; icpx through ONEAPI_ROOT or
the path, or the open source DPC++'s clang++ through DPCPP_ROOT) as an
external project: Embree 4.4 with EMBREE_SYCL_SUPPORT, linked statically and
bound inside (-Bsymbolic, only its C functions exported, so it never meets
the servers' own Embree), and the GPU kernels (nearest hit skipping a ray's
triangle, any hit), into libdlu_embree_sycl next to the servers
- UgcRaysEmbreeGpu loads it the first time embree-gpu is asked for; one GPU for
the process (embree_gpu_device picks it), the workers take turns, the
occlusion rays in batches as for hiprt
- without the build, the library or a supported Intel GPU it falls back to
embree and says why (the UGC server's log at start, --make-model on stderr)
- the option names, the settings page, the dashboard's picker, /reprocessproperty
Verified here: the default build and ctest; the SYCL build with the open source
DPC++ 7.1.0 (compiles, links against oneAPI's libsycl.so.9, exports only its C
functions); on this machine (no Intel GPU) it loads, finds no GPU and falls
back to embree. Not verified: tracing on an Intel GPU (none here).
Check: on a machine with an Intel Arc or Xe GPU and oneAPI, configure with
-DDLU_EMBREE_SYCL=ON and run UgcServer --make-model x.lxfml out embree-gpu;
the UGC tests then compare it with Embree.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Embree 4 (always built) replaces the UGC server's own bounding volume
hierarchies (the nearest-hit one and the occlusion rays' any-hit one), with no
fallback to them. ray_backend is embree (default) or hiprt (optional build;
Embree when it can't be used).
Settings, stored options and stats that say builtin still read: it is embree
(UgcRays::Parse, UgcProcessOptions::Parse). The dashboard's picker,
/reprocessproperty and --make-model offer embree and hiprt.
Tests: the backends are compared with Embree (hiprt when built), Embree against
rays whose hits are known, and the clutter's occlusion against what the old
hierarchy worked out (296 vertices summing to 114.5, 26 open, 183 dark); the
pinned model hashes are unchanged with Embree.
Check: ray_backend=builtin in an ini still starts and uses embree; the
settings page offers Embree and HIPRT.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The path tracer that imitated LU Toolbox's Remove Hidden Faces took 17 to 100
times as long as the renders from 42 directions around the model, too heavy to
use. It goes with its settings (hsr_method, hsr_samples, hsr_bounces,
hsr_sample_spacing, hsr_min_points), its side by side tracing for GPUs and its
tests. The renders (UgcRender::VisibleFromAround) remove hidden faces as before
that method existed; their size is hsr_resolution (1024), and the memory
estimate counts their buffers again.
The hidden-face method is no longer a processing option: the dashboard's picker,
/reprocessproperty and --make-model take only the ray backend (the occlusion's)
and denoising. Options stored before (made_options, process_options,
ugc_process_runs, --make-model arguments) that name toolbox or fast still
parse; the word is skipped.
Check: the settings page has no hsr_method or path settings and has
hsr_resolution; /reprocessproperty embree toolbox still works (toolbox
ignored); models made with the defaults keep their hashes (UGC tests).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Staff can make models again with other processing options than the UGC
settings' (ray backend, hidden-face method, denoising) and every make records
what made it, for comparing the options.
- migration mysql 99 / sqlite 82 (ugc_process_options): ugc.process_options
(picked for the next make, cleared once made), ugc.made_options (what made
the current files) and ugc_process_runs (every successful make: options,
wall and CPU time, hidden faces', occlusion's and icon's time, bricks,
triangles before and after)
- IUgc: ResetUgcModelProcessing and ResetPropertyUgcModelProcessing take the
options; PendingModel carries them; RecordUgcModelRun, GetUgcRunSummaries;
list entries have madeOptions and processOptions
- the UGC server applies a job's options over its settings and records the run
- /api/ugc/reprocess takes options ("embree fast oidn"); /api/ugc/options
lists the choices, the settings' defaults and averages per combination
- /reprocessproperty [builtin|embree|hiprt] [toolbox|fast] [off|oidn], any
order, all optional
Check: run the migration on MySQL and SQLite; /reprocessproperty embree fast
on a property, then the models' made_options and ugc_process_runs rows;
/api/ugc/options.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When staff approve or reject a character name while the player is online, the player gets a private announcement
popup ("Name approved" / "Name not approved") as well as the chat line. Players who are offline still aren't told
at their next login. Check: approve and reject a pending name while its player is online.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Members' clients read the chat text up to a terminating zero, as the client itself sends it; /g sent the text
without one, so the client read past it and showed extra characters ("test0>"). Check: guild chat shows exactly
what was typed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The world side of guilds (docs/Guilds.md):
- TMP_GUILD_CREATE, which the client's guild create box sends to its world, goes to the chat server as GUILD_CREATE.
- The character component reads the character's guild from the database when it loads (instead of the old unused "gn"
and "gid" charxml attributes) and takes GUILD_GET_STATUS from the chat server; a change is serialized once (the
client redraws the name billboard every time it reads a guild name). A name waiting for moderation isn't shown.
- Slash commands (the client has none for guilds): /g and /guild (guild chat; the client's guild tab sends /g, which
goes through the chat filter and mute like zone chat, then to the chat server as channel 10), /guildcreate (opens the
create box with DisplayGuildCreateBox), /gkick, /grank <name> <officer|veteran|recruit>, /gleader and
/gdisband confirm.
- The Guild Master script (LOT 3001, L_GUILD_CREATE.lua: using it opens the create box). Live never placed it; GMs
can spawn it.
Check in game (two accounts, a client with a FeatureGating row "guilds", 1, 0, 0 in its cdclient.fdb): /guildcreate,
make a guild (the name shows under yours, the guild button appears on the status bar); invite the second character
from the guild window, accept, both lists show both; guild chat tab; /grank, /gkick, /gleader; leave from the window;
log the second character out and in (guildmate logged off/in lines).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Connects GuildManager to the chat server (ChatGuilds): the client's GUILD_INVITE, GUILD_INVITE_RESPONSE, GUILD_LEAVE
and GUILD_GET_ALL (routed through its world), the worlds' GUILD_CREATE, GUILD_KICK, GUILD_SET_RANK and GUILD_DISBAND,
guild chat (GENERAL_CHAT_MESSAGE in channel 10, sent to the online members as channel 10 private chat, which the
client's guild tab shows, and logged as "guild"), guildmates told when a member logs in, out or changes worlds, and a
member's pending invite dropped when they log off. Client packets go through the member's world as WorldRoutePacket;
GUILD_GET_STATUS goes to the world itself.
Guild names: the chat filter's deny list refuses a name (dChatFilter::HasDenyList, since without a deny list the deny
check refuses everything); a name the allow list doesn't cover waits for moderation, as pet names do.
The dashboard's new player action GUILD_CHANGED goes from master to the chat server only (answered 1 when chat is
connected), which catches online members up. chatconfig.ini: guild_max_members (100) and guild_invite_timeout (600
seconds).
Check in game: nothing yet on its own (the world side makes guilds reachable). Server log: the chat server starts and
logs no unhandled guild packets.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Live pushed a mail NotificationResponse (NewMail, the unread count)
right after the respawn checkpoint in loads where the player had
unread mail, without the client asking (e.g. a new character's first
load: ServerDoneLoadingAllObjects, PlayerReachedRespawnCheckpoint,
NotifyMissionTask, MAIL). DLU had that call commented out, so the
mailbox icon only lit up once the client asked or new mail arrived.
It is sent only when there is unread mail (inferred: the 223 captured
loads without one are loads without unread mail; only 3 of 226 loads
carry it).
Check: send a character mail, log it out and back in without opening
the mailbox: the new-mail indicator shows right after loading; a
character with no unread mail shows none.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Live DropClientLoot (12,501 decoded): use_position is set only when a
player is the source (1,432, all activity rewards such as the dragon,
Frakjaw and BONS chests); drops from enemies, smashables and chests
leave it clear, so the client spawns the loot at the source object as
it sees it and falls back to spawn_position only if that object is
gone (LWODestroyableComponent::GetLootSpawnPos). Coins land where they
spawn (final = spawn in all 3,158 coin drops); items land 10 units
away (median 9.998, uniform direction) at spawn height. DLU always set
use_position and scattered everything 4.2 units away.
Inferred: the 10-unit rule for player-sourced items (live spread there
ranges 0-58 units, source unknown).
Check: kill enemies and smash crates: coins pop out of the object,
items fly about 10 units away; open an activity chest (e.g. after a
dragon or Frakjaw) and the loot still appears around you.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Live filled SetCurrency's source fields per source type (16,816 decoded
live SetCurrency): pickups carry the position the client picked the
coins up at (the one in its PickupCurrency) and no source; mission,
achievement and activity rewards name the player (object and LOT 1);
vendor buys, sells and buybacks name the vendor and its LOT; trades
carry the trade ID; death, mail and everything else name nothing. No
live SetCurrency sets loot_type. DLU wrote a present source LOT of 0,
no object and a zero position for all of them.
The position is the one the client uses: for pickups it projects it to
the screen as the start of the coin counter animation
(LWOCharacterComponent::SetCurrency), so it started from the world
origin before. PickupCurrency now reads that position (the client
always sends it). The trade ID is an 8-byte object ID in the client;
DLU declared it 4 bytes, which only never broke because it was never
set.
Check: pick up coins and watch the coin counter animate from where the
coins were; buy and sell at a vendor, finish a mission and trade coins
with another player, and the coin total updates each time.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
EventGating (eventName, date_start, date_end; 8 rows) is only in the
CDClient for the server: the 1.10.64 client has no string for it. The
dates are Unix times (UTC), inclusive: pirateDay 2011-09-19 00:00 to
2011-09-20 23:59:59 (Talk Like a Pirate Day), buildNexusTower 2010-03-09
to 2011-03-14, buildNexusTowerEnd, frostburgh2010_notused and test rows.
The live Lua asked for an event with GetHolidayEvent{eventToCheck =
name}.isValid; the only callers are the Crux Prime random spawners
(L_BASE_RANDOM_SPAWNER.lua, L_AM_ZONE_SERVER.lua), whose only event is
pirateDay. DLU's BaseRandomServer::CheckEvents was a TODO.
- CDEventGatingTable loads the table; HolidayEvents::IsActive(name) is
true while the dates include now, or when event_1..event_8 names the
event (every live date is in 2010/2011, so this is how a server turns
one on; the same settings already switch gatingOnFeature objects).
- BaseRandomServer::CheckEvents: during pirateDay the str and zip areas
use the event loads from the Lua (80%: 5 pirates on each of type1-3,
20%: 5 admirals each; multipliers secA 1, secB 1, secC 1.2 for str).
Inferred: every spawner's Lua put the whole {str, zip} table in its
zones, which its getRandomLoad walks with ipairs and so finds no load;
DLU uses each table for the area it names, as L_AM_ZONE_SERVER.lua's
layout intends.
- sharedconfig.ini says event_N also takes EventGating names.
Check in game: with event_2=pirateDay in sharedconfig.ini, the Crux Prime
areas filled by the str and zip random spawners (spawner networks
em_str_* and em_zip_*) spawn only Maelstrom pirates and admirals; the
other two areas and a server without the setting spawn the usual mix.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Adding or removing a grant sends the existing PLAYER_ACTION refresh through master to every world: REFRESH_ACCOUNT
for an account's grants, REFRESH_CHARACTER for a character's. A world with that account's sessions (or the loaded
character) drops the grants it kept, so the next command reads them again: no relog. The dashboard toasts whether the
player was online. The dashboard's own open pages already get the new rights on their next request, and their
WebSockets are checked again.
Check: with a character in game, grant it /spawn on the dashboard (toast: "Applied in game at once") and use /spawn
without relogging; remove the grant and /spawn is refused again at once. With the player offline the toast says it
applies when they next play.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A command may be used when the character's GM level allows it or a grant does: a grant of the command, of every
command up to a GM level, or of the dashboard permission the command follows (so accounts_kick covers /kick). A deny
of any of those takes it away even when the level allows it, except from GM 9 accounts (also while they play at a
lower level). Grants never take anyone below a command's floor above GM 1 (/execute), and commands the client handles
keep their fixed level. Expired grants count for nothing. The grants are the account's and the logged-in
character's, read from the database the first time a command is used and kept on the User. /help lists the commands
a player may use this way. The self and rank rules for commands that act on another player count grants of self_* and
manage_equal_rank too. A denied command says it was taken away.
Check: grant a GM 0 account /spawn (or "every command up to GM 8") on the dashboard, relog, and /spawn works and
shows in /help; deny /spawn from a GM 8 character: "it was taken away"; grant accounts_kick and /kick works (the rank
rules still apply to whom). dGameTests SlashCommandGrantsTest.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A client asks for a served model's HKX and is answered with the model's
LXFML, so it builds its own collision. That build also writes the
client's own .nif over the served one it downloaded and caches that
file's MD5 as the NIF's manifest info (client 1.10.64:
LWOBBBInterface::GenerateModelFromLxfml 0x00b6c220, MainThread_Process-
ModelResponse 0x00b5a1e0; valid entries never expire, 0x010186d0). On the
next load of the property the client used its cached checksum and file
without asking, and drew its own build instead of the served mesh.
Now, before the property's objects are constructed for a player, the
world sends them the served NIF checksum of every made model placed there
(UgcManifest::OnPropertyLoading). A client whose cached checksum is its
own build's downloads the served mesh again; one that has it already
downloads nothing. Nothing is sent unless ugc_manifest and
ugc_manifest_models are 1.
Check in game (ugc_manifest=1, ugc_manifest_models=1): visit a property
with made models, leave, come back in the same session (and after a
client restart): the models show the served mesh every time, and still
have collision.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
When a model finishes on the UGC server while players are on the property,
each client was sent the served NIF checksum, NotifyClientUGCModelReady and
the model constructed again at once. A client that was sent the model's
LXFML shortly before (the property load, a request for a model not made
yet, the owner's brick by brick save) is still building it then: the build
writes its own .nif over BrickModels/UserMade/<id>.nif and caches that
file's MD5 as the NIF's manifest info when it finishes (client 1.10.64:
LWOBBBInterface::GenerateModelFromLxfml 0x00b6c220, MainThread_Process-
ModelResponse 0x00b5a1e0), undoing the switch, so it kept its own build.
That is the usual case: another player's client asks for a model the owner
just placed, gets its LXFML, and that request makes the UGC server make the
model at once.
Now such a client is switched 30 seconds after the last LXFML sent to it
(UgcManifest::ServedMeshSwitches, BUILD_SETTLE); another LXFML sent
meanwhile starts the wait again. Other clients are switched at once, as
before. The served checksum is looked up when the switch happens. HKX
requests are still answered with the LXFML, so collision stays the
client's own.
Check in game (ugc_manifest=1, ugc_manifest_models=1, two accounts on one
property): the owner saves a new model; within about 30 s of the UGC
server making it, the other player's copy blinks out and back with the
served mesh (vertex AO, look of the dashboard's viewer), and still has
collision (walk into it). The owner's copy switches the same way. The
world log shows "Switching ... to the served mesh of model ...".
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With new binaries in place, master moves everything onto them while the
server keeps running (the dashboard's Live update, /liveupdate or SIGUSR2 to
master):
- database migrations of the new build first; a failure stops there
- UGC finishes the jobs it is running (queued rows stay pending), auth
restarts, chat hands its teams to master for the next chat server; master
starts the new processes and retries ones that don't come back
- once the new chat server is up (CHAT_SERVER_READY) every world connects at
once and sends its players again (LoginSessionNotify resync, no login logged)
- every world instance is replaced with an instance migration: public worlds
and private ones (same password) at once, properties after the old instance
saved and froze the property (MIGRATE_PREPARE: no building, claiming or
saving there any more), activity zones and character select once their
players left or after a wait; empty instances just stop, zones in
prestart_worlds get a new one first
- the dashboard restarts last and picks the status up again
Players land where they stood (position carried in CarriedPlayerState, also on
properties and Moon Base). Draining instances get no new players
(InstanceMigration::AcceptsNewPlayers) and show as "Moving players" in the
world list. The order lives in LiveUpdateMachine.h without master state and is
unit tested; master's glue is LiveUpdateCoordinator. Master itself is not
replaced. Message IDs are appended only.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A GM 8 command: every brick built model on the property you are on goes
back to the UGC server's queue. Once all are made (at most 15 minutes),
everyone on the property is sent the new mesh checksums and transferred
back into the same zone and clone, so their game loads the property again
and downloads the new meshes. Swapping a model's mesh in place doesn't
work in the client, so the whole property is loaded again.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
property_bff_build (off by default, as live): while the owner is in build
mode, their best friends can join it and place, move and pick up models,
build brick by brick and edit behaviors, each from their own inventory.
Only the owner starts build mode; a best friend still in it when the owner
leaves keeps building until they leave it.
The client lets only the property's owner edit, so each player now gets
their own DownloadPropertyData, with their own id as the owner while they
can build, sent again whenever that changes. SetBuildModeConfirmed and
GetModelsOnProperty go to each player instead of everyone. The first
builder makes the property private and pauses the models; the last one
restores them. Saving no longer needs the owner in the world.
A model taken off the property goes to whoever placed it; if they aren't
here it stays placed. Privacy and the property's name stay owner only
(neither was checked before).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Sending every model's LXFML at property load and then switching to the
served mesh (served checksum, NotifyClientUGCModelReady, the model
constructed again) left the client drawing its own build. Now made models'
LXFML is left out of the property load again: the client downloads and
draws the served mesh, and when it asks for the HKX it gets the LXFML,
builds the model and loads its own HKX, so the model has collision while
the mesh drawn stays the served one. The load-time switch is removed; the
switch for a model made again while players are there stays.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Switching characters keeps the connection (the client sends
CHARACTER_LIST_REQUEST to the world it's in), so the per-client UGC state
(waiting requests, pending mesh switches, the 3-switches-per-model cap)
carried over to the next character. It's now dropped as on a disconnect.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Taken down and constructed again in the same batch, the client dropped the
construction of the object it hadn't deleted yet, and most models
disappeared. The construction now follows a second later.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
