Registered jobs are third party callbacks and were invoked bare. An error inside
one propagates out of OnTime into Tick, so Tick never reaches its
SetTimeout(60000, Tick) at the end and nothing ever reschedules it. From that
moment cron is dead for every resource on the server - paychecks, cleanups,
anything registered through cron:runAt - until a restart. The operator sees one
stack trace and then silence.
Wrapping the callback keeps the loop and the reschedule intact, and names the job
that failed.
Simulated the scheduler chain in standalone Lua with three jobs where the first
raises: before, the tick died before SetTimeout and the two later jobs never ran
again; after, the failure is reported and both later jobs fire.
xLib.name was "xLib", which is not a resource, so the import metatable fed it to LoadResourceFile and resolved nothing. It failed silently on Legacy and hard-errored on Enhanced (Resource 'xLib' not found). Point it at the real resource name.
Adds xLib.pubsub for server-driven topic multicast: a producer resource
subscribes players to a topic server-side and publishes data that is pushed
only to the subscribed players. Clients listen with xLib.pubsub.on and never
subscribe or publish themselves.
The subscription registry is a single shared instance in the esx_lib server VM,
reached cross-resource through exports; the client side is a per-VM listener
over one net event. Subscriptions are cleaned up on playerDropped, so a reused
serverId never inherits them, and on the owning resource stopping.
Also adds xLib.triggerClientEvent, which packs the payload once when sending to
many players; publish uses it so a broadcast to N subscribers serialises once.
Adds a generic xLib.cache (ped, vehicle, seat, weapon, coords) modelled on
ox_lib cache and framework agnostic, and rewires es_extended to source ped and
weapon from it which removes the per-frame ped poll. The vehicle enter/exit
state machine and every esx: event are kept unchanged through a thin glue.
Moved getPlayersInArea / getClosestPlayer / getPedsInArea / getObjectsInArea /
getVehiclesInArea / getClosestPed / getClosestObject / getClosestVehicle to
xLib; es_extended re-exposes them through server compat shims.
Flagged: getNearbyPlayers iterates ESX.Players (es_extended server state),
which is not portable across the lib boundary - the lib relies on that global
being present at call time.
Integrated fix during the move:
- spawnVehicle uses the computed isNetworked for the network-id/migrate guard,
fixing the original that branched on the raw networked argument
modSmokeEnabled preserved as-is in get/setVehicleProperties (flagged: the set
path toggles mod 20 without applying a smoke color, matching legacy behavior).
Non-portable substitution documented:
- ESX.PlayerData.ped -> PlayerPedId() (ESX.PlayerData is not available in the lib VM)
Integrated fixes during the move:
- monotonic handle counter (handleCount) instead of count+1; no handle
collision after a point is removed
- unified single proximity loop: per-frame inside() plus a 500ms enter/leave
scan, adaptive Wait(next(insidePoints) and 0 or 500), replacing the previous
two separate loops
- emptiness detected via next() instead of # on a sparse-keyed table
Non-portable substitutions documented:
- ESX.PlayerData.ped -> PlayerPedId() (ESX.PlayerData is not available in the
lib VM)
- GetGameTimer() used to gate the 500ms scan within the single loop
Selt