This required some minor shared task refactoring
- Shared task expiration handling was moved to world. Previously this
was handled by zone and had clients self remove when a shared task
expired. This resulted in wrong messages when a task ended.
- FailTask is now implemented for shared tasks which previously only
made the client quit the shared task. It now fails the task for all
members by ending the shared task.
The `Client::EndSharedTask` api will end the client's current shared
task (removing all members). This is similiar to the `FailTask` api
except it doesn't require a `task_id` and the red task fail banner is
optional (live doesn't use it for shared tasks).
The global `end_dz_task` api was added for when a client context isn't
available. This will end a shared task if the current zone is a dynamic
zone for a shared task mission. Currently only shared tasks use dynamic
zones but this api can be expanded if support is added for tasks/quests.
The global `get_dz_task_id` was added to conveniently get the task id
for the current dynamic zone if it's used for a mission. Note this is
a database hit since that information is not available at zone level.
* Combine task_activity item and npc fields
This will make tooling easier.
While denormalizing goallists may not be ideal, it decouples tasks from
rewards which share the table and removes a redundant column in favor
of a using the delimited string which better matches live packet data.
* [Tasks] Deprecate goallists table, migrate reward goal lists, simplify logic
* Update 2022_09_25_task_concat_matchlists.sql
* Update 2022_09_25_task_concat_matchlists.sql
* Tweaks
* Fix reward column name in conversion script
* Task reward stacking
* Update task_client_state.cpp
* Implement stack counts
* Fix reward item instance memory leak
* Validate reward item instance
* Fix item reward message
* Fix findtask
Co-authored-by: Akkadius <akkadius1@gmail.com>
Some live npcs ignore the request cooldown timer (tutorialb)
A separate function had to be used for perl because the apis use an
array instead of array reference which won't allow a bool overload
This also replaces the fixed array and count args with a vector
* Add faction logging category
Probably should use this for more things
* Add FactionAssociation struct
This is simply just a struct that contains an array of faction ids and
multiplier. This can hold a maximum of 10 entries (Seru hit is 8, so 2
extra) this can be raised if need be.
* Add database changes and other data point changes
This is all the database changes and loading changes
Included is an optional SQL that will be used as a starting point, there
is likely errors or typos, but we will fix those as they are discovered.
* Add Client::RewardFaction function
This just takes the faction ID and the magnitude of the primary faction
hit and calculates the rest.
The minimum change will be either 1 or -1. We stop processing after we
see an ID of 0 and assume there will be no later entries.
The primary faction ID will always receive a hit even if there is no
faction association entries
* Add users of RewardFaction to NPC death, tasks, and QuestRewards
This will only use the new system if the magnitude is set, otherwise we
will just use the old system still
* Add quest system calls and lua QuestReward support
* Add #factionassociation command
This just calls RewardFaction, mostly useful for debugging
This allows removal of the task id limit (MAXTASKS)
There's some suspect places task data isn't verified but left unchanged
If memory use becomes too high once more live data is added tasks can be
stored in shared memory instead
The task goal system made implementing tasks a little confusing since
the goal could be ambiguous depending on type. This also didn't support
filtering on multiple goals (e.g. looting items from matching npc names
inside an area). Deliver types could specify an npc id in `delivertonpc`
but the database may have multiple npcs with the same name or a task
might want to match partial npc names.
This replaces goalids with explicit fields for npcs, items, proximity
areas, and touch switch ids. These changes make managing task data
easier without needing to update multiple tables and allows filtering
task updates by multiple criteria. To mitigate any performance impact
from merging task proximities, only clients with explore tasks in the
current zone are checked during client movement updates.
Items and npcs still support goallists but it would be possible to
denormalize entries into delimited strings to combine with the match
lists. This would also decouple task goals from reward lists.
The client task update functions were refactored to run through a single
filtering function which significantly reduces duplicated code from the
legacy task system. This will also make it easier to later implement
any unhandled types.
Since the new fields will handle filtering single entries and lists
based on having values set, `goalmethod` now only distinguishes quest
controlled from source controlled.
This is a breaking api change, `taskexploredarea` has been removed
since explore ids no longer exist.
Some live tasks make new elements available without requiring all
currently active ones to be completed first.
This adds the `req_activity_id` field to task activities which will mark
an element active if its required activity id is completed. If a valid
value is set then it's used instead of checking the current step.
The `step` field may still be set on rows with a valid `req_activity_id`
to specify its logical step and prevent later steps from becoming active
until completed. It's only ignored when deciding if the current element
is active.
The legacy task logic for unlocking activities was completely refactored
for this. A common method has been added so both zone and world can make
use of it to determine which elements are currently active. The previous
step system should remain unchanged.
The world logic for locking shared tasks when an element became active
did not account for "sequential" mode (all steps 0), unordered steps, or
gaps in step numbers. This also resolves that issue.
Functionally nothing should change in this patch. This refactors a lot
of formatting and code duplication in requests and member removals.
The long SharedTaskMessage namespace and its constants were contributing
to a lot of formatting that was difficult to read. These were shortened
since they're not all shared task specific anyway (though they should
probably just be moved to string_ids.h with other eqstrs now)
Shared task requests were refactored a little to remove an unnecessary
function that filled potential members. The separate functions used for
/taskquit and /taskremove were also combined into a single function to
remove a member while preserving live message differences.
* Add task reward points field
This replaces the separate DoN crystal reward fields with points and
point_type fields. This will make it easier to import data from
packet/client captures and possibly better support any future clients
or tasks that don't reward through the newer reward window.
* Fix manifest column check
If a member is offline (or possibly during a race while zoning?) when
the shared task is completed they will not receive the reward. On live
the character receives their reward (with an updated replay timer) if
they enter back into game while the shared task is still active. They
keep the original replay timer if the shared task is no longer active
and do not receive a reward.
This makes it so clients are issued rewards (and a task completed
event is dispatch) if the client's task state was out of sync with a
completed shared task. To prevent characters being rewarded more than
once in case of bad sync checks, a 'was_rewarded' field has been added
to the character_tasks table and updated when rewards are assigned.
This fixes a couple bugs so the character_activities table is correctly
updated with shared task states to better detect when out of sync:
- The character_activities table is now flagged to update after syncing
shared task states. This table was not being updated if a client was
offline or inaccessible for a shared task element update.
- The character_activities table is now updated when a task element is
completed. This was only being updated for activity increments and on
completing the entire task. SaveClientState is now called at the end
of ClientTaskState::IncrementDoneCount to cover all cases.
This also has a cosmetic change to show replay timers before rewards
like live, though this will not work for shared tasks until refactoring
world code
* Shared tasks WIP; lots of logging; shared tasks and tasks work internally the same for now; lots to cleanup yet
* Update task_manager.cpp
* Add tables
* World message handler
* Zone message handler
* More messaging
* More rearranging
* Task creation work (wip)
* Tweaks
* Decoupled things, added a shared task manager, moved logic to the manager, created the shared task object, now creating a sense of state on creation and members, zero validation, happy path
* Cleanup unnecessary getter
* More work on shared task persistence and state loading
* Add int64 support into repositories
* More state handling, creation loads all tables
* Wrap up shared task state creation and removal
* Move more lookup operations to preloading (memory). Restore shared task state during world bootup
* Implement shared task updates
* Add members other than just leader in task confirmations
* Update shared_task_manager.cpp
* Hook task cancellation for shared task removal (middleware)
* Remove dynamic_zone_id from SharedTasks model in repositories (for now) since we will likely be one to many with DZ objects
* Get members to show up in the window on creation
* Add opcodes, cleanup
* Add opcode handlers
* Split some methods out, self removal of shared task and updating members
* Implement offline shared task sync
* Style changes
* Send memberlist on initial login; implement remove player from shared task window
* Refactorings, cleanup
* Implement make leader in shared tasks window
* Implement add player, sync shared task state after add
* Add opcodes for remaining clients
* Shared task invite dialogue window implementation and response handling (including validation)
* Logging
* Remove comment
* Some cleanup
* Pass NPC context through shared task request logic
* Remove extra SharedTaskMember fields
* Add message constants
* Remove static
* Only use dz for expedition request
This passes expedition creation parameters through DynamicZone instead
of injecting ExpeditionRequest since it can hold creation data now
* Store expedition leader on dz
This shifts to using the leader object that exists in the core dynamic
zone object. It will be moved to the dynamic zone table later with other
columns that should just be on the dz to make loading easier.
Expeditions are probably the only dz type that will use this for window
updates and command auth. Other systems on live do fill the window but
don't keep it updated
* Store expedition name on dz
This uses the name stored on dz (for window packets) instead of
duplicating it. This will be moved completely to dz table later
* Store uuid on dynamic zone
This lets dynamic zones generate the uuid instead of expeditions. Other
dz type systems may want to make use of this. Lockouts should also be
moved to dynamic zones at some point in the future so this will be
necessary for that
* Move expedition db columns to dz
These columns should just belong to the core dynamic zone. This will
simplify loading from the database and in the future a separate
expedition table may no longer be necessary.
* Move window packet methods to dz
It makes more sense for these methods to be in the core
This will also allow support for other systems to use the window, though
live behavior that updates the window for shared task missions when not
in an expedition is likely unintended since it's not updated on changes.
* Store dynamic zone ids on clients
These will now be used for client dynamic zone lookups to remove
dependency on any dz type system caches
* Move member management to dz
This moves server messaging for adding and removing members to internal
dynamic zone methods
Set default dz member status to Unknown
* Move member status caching to dz
This moves world member status caching into internal dz methods
Zone member updates for created expeditions are now async and sent after
world replies with member statuses. Prior to this two memberlist packets
were sent to members in other zones on creation to update statuses.
This also fixes a bug with member statuses being wrong for offline raid
members in the zone that created an expedition. Note that live kicks
offline players out of raids so this is only to support emu behavior.
* Move member status updates to dz
* Set dz member status on all client dzs
This also renames the zone entry dz update method and moves window
update to a dynamic zone method. Eventually expedition components
should just be merged with dz and handled as another dz type
* Save instance safe return on characters
Add character_instance_safereturns table and repository
Previously dz safe return only worked for online characters via the dz
kicktimer or offline characters with a workaround that moved them when
an expedition was deleted. There were various edge cases that would
cause characters to be moved to bind instead (succoring after removal,
camping before kick timer, removed while offline, bulk kickplayers
removal with some offline)
This updates a character's instance safereturn every time they enter a
zone. If a character enters world in an instance that expired or are no
longer part of they'll be moved to their instance safereturn (if the
safereturn data is for the same zone-instance). Bind is still a fallback
This may also be used for non-dz instancing so it's named generically
This removes the expedition MoveMembersToSafeReturn workaround which
deprecates the is_current_member column of dynamic_zone_members and
will be removed in a followup patch.
* Remove is_current_member from dz members
This was only being used in the workaround to move past members to
dz safereturns if they were still inside the dz but not online
* Let dz check leader in world
This moves expedition leader processing in world to the dynamic zone.
This is a step in phasing out the separate expedition class for things
that can run off the dynamic zone core with simple dz type checks
This greatly simplifies checking leader on member and status changes
without needing callbacks. Other dz types that may use the dz leader
object can just handle it directly on the dz the same as expeditions
* Let dz handle member expire warnings
This moves expire warning checks to dz. This will make it easier for
other dz types to issue expire warnings if needed
* Use separate dynamic zone cache
Dynamic zones are no longer member objects of expeditions and have been
placed into their own cache. This was done so other dz types can be
cached without relying on their systems. Client and zone dz Lookups are
now independent of any system
This continues the process of phasing out a separate expedition cache.
Eventually expeditions can just be run directly as dynamic zones
internally with a few dz type checks.
Add dz serialization methods (cereal) for passing server dz creation
Modify #dz list to show cache and database separately. Also adds #dz
cache reload. This command will reload expeditions too since they
currently hold references to the dz in their own zone cache.
Add a dynamic zone processing class to world to process all types and
move expedition processing to it
* Move expedition makeleader processing to dz
* Let dz handle expedition deletions
This removes the need for separate expedition cache in world
This will greatly simplify world dynamic zone caching and processing.
Dynamic zones that are expeditions can just handle this directly. Once
lockouts and other components are completely moved to dynamic zones the
separate expedition cache in zone will also no longer be necessary
* Remove ExpeditionBase class
Since world no longer caches expeditions this will not be necessary
* Fix windows compile
* Implement task dz creation
Prototype dz creation for shared tasks
* Add and remove shared task members from dz
Also keep leader updated (used in choose zone window)
* Fix client crash on failed shared task
* Fix linux compile and warning
* Check client nullptr for dz message
This was accidently removed when expedition makeleader was moved
* Disable dz creation for solo tasks
* Add shared task repository headers to CMakeLists
* Add shared task dynamic zones table
* Add shared task dz database persistence
* Get members from db on shared task dz creation
This fixes a case where removing a member from a shared task dz would
fail if the member's name was empty. This could happen if the shared
task dz was created while a member was offline.
This also changes the dz member removal method to only check id. It
might be possible to change all dz member validations to only check
ids since names are primarily for window updates, but shared task dz
member names need to be non-empty anyway to support possible live-like
dz window usage in the future.
* Add character message methods to world
Add simple and eqstr message methods to ClientList
Add shared task manager methods to message all members or leader
* Add SyncClientSharedTaskState and nested sync strategies to cover M3 work
* Fix whitespace
* Implement task request cooldown timer
This implements the task request cooldown (15 seconds) that live uses
when a task is accepted. This will also need to be set when shared
tasks are offered (likely due to additional group/raid validations)
* Implement shared task selector validation
This implements the validation and filtering that occurs before the task
selection window is sent to a client for shared tasks
To keep things live-like, task selectors that contain a shared task will
be run through shared task validation and drop non-shared tasks. Live
doesn't mix types in task selections and this makes validation simpler.
Also note that live sends shared task selectors via a different opcode
than solo tasks but that has not been implemented yet
* Add separate shared task select opcodes
Live uses separate opcodes for solo and shared task selection windows
* Convert ActivityType to enum class
* Refactor task selector serialization
This adds serializer methods to task and task objective structs for the
task selection windows. This combines the duplicate task selector
methods to reduce code duplication and simplify serialization
* Add shared task selector
This sends shared task selection window using the shared task specific
opcode and adds an opcode handler for shared task accepts which are sent
by client in response to setting selection window to shared task type.
* Refactor task objective serialization
This adds a serialization method to the task objective struct for
serializing objectives in the window list and combines the separate
client-based methods to reduce duplicated code.
* Add task level spread and player count columns
* Implement shared task accept validation
This adds a common method for shared task character request queries
* Add task replay and request timer columns
* Add character task timers table
* Use shared task accept time on clients
This overrides client task accept time with shared task's creation time.
This is needed for accurate window task timers and lockout messages
especially for characters added to shared tasks post creation
* Implement task timer lockouts
This implements replay and request task timers for solo and shared tasks
* Add solo and shared task timer validation
* Remove logging of padding array
This gets interpreted as a c string which may not be null terminated
* Implement /kickplayers task
This also fixes current CancelTask behavior for leader which was
performing kickplayers functionality through the remove task button
* Implement /taskquit command
* Implement shared task invite validation
Remove active invitation before invite accept validation
* Remove local client db persistence during SyncClientSharedTaskRemoveLocalIfNotExists
* Add missing accept time arg to assign task
* Only validate non-zero task invite requirements
* Fix task error log crash
* Separate task cooldown timer messaging
* Use method to check for client shared task
* Avoid unneeded task invite validation query
Only need to query character data for levels for non-zero level spread
* Implement /tasktimers command
May want to add some type of throttled caching mechanism for this in
the future
* Add /tasktimers rate limiter
* Intercept shared task completion; more work to come
* Change SharedTaskActivityState and SharedTasks time objects to datetime
* Add updated_time updates to SharedTaskActivities
* Mark shared tasks as complete when all activities are completed
* Save a database query on shared task completion and use the active record in memory
* Don't record shared task completions to the quest log
* Implement RecordSharedTaskCompletion, add tables, repositories
* Update shared_task_manager.cpp
* Update shared_task_manager.cpp
* Add shared task replay timers
This is still not feature complete. On live any past members that ever
joined the shared task will receive a replay timer when it's completed
* Create FindCharactersInSharedTasks that searches through memory
* Remove namespace shorthand and formatting
* More minor cleanup
* Implement PurgeAllSharedTasks via #task command
* Add #task purgetimers
* Decrease m_keepalive time between processes
* Remove type ordering in /tasktimer query
* Add comment for task packet reward multiplier
This is likely a reward multiplier that changes text color based on
value to represent any scaled bonus or penalty
* Add replay timers to past members
This implements the live behavior that adds replay timers to any
previous member of a shared task. This likely exists to avoid possible
exploits.
Shared task member history is stored in memory and is used to assign
replay timers. This history will be lost on world crashes or restarts
but is simpler than saving past member state in database.
This also makes world send shared task replay timer messages since
past members need to be messaged now
* Move PurgeTaskTimers client method to tasks.cpp
* Remove dz members when purging shared tasks
Server dz states need to be updated before shared tasks are deleted
* Use exact name in shared task invites
This removes the wildcards from shared task invite character queries
which was sometimes selecting the wrong character
Taskadd validation is called even for invalid characters to allow for
proper messages to occur
* Clear declined active shared task invitations
This also notifies leader for declined shared task invites
* Store shared task member names
This adds back the character name field to SharedTaskMember. This should
make serialization easier in the future and reduce database lookups when
names are needed for /task commands
* Implement /taskplayerlist command
* Replace queries with member name lookups
Now that shared task members store names these queries are unnecessary
This also adds not-a-member messages for /taskremove and /taskmakeleader
* Implement shared task member change packet
This avoids sending the full member list to members when a single member
is added or removed and lets the client generate chat messages for it.
* Serialize shared task member list from world
This uses cereal to serialize the full member list from world and
removes the zone query workarounds
* Initialize client task state array
This was causing sql query errors on client state reloads
The client task information array was uninitialized resulting in being
filled with 0xcdcdcdcd values in msvc debug builds. Under release builds
this may have resulted in indeterminate values
A better fix would be to refactor some of this legacy code
* Add shared task command messages
Add messages for non-leader task commands
This adds taskadd, taskremove, taskmakeleader, and taskquit messages
The leader receives double messages for taskremove like live due to the
client generated message as well as the explicit one. It also receives
double server messages if the leader /taskremoves self.
* Replace some task messages with eqstrs
This also updates to use live colors
* Avoid shared task invite leader lookup query
Since member names are stored now this query is also unnecessary
* Avoid reloading client state on shared task accept
This was unnecessarily reloading client task state when added to a
shared task.
This also resulted in all active tasks being resent to shared task
members on creation. The shared task itself is the only task that
needs to be sent which is handled by AcceptNewTask.
* Remove active shared task invite on zone
Live doesn't re-send shared task invites after zoning like it does for
expeditions so there's no need to keep these around. This fixes active
invitations never getting reset on characters that zone or go offline.
* Choose new shared task leader if leader removed
* Add separate shared task kickplayers method
* Enable EVENT_CAST_ON for clients
This will be required for a shared task objective (The Creator) in DoN
* Revert "Avoid reloading client state on shared task accept"
This reverts commit 3af14fee2de8b109ffb6c2b2fc67731e1531a665.
Without this clients added to a task after some objectives have been
completed don't get updated state. Will need to investigate this later
* Disallow looting inside a dz by non-members
Non-members of a dynamic zone should not be allowed to loot npcs inside
it. This should have been disabled for expeditions already but was still
allowed due to an oversight (or live behavior changed). This is less
critical for shared tasks since members can be added and removed at will
without leaving a dz but still an important feature.
* Change load where criteria
* Increase task completion emote column size
* Use eqstr for task item reward message
* Implement radiant and ebon crystal rewards
This adds reward columns for radiant and ebon crystals to the tasks
table and updates task description serialization
* Send task completion emote before rewards
This matches live and makes it a little easier to see item rewards when
tasks have a long completion emote. This also changes it to send via the
same normal message opcode that live uses.
* Do not send a shared task in completed task history
* Allow EVENT_TASK_STAGE_COMPLETE for quest goals
This invokes event_task_stage_complete for task elements flagged with a
quest controlled goal method. It should be expected behavior that a
completed task stage always fires this event even if a quest controls it
* Add SyncSharedTaskZoneClientDoneCountState
* Swap return for continue in this case
* Formatting
* Simplify
* Formatting
* Formatting
* Formatting
* Remove errant check
* Formatting, add setter for shared tasks
* Remove debugging
* Comments in PR
* More PR follow up
* Formatting
* Cleanup
* Update packet comments
* Comments
* More cleanup
* Send command error message if not in shared task
/taskadd is the only command with this feedback on live. Newer live
clients also generate this instead of the server sending the message
* Implement expire_time on SharedTask object and add a purge on world bootup
* Comment
* Add SyncClientSharedTaskStateToLocal where clients fall out of sync and no longer have a task locally
* Clamp shared task activity updates to max done count and discard updates out of bounds
* Fix packet send
* Revert packet send
* Adjust clamping OOO for completed time check. Add completed tables to purge truncation
* Refactor kill update logic so that shared task kill updates only update one client instead of all clients
* Cleanup how we're checking for active tasks
* Forward task sets that contain shared tasks
This forwards task sets that contain a shared task to shared task
selector validation like normal task selectors
* Change eqstr for empty solo task offers
This is the message live appears to use if all task offers are filtered
out by solo task validation
* Fix max active tasks client message
This message starts at the third argument. It was maybe intended to be
an npc say message but live just sends it as a normal eqstr with the
first two arguments nulled.
* Load client task state after zoning complete
This fixes a possible race where a character removed from a shared task
while zoning would be stuck with an incorrect character activities state
after zoning was completed.
This was caused by the character loading task state to early on zone
entry but never receiving the remove player message from world since
they are missing from the world cle until zoning is completed.
Loading client state after zone connection is completed makes sure the
client has the latest state and available to the world cle
* Send message to clients removed while zoning
This message should usually only be sent to characters that were
removed from a shared task while zoning but will occur for any sync
state removals where a message wouldn't have already occured.
* Post rebase fix
* HG comment for checking active task
* Addressing HG comments around zeroing out a shared task id
* Remove errant comment
* Post rebase database manifest updates
* Update eqemu_logsys_log_aliases.h
* More rebase catches
* Bump database version for last commit
Co-authored-by: hg <4683435+hgtw@users.noreply.github.com>
