Commit Graph

2724 Commits

Author SHA1 Message Date
Justin Clark-Casey (justincc)
337ea019bd If a part has a sit target and an avatar is already sitting, allow another avatar to sit in the position given if no sit target was set.
Previous behave was that the second avatar could not sit.
This matches behaviour observed on the LL grid.
2012-07-10 23:55:22 +01:00
Justin Clark-Casey (justincc)
cdea572d2e refactor: move management of SOP.SitTargetAvatar into SOP.AddSittingAvatar() and SOP.RemoveSittingAvatar() 2012-07-10 23:50:04 +01:00
Justin Clark-Casey (justincc)
11e0ad6dc8 Revert "refactor: Add SOP.IsSitTargetOccupied to improve readability"
This reverts commit c8f0d476d2.
On reconsideration, I think this is less readable since immediately following code still sets SitTargetAvatar directly
2012-07-10 23:39:05 +01:00
Justin Clark-Casey (justincc)
c8f0d476d2 refactor: Add SOP.IsSitTargetOccupied to improve readability 2012-07-10 23:34:40 +01:00
Justin Clark-Casey (justincc)
69a6f6e3cd refactor: use sit orientation argument passed in to SP.SendSitResponse() rather than creating a new copy
There are no issues with side-effects since this is a struct.
2012-07-10 23:26:40 +01:00
Justin Clark-Casey (justincc)
e8347b7095 Move common code to detect whether a part has a valid sit target into a SOP property rather than being repeated in SP.
This also makes the detection in SP.FindNextAvailableSitTarget() and SendSitResponse() identical.
Previously they varied slightly (SendSitResponse didn't check for an older type of invalid quaternion) but the practical effect is most probably zero.
2012-07-10 23:21:39 +01:00
Justin Clark-Casey (justincc)
506437b684 Remove log line accidentally left in SP.SendSitResponse() 2012-07-10 23:06:34 +01:00
Justin Clark-Casey (justincc)
58869e5aa0 Fix recent SOP.GetSittingAvatars() to return null if there are no sitting avatars rather than throwing an exception.
Extends sitting avatar regression tests to test new sitters information
2012-07-10 23:03:52 +01:00
Justin Clark-Casey (justincc)
f3134b5cf6 When an attachment is detached to inv or derezzed, stop the scripts, update the known item with script state still in the script engine and then remove the scripts.
This is to fix a regression starting from 5301648 where attachments had to start being deleted before persistence in order to avoid race conditions with hud update threads.
2012-07-10 22:41:11 +01:00
Justin Clark-Casey (justincc)
d6f563794e Don't allow a prim to be sat upon if its part of an attachment 2012-07-09 21:43:44 +01:00
Justin Clark-Casey (justincc)
2eaa6d5ace Do not allow a script to attach a prim if its being sat upon.
This prevents a stack overflow where a get position on the avatar will refer to the attachment which will in turn refer back to the avatar.
This required recording of all sitting avatars on a prim which is done separately from recording the sit target avatar.
Recording HashSet is null if there are no sitting avatars in order to save memory.
2012-07-09 21:24:32 +01:00
Justin Clark-Casey (justincc)
7ff4eec79c Remove redundant SetScene() function in Scene.AddSceneObject()
This is always done later on in SceneGraph.AddSceneObject() if the call hasn't failed due to sanity checks.
There's no other purpose for this method to exist and it's dangerous/pointless to call in other conditions.
2012-07-07 00:02:45 +01:00
Justin Clark-Casey (justincc)
f1f390cfdf Remove now duplicate interregion object check that should have been removed a few commits ago in 43a2da9 2012-07-06 23:22:40 +01:00
Justin Clark-Casey (justincc)
056c9a59b2 Add assert to attachment regression tests to check that number of objects in the scene graph 2012-07-06 23:07:50 +01:00
Talun
1b1f841c6a Mantis 6063 osNpcTouch.
Allow NPCS to touch obects.
2012-07-06 22:37:19 +01:00
Justin Clark-Casey (justincc)
43a2da9edb Pull prim crossing/teleport checks up into Scene.IncomingCreateObject() from Scene.AddObject()
Only IncomingCreateObject() needs these checks.  General object adding does not need to perform crossing perm checks
2012-07-06 22:33:16 +01:00
Justin Clark-Casey (justincc)
951b45b80f Add OSSL function osForceAttachToAvatarFromInventory()
This works like osForceAttachToAvatar() but allows an object to be directly specified from the script object's inventory rather than forcing it to be rezzed in the scene first.
Still only attaches objects to the owner of the script.
This allows one to bypass the complicated co-ordination of first rezzing objects in the scene before attaching them.
Threat level high.
2012-07-05 00:05:06 +01:00
Justin Clark-Casey (justincc)
f7b4802577 Correct spelling mistake m_BanedViewers to m_BannedViewers 2012-07-03 23:26:02 +01:00
Justin Clark-Casey (justincc)
8183c2926d minor: Add some method doc to HasGroupChanged and Schedule GroupForFull/PartUpdate() to indicate when region modules need to invoke them 2012-07-03 23:19:11 +01:00
Justin Clark-Casey (justincc)
3c9b9a848f Fix issue in database tests where sogs being stored are not in a scene.
This puts an extra m_part.ParentGroup.Scene == null check at the top of SceneObjectPartInventory.QueryScriptStates()
2012-07-03 22:58:58 +01:00
Michelle Argus
3399596e0e Adds a list of viewers that are allowed or banned from the region.
Signed-off-by: Melanie <melanie@t-data.com>
2012-07-02 18:45:11 +01:00
Melanie
d32cf21576 Add preservation of running state of scripts when drag-copying. 2012-07-01 18:30:59 +01:00
Justin Clark-Casey (justincc)
a4551b027b Removing unused handling of incoming create object by userID and itemID only.
It appears this was never actually used since attachments were rezzed in other code.
This was never available on remote simulator comms, only local.
2012-06-30 01:14:49 +01:00
Justin Clark-Casey (justincc)
56c776066c Remove code listed for removal in 0.7.3 that handled script restart for incoming attachments from pre-fatpack regions (versions of OpenSimulator more than a year old) 2012-06-30 01:06:37 +01:00
Justin Clark-Casey (justincc)
f202c36106 Add IScene.Name for code clarity to replace the RegionInfo.RegionName used in many, many log messages. 2012-06-29 00:03:22 +01:00
Justin Clark-Casey (justincc)
bfa6896678 Change AttachmentsModule.DetachSingleAttachmentToInv() to accept a SOG directly instead of an item ID to then shuffle through attachments, saving CPU busywork.
Almost all callers already had the sog to hand.
Still checking that it's really an attachment, but now by inspecting SOG.AttachedAvatar
2012-06-28 23:31:23 +01:00
Justin Clark-Casey (justincc)
d043213317 refactor: Move ScenePresence <-> AgentData attachments copying code into AttachmentsModule. 2012-06-27 00:41:46 +01:00
Justin Clark-Casey (justincc)
e5b739aaeb When attachments are being saved and deleted for a closing root agent, delete first to avoid a hud race condition with update threads.
If delete doesn't occur first then the update thread can outrace the IsAttachment = false necessary to save attachments and send hud artifacts to other viewers.
2012-06-25 22:48:13 +01:00
Justin Clark-Casey (justincc)
5301648cff In AttachmentsModule.DetachSingleAttachmentToInvInternal(), remove attachment before changing properties for correct inventory serialization.
Serialization of attachments requires IsAttachment = false so that correct positions are serialized instead of avatar position.
However, doing this when a hud is still attached allows race conditions with update threads, resulting in hud artifacts on other viewers.
This change sets SOG.IsDeleted before serialization changes take place (IsDeleted itself is not a serialized property).
LLClientView then screens out any deleted SOGs before sending updates to viewers.
2012-06-25 21:08:19 +01:00
BlueWall
6d3ee8bb39 Fix script "Running" behavior
Unchecking "Running" box in script editor now persists. This fixes http://opensimulator.org/mantis/view.php?id=6057
2012-06-23 04:15:14 -04:00
Justin Clark-Casey (justincc)
fda39c11bf Fix bug where attachments would not retain position if just rotated and not moved.
This was because we were not setting AttachedPos in SOG.UpdateGroupPositionPR, unlike UpdateGroupPosition
2012-06-22 01:33:27 +01:00
Justin Clark-Casey (justincc)
68ea096f1b Use HasPrivateAttachmentPoint properties in SOG.DeleteGroupFromScene() instead of magic numbers 2012-06-20 23:25:07 +01:00
Justin Clark-Casey (justincc)
714db90832 refactor: use SOG.HasPrivateAttachmentPoint in SOP.SendTerseUpdateToClient() instead of attachmentpoint magic numbers. 2012-06-20 22:46:01 +01:00
Justin Clark-Casey (justincc)
0fa303b1cf Log how many scripts are candidates for starting and how many are actually started.
Adds DebugLevel infrastructure to XEngine though currently commented out and unused.
2012-06-20 00:10:19 +01:00
Justin Clark-Casey (justincc)
9737e6d52e If RegionReady is active, don't falsely say that logins are enabled in the main scene loop before RegionReady is signalled when initial script compilation finishes.
Also raises this logging level to Info from Debug since this information is of high importance.  This matches the behaviour of the RegionReady module
2012-06-20 00:07:03 +01:00
Justin Clark-Casey (justincc)
e23d7ff9c0 minor: If logging because mesh/sculpt data isn't present for an object, log object UUID rather than local id, since UUID doesn't potentially vary between simulator starts. 2012-06-20 00:05:48 +01:00
Justin Clark-Casey (justincc)
a4290048e5 Add SOG.HasPrivateAttachmentPoint to tell if a SOG has a private attachment point. HUDs attachment points are private.
Change SOP.SendFullUpdateToClient() and SoundModule.PlayAttachedSound() to use this rather than different magic number formulations.
This also corrects a bug in PlayAttachedSound() where the code assumed that all attachment points over 30 were HUDs.
It appears this is no longer true with Neck and Root (Avatar Center)
2012-06-14 01:20:55 +01:00
Justin Clark-Casey (justincc)
1b1f0a2d77 OnConnectionClosed listeners, retrieve data from IClientAPI.SceneAgent rather than scanning all scene for the presence with the right id
Stop checking IsLoggingOut on these listeners, if called with a root agent then we always want to perform these actions.
This covers cases where the client is closed due to manual kick, simulator shutdown, etc.
2012-06-12 02:43:33 +01:00
Justin Clark-Casey (justincc)
8c7149063b In PresenceDetector.OnConnectionClose(), use the IsChildAgent check already available on IClientAPI.SceneAgent rather than retrieving it again by scanning all scenes. 2012-06-12 01:25:09 +01:00
Justin Clark-Casey (justincc)
bab7dab4c5 Comment out the scene presence sitting debug log messages for now 2012-06-11 23:37:16 +01:00
Justin Clark-Casey (justincc)
30f4a33f01 Don't make duplicate call to ScenePresence.Close() separately in ETM.DoTeleport() if an agent needs closing.
This is always done as part of Scene.RemoveClient()
Also refactors try/catching in Scene.RemoveClient() to log NREs instead of silently discarding, since these are useful symptoms of problems.
2012-06-08 01:26:43 +01:00
Justin Clark-Casey (justincc)
5c162ccd57 Go back to calling IncomingCloseAgent() in the "kick user" command for consistency instead of IClientAPI.Close() directly.
This no longer double counts child agent removals
2012-06-08 00:59:39 +01:00
Justin Clark-Casey (justincc)
d547bcf8d1 Remove duplicate update of user count in Scene.IncomingCloseAgent()
This is already done in Scene.RemoveClient() which IncomingCloseAgent() always ends up calling.
2012-06-08 00:40:38 +01:00
Justin Clark-Casey (justincc)
b56673c920 Fix bug with "kick user" reducing agent counts by 2 instead of 1.
This is done by making the kick user command call IClientAPI.Close() rather than routing through Scene.IncomingCloseAgent(), which also called IClientAPI.Close()
DisableSimulator for child agents is moved from IncomingCloseAgent() to RemoveClient(), this is not a functional change since IncomingCloseAgent() always ends up calling RemoveClient()
2012-06-08 00:18:25 +01:00
Justin Clark-Casey (justincc)
0c5fefacb4 Record the fact that child agents can have asset transactions.
Also change code to grab the agent asset transaction module once.
2012-06-07 23:51:04 +01:00
Justin Clark-Casey (justincc)
de87e4871b Don't send kill object messages to clients when a child agent is closed. 2012-06-07 23:35:21 +01:00
Justin Clark-Casey (justincc)
96d1ba90d7 Scale down per frame MS stats to match scaled simulator FPS stat.
This makes frame time stats properly tally with fps, which saves confusion and makes it easier to interpret numbers.
In some ways this is not so artifical - physics FPS runs at the higher rate.
2012-06-05 00:27:51 +01:00
Justin Clark-Casey (justincc)
655625ab87 Start sending spare frame time MS viewer stat. Make frame time correctly display total frame time, not just non-spare time.
This makes it easier to see when components of frame time exceed normal permitted frame time.
Currently reflect scene frame times.
2012-06-05 00:17:55 +01:00
Justin Clark-Casey (justincc)
c3d9acc9a9 Merge branch 'master' of ssh://opensimulator.org/var/git/opensim 2012-06-05 00:15:29 +01:00
Justin Clark-Casey (justincc)
a7f4804f53 Properly show per frame millisecond statistics per frame, not as amount of time taken per second.
This is to make these statistics actually match their names (and also be more accurate as number of frames can vary under heavy load)
Currently using scene frames (11.23 every second) instead of physics frames (56.18 per second)
2012-06-04 23:07:53 +01:00