aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* Merge branch 'master' of melanie@opensimulator.org:/var/git/opensimMelanie2010-06-097-67/+63
|\
| * Refactor SendCoarseLocations for better performance. Instead of computing ↵Dan Lake2010-06-086-67/+59
| | | | | | | | list of all locations fresh for every scene presence on every frame, we will instead compute the list once every 50 frames and send to all connected presences at that time. Also, we only add 60 items to the list when there are more than 60 presences in the scene. For 1000 users, this change yields a 99.8% reduction in list processing and a 98% reduction in network bandwidth for coarse locations.
| * * Changed CreatorIDs consistently to varchar(128)Diva Canto2010-06-081-0/+4
| | | | | | | | | | * Deleted redundant migration for assets in SQLite * Rewrote XInventory migrations in SQLite in the new style
* | If a script is deleted before it gets compiled, don't even bother to tryMelanie2010-06-091-0/+19
|/ | | | compiling it
* Comment out the attachment update reordering mechanism in LLClientView since ↵Justin Clark-Casey (justincc)2010-06-081-31/+31
| | | | this doesn't appear necessary when the prim priorities are adjusted so that root prim updates are always sent before child prim updates.
* Adjust object update priorities such that root prims are sent to the viewer ↵Justin Clark-Casey (justincc)2010-06-081-6/+44
| | | | | | | | | before child prims. This was originally a fix for huds, since child prims fail to display if the viewer doesn't receive the root prim update first. However, on the advice of jhurliman, this has been done for all objects, both ordinary and attachments. The separate mechanism in LLClientView which prevents child prim updates being sent out first is still present temporarily. This is a foreport of the equivalent fix in 0.6.9-post-fixes, though that was for attachments only.
* Reduce number of full updates sent on region crossing for attachments/huds ↵Justin Clark-Casey (justincc)2010-06-084-12/+86
| | | | | | | | | to 1 from 3 This is one step towards reducing hud glitches on region crossing, since the viewer fails to display prims if it receives child full updates before the root prim full update This commit also introduces a mechanism in LLClientView to stop child attachment updates ever going out before the root one This is a very temporary mechanism and will be commented out when the next step of the fix (to give root prims higher udpate priority) is committed This code is a foreport from the equivalent changes in 0.6.9-post-fixes
* Add a call to SOG.ResumeScripts() after region crossing / teleportMelanie Thielker2010-06-071-0/+2
|
* * Bug fix in TP home: typo in unpacking of GridUserInfo.Diva Canto2010-06-052-0/+13
| | | | * Bug fix in TPs across neighboring regions: bug was introduced when getting rid of crashed sessions.
* Bug fix: update the list of friends upon MakeRoot, because child agents ↵Diva Canto2010-06-051-1/+1
| | | | don't get notified of new friendships.
* various refactorings on methods to replicate IAR paths to user inventory in ↵Justin Clark-Casey (justincc)2010-06-052-82/+128
| | | | order to improve readability
* refactor: do pathname trimming outside of ReplicateArchivePathToUserInventory()Justin Clark-Casey (justincc)2010-06-042-11/+8
|
* add ability for load iar/save iar to take in arbitrary optionsJustin Clark-Casey (justincc)2010-06-042-20/+76
| | | | not used for anything yet
* bring TestLoadIarV0_1AbsentUsers() back onlineJustin Clark-Casey (justincc)2010-06-041-63/+56
|
* bring TestIarV0_1WithEscapedChars() back onlineJustin Clark-Casey (justincc)2010-06-041-89/+72
|
* bring test TestLoadIarV0_1ExistingUsers() back onlineJustin Clark-Casey (justincc)2010-06-041-81/+76
|
* remove pointless mock user account service since the real one can now be ↵Justin Clark-Casey (justincc)2010-06-042-2/+2
| | | | easily configured for test purposes
* Merge branch 'master' of ssh://opensimulator.org/var/git/opensimJustin Clark-Casey (justincc)2010-06-043-29/+44
|\
| * Merge branch 'master' of ssh://diva@opensimulator.org/var/git/opensimDiva Canto2010-06-041-2/+2
| |\
| * | Fixed a couple of buglets in Friendship offers / acceptance / decline when ↵Diva Canto2010-06-043-29/+44
| | | | | | | | | | | | avies are in different sims.
* | | get TestSaveIarV0_1() working again by setting up an ↵Justin Clark-Casey (justincc)2010-06-042-3/+5
| |/ |/| | | | | | | | | OpenSim.Data.Null.UserAuthenticationData plugin additional tweaks to get this working properly
* | minor: comment out region interface registration log msg I accidentally left ↵Justin Clark-Casey (justincc)2010-06-041-2/+2
|/ | | | | | in last week also changes one log message to print out full exception stack trace on both mono/.net instead of just .net
* Add a forgotten security check. Make the new functions only work on primsMelanie Thielker2010-06-021-0/+6
| | | | owned by the host prim owner.
* Add two new osFunctions:Melanie Thielker2010-06-025-0/+50
| | | | | list osGetPrimititveParams(key prim, list rules); osSetPrimitiveParams(key prim, list rules);
* One should not lock null objects.Melanie Thielker2010-06-011-5/+8
|
* Lock the object queue when dequeueingMelanie2010-06-011-78/+81
|
* Comment and remove JScript support. Mono 2.7Dev and 2.8 no longer include theMelanie2010-06-011-19/+19
| | | | needed libraries
* Fix prim returns I broke earlierMelanie Thielker2010-06-011-1/+1
|
* Fix prim returns I broke earlierMelanie Thielker2010-06-011-4/+7
|
* Continuing refactor. Refactor DeRezObject to deal with multiple objectsMelanie Thielker2010-06-015-71/+118
|
* Add a method to get the bounding box and root prim offsets within it forMelanie Thielker2010-06-011-0/+44
| | | | a group of prims.
* Split GetAxisAlignedBoundingBox into two methods to allow calculation ofMelanie Thielker2010-06-011-3/+20
| | | | combined bounding boxes and offsets
* Change the handling of CreateSelected. Only send it on real creation, notMelanie Thielker2010-06-011-1/+2
| | | | for each prim coming into view.
* Fix create selection getting overwritten by multiple updates for the same prim.Melanie Thielker2010-05-315-6/+13
|
* Fix a nullref in EventManager caused by RegionReady not setting the sceneMelanie Thielker2010-05-301-0/+1
|
* Changes osFunction permissions again. Allow_ with a list of UUIDs now againMelanie Thielker2010-05-301-16/+51
| | | | | | refers to prim OWNERS. A new option set, Creators_, is added to allow selection by script creator. For existing installs, this means no functional change. The warning from my prior commit doesn't apply anymore.
* Changes OSSL Api permissions for the case of UUID list. In 0.6.9, the UUIDsMelanie Thielker2010-05-301-3/+18
| | | | | | | | | | | | | would be the IDs of the prim owners in whose prims these functions would run. This changes it so the UUID is the SCRIPT CREATOR instead. Further, osfunctions limited by uuid will not run if the creator and owner differ and the owner has mod rights on the script. There is still a danger in passing moodifiable scripts to others, as they can insert a harmful function, then remove the mod rights to make it runnable. As before, care needs to be taken, but where it was modable prims that were the risk before, modable scripts are the weak spot now. In cases where prim owner == script creator == script owner, nothing will change.
* get TestSaveIarV0_1() uncommented but not running as a test yet since I ↵Justin Clark-Casey (justincc)2010-05-282-106/+101
| | | | didn't get the authentication server to work and my brain is about to fizzle out my ears
* restore InventoryArchiverTests.TestReplicateArchivePathToUserInventory() to ↵Justin Clark-Casey (justincc)2010-05-281-51/+37
| | | | work with the new UserAccountService/InventoryService
* fission UserAccountService.HandleCreateUser() into two methods, one which ↵Justin Clark-Casey (justincc)2010-05-281-52/+50
| | | | handles user command parsing and another which actually does the work
* Add ability to load IARs directly from URIsJustin Clark-Casey (justincc)2010-05-283-7/+43
| | | | | | | | So, something like load iar Justin Clark-Casey / PASSWORD http://justincc.org/downloads/iars/my-great-items.iar Will load my IAR directly from the web.
* minor: remove mono compiler warningJustin Clark-Casey (justincc)2010-05-281-1/+1
|
* tidy up help information on load oar and save oarJustin Clark-Casey (justincc)2010-05-281-5/+11
|
* refactor: move GetStream and URI methods from ArchiveReadRequest -> ↵Justin Clark-Casey (justincc)2010-05-283-64/+65
| | | | ArchiveHelpers
* Adjust Scene.DeleteAllSceneObjects() to not delete objects attached to avatars.Justin Clark-Casey (justincc)2010-05-281-2/+6
| | | | | This is going to be the right behaviour in all cases, I should think. This means that avatars in region when an oar is loaded do not lose their attachments
* Simplify AddFullUpdateToAvatars()/AddPartialUpdateToAvatars() by calling the ↵Justin Clark-Casey (justincc)2010-05-281-2/+5
| | | | object's corresponding single avatar update method, rather than calling the sceneviewer directly
* remove redundant ScenePresence.QueuePartForUpdate() - every place in the ↵Justin Clark-Casey (justincc)2010-05-281-9/+0
| | | | code calls SceneViewer.QueuePartForUpdate() directly
* Prevent a null refMelanie Thielker2010-05-271-0/+3
|
* Revert "Comment noisy "CONNECTION DEBUGGING" messages, because they push more"Melanie2010-05-272-14/+2
| | | | | | Some other stuff snuck in. This reverts commit 4cc533e7ad94d148351c16f48afd2a688a64c48a.
* Comment noisy "CONNECTION DEBUGGING" messages, because they push moreMelanie Thielker2010-05-272-2/+14
| | | | important stuff off screen too fast