aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region (unfollow)
Commit message (Collapse)AuthorFilesLines
2010-10-03I really have to say it again and again: e.Message is NOT enough data forMelanie1-2/+1
troubleshooting. It MUST be combined with e.StackTrace, or e.ToString() must be used. Fix one more instance of that cr.p
2010-10-03Adapt to core changesMelanie1-1/+1
2010-10-03Addign the new / renamed files for previous commitMelanie8-0/+2655
2010-10-03Initial port of the Warp3D map tile rendererMelanie7-1292/+4
2010-10-02Attempt at fixing mantis #5057 -- CPU pinned high with HGAssetBroker.Diva Canto1-21/+34
2010-10-02So, the client can have an old idea of the object properties for the object ↵Melanie1-0/+24
when it goes to buy. This can cause a problem in the buy process. Additionally Hazim mentioned that the buy packets are spoofable. The core modules are the crowing glory example of best practice :P, so therefore, setting the example here, Validate Client sent Buy Data. WebAppSecurity 101, Never trust a client. Validate Validate Validate! Or you'll have problems whether intentional or not.
2010-10-02Reapplying the parts of the prior revert that were not derived from theMelanie3-5/+5
original patch
2010-10-02Revert "Forward-port 0.6 fix"Melanie3-17/+5
This reverts commit 90b51dc7d67507e27c4baa529e979de19dce8de1.
2010-10-02Replace CalculateMass with a more accurate version, contributed by Ubit.Melanie1-234/+219
Thank you.
2010-09-30Prevent dropping someone else's attachmentsMelanie1-0/+3
2010-09-30Forward-port 0.6 fixMelanie3-5/+17
2010-09-30Fix an inventory fetch issueMelanie1-0/+3
2010-09-29Cleanup: Removed an unreferenced variable from ActivityDetector.csJohn Hurliman1-11/+1
2010-09-29Part two of the previous commit, making IGridUserService.LoggedOut() more ↵John Hurliman3-5/+5
flexible without changing current behavior
2010-09-29Made IGridUserService.SetLastPosition() more flexible to allow different ↵John Hurliman3-6/+11
possible implementations. This doesn't change any behavior or any Robust code, but it simplifies the SimianGrid connector by using the standard OpenSim interface for setting last position now
2010-09-27Restrict top colliders / top scripts to estate managersMelanie1-0/+3
2010-09-27Prevent setting arbitrary groups on your objects.Melanie1-0/+3
2010-09-27Fix build breakMelanie1-37/+38
2010-09-27Quash the region ID on intersim messages. That prevents people from determiningMelanie1-1/+1
another avatars location
2010-09-27if you can't edit a prim's inventory, you don't need to see the asset ids.Melanie1-3/+11
Prevents stealing IDs of animations, sounds and textures from prim inventories. Prevents copybot from gathering the wearable UUIDs needed for pirating things from vendors.
2010-09-27Restrict inventory items added in a transaction to folder belongingMelanie1-0/+6
to the connected client.
2010-09-26When receiving intersim IM, don't let the spoof protection bombMelanie1-1/+2
2010-09-26Consistency patch: use Path.Combine() instead of + and eliminate the needMelanie3-8/+9
for a trailing slash on exports/
2010-09-26Typo fixesMelanie2-7/+7
2010-09-26adding configurable j2kDecodeCache pathBlueWall4-8/+18
allowing the decoded sculpt map cache path to be defined in the configuration files. Use DecodedSculpMapPath in the [Startup] section to set the path. The default is still ./bin/j2kDecodeCache
2010-09-26Add configurable path to script engine assembliesBlueWall4-15/+25
Adding ability to place script engine assemblies outside the codebase directories. Uses new [XEngine] option: ScriptEnginesPath = "path_to_assemblies" Signed-off-by: Melanie <melanie@t-data.com>
2010-09-25Prevent spoofing the agent name on text IMMelanie1-1/+2
2010-09-25Fix a minor economy issueMelanie1-1/+4
2010-09-25Add prim name to OdePrim Error and Warning messages.Justin Clark-Casey (justincc)1-42/+45
This aims to make it easier to identify and remove rogue prims that are causing ODE to fall over.
2010-09-25Add UUID to physics prim name parameter so that diagnostic messages can be ↵Justin Clark-Casey (justincc)2-3/+3
made more useful. If a separate UUID parameter is better for the future then this can be added later on.
2010-09-24If the uuid gatherer fails to find the asset containing gesture metadata, ↵Justin Clark-Casey (justincc)1-0/+7
then simply return rather than generating an exception This matches existing behaviour, though better diagnostics for missing assets may be good later on. This addresses http://opensimulator.org/mantis/view.php?id=4977
2010-09-24Added delay for llSetPrimitiveParams() and llSetLinkPrimitiveParams() ↵Kevin Cozens1-100/+104
functions per the LSL wiki. Signed-off-by: Melanie <melanie@t-data.com> This patch also fixes a large amount of trailing whitespace. While this is beneficial, it should really be in a separate patch that fixes whitespace only. Just good practice.
2010-09-21Fix a typoMelanie1-3/+3
2010-09-21Refactor script create permission into the perms moduleMelanie2-15/+23
2010-09-21If the uuid of a SceneObjectGroup (RootPart) is changed before adding to the ↵Justin Clark-Casey (justincc)2-2/+41
scene, remove the old uuid reference from m_parts as well as adding the new one. The separate remove and set operations is SOG.set_UUID() are both locked under m_parts.SyncRoot since they are logically atomic (though this isn't such an issue if the SOG isn't part of a scene) Added unit test for this behaviour. Also changed the second m_parts.AddOrReplace() to m_parts.Add(). As the old reference is now removed we never end up replacing an identical uuid. And if we replace a uuid that's already there (from a child part) then this is an error.
2010-09-18Clarify help text for use of file paths with console command "create region".Marck1-1/+1
2010-09-18Stop broadcasting non 0/DEBUG ChatTypeEnum.Region messages to all avatarsJustin Clark-Casey (justincc)1-3/+1
This allows non public/debug region wide messages to be sent to scripts but not be broadast to avatars
2010-09-17For all Flotasm group module XMLRPC calls, correct parameter ↵Justin Clark-Casey (justincc)1-9/+2
requestingAgentID to RequestingAgentID This was stopping the get group member roles call from working, and may have affected other things
2010-09-17Applying the llParseString2List() patch from #5036 that Melanie claims was ↵John Hurliman1-81/+22
already applied
2010-09-17Send kill packets for avatars, tooMelanie1-0/+4
2010-09-17Revert "* Changed 11 calls for session info to the more optimized API method"root3-35/+107
This reverts commit 5dc9ea2f2487804d788b4b80d40d91bd792de4c2. Also makes online indicators and IM more robust
2010-09-16Fixed a regression in SOG.Copy()John Hurliman1-1/+1
2010-09-16Changed SceneObjectGroup to store parts with the fast and thread-safe ↵John Hurliman25-1191/+856
MapAndArray collection
2010-09-17Fix build break by replacing Items.LockItemsForWrite() with lock (Items) {}Justin Clark-Casey (justincc)1-22/+21
2010-09-17Send KillPackets on the Task queue rather than the State queueJustin Clark-Casey (justincc)3-2/+17
Object updates are sent on the task queue. It's possible for an object update to be placed on the client queue before a kill packet comes along. The kill packet would then be placed on the state queue and possibly get sent before the update If the update gets sent afterwards then client get undeletable no owner objects until relog Placing the kills in the task queue should mean that they are received after updates. The kill record prevents subsequent updates getting on the queue Comments state that updates are sent via the state queue but this isn't true. If this was the case this problem might not exist.
2010-09-16Removing debugroot1-1/+0
2010-09-16JustinCC is evil. f7b28dd3 broke script persistence. This fixes it.root4-1/+40
2010-09-15extend m_entityUpdates.SyncRoot lock in LLClientView.ProcessEntityUpdates() ↵Justin Clark-Casey (justincc)1-49/+49
to reduce scope for kill/update race conditions This is necessary because it was still possible for an entity update packet to be constructed, the thread to pause, a kill to be sent on another thread, and then the original thread to resume and send the update This would result in an update being received after a kill, which results in undeletable ghost objects until the viewer is relogged Extending the lock looks okay since its only taken by kill, update and reprioritize, and both kill and update do not take further locks However, evidence suggests that there is still a kill/update race somewhere
2010-09-15rename SceneObjectGroup.DeleteGroup() to DeleteGroupFromScene() to improve ↵Justin Clark-Casey (justincc)4-4/+4
code readability
2010-09-15Instead of locking SOG.Children when a group is being removed from the ↵Justin Clark-Casey (justincc)1-15/+17
scene, iterate over an unlocked list instead Previously, deadlock was possible because deleting a group took a SOG.Children lock then an m_entityUpdates.SyncRoot lock in LLClientView At the same time, a thread starting from LLClientView.ProcessEntityUpdates() could take an m_entityUpdates.SyncRoot lock then later attempt to take a SOG.Children lock in PermissionsModule.GenerateClientFlags() and later on Taking a children list in SOG appears to be a better solution than changing PermissionsModule to not relook up the prim. Going the permission modules root would require that all downstream modules not take a SOG.Children lock either
generated by cgit v1.1 at 2024-09-03 05:20:21 +0000