aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region/ClientStack (follow)
Commit message (Collapse)AuthorAgeFilesLines
* simplify http textures and meshs ThrottlesUbitUmarov2016-07-162-100/+24
|
* lib omv now supports SeeAvs fields, so use it, thx Gavin HirdUbitUmarov2016-07-151-16/+5
|
* revert changes to Select/DeSelect, using the SelectObjects as control seems badUbitUmarov2016-07-141-29/+18
|
* use geometric center and not sog root position in cullingUbitUmarov2016-07-131-4/+6
|
* Oops that was not slowerUbitUmarov2016-07-131-2/+2
|
* slow down culling kills even moreUbitUmarov2016-07-131-1/+5
|
* don't overlap Select and Deselect, fix deselect prims accountUbitUmarov2016-07-131-14/+29
|
* ignore camera on culling for nowUbitUmarov2016-07-131-12/+26
|
* ugly cullingUbitUmarov2016-07-131-70/+121
|
* a few changes to cullingUbitUmarov2016-07-131-26/+28
|
* be nicer to PIs (actually linux)UbitUmarov2016-07-121-5/+9
|
* reduce some locks on culling checksUbitUmarov2016-07-121-37/+39
|
* fix ObjectPropertiesFamilyPacket LastOwnerIDUbitUmarov2016-07-101-1/+1
|
* move object inventory send to udp queue asset ( was task). Remove a callUbitUmarov2016-07-071-22/+22
| | | | | to RefreshGroupMembership() on start. It was too soon and its Groups module job
* the bug was actually hereUbitUmarov2016-07-061-2/+4
|
* add test option ObjectsCullingByDistance. In future, if true, it mayUbitUmarov2016-07-061-2/+126
| | | | | prevent sending objects outside view range to viewers. DO NOT SET TRUE unless testing it. Code still not completei!!!
* Merge branch 'master' of opensimulator.org:/var/git/opensimMelanie Thielker2016-07-043-159/+133
|\
| * dont send bad profiles to viewerUbitUmarov2016-07-021-0/+3
| |
| * reduce absurd viewers blind retries of GroupProfileRequestUbitUmarov2016-07-021-40/+57
| |
| * cleanup a bitUbitUmarov2016-07-013-86/+2
| |
| * save one call to grid to get agent groups powers (xmlrpcgoups only for nowUbitUmarov2016-07-011-1/+1
| |
| * add a few methods to modify clientview group powersUbitUmarov2016-07-011-24/+65
| |
| * cleanup a bit AgentGroupDataUpdate sendingUbitUmarov2016-07-011-35/+32
| |
* | Finally remove the requirement for an InventoryItem/FolderBase object toMelanie Thielker2016-07-041-1/+1
|/ | | | | | | | be passed into inventory queries. This makes the API more homogenous and also will more clearly show coding errors related to HG inventory where the .Owner field has a meaning for a query but wasn't always set.
* revert several changes to groups modules (agentGroupData is private). change ↵UbitUmarov2016-07-012-13/+3
| | | | interregions invite/eject messages, etc
* Make some more modules inheritableMelanie Thielker2016-06-211-13/+13
|
* Mantis #7920: group info was not being updated in many situations. ↵Diva Canto2016-06-192-12/+12
| | | | (regression) Putting back the heavy messaging.
* AgentGroupData was being sent by both UDP and EQ -- an overkill. Turn off ↵Diva Canto2016-06-191-1/+1
| | | | UDP messaging.
* Restored sending group membership data via EventQueue -- this had been a ↵Diva Canto2016-06-192-1/+55
| | | | regression. Took the opportunity to refactor that code, so that both Groups V2 and XmlRpcGroups can use the same function.
* Still related to mantis #7917, parcel prims info. Now sending the correct ↵Diva Canto2016-06-191-6/+1
| | | | numbers for "Simulator primitive usage" related to the parcel owner, not the entire region.
* Mantis #7917: regression on prim limits module.Diva Canto2016-06-181-2/+5
|
* Fields should be protected unless private is requiredMelanie Thielker2016-06-141-1/+1
|
* Make the LLUDPServerShim inheritable. There is absolutely no reason forMelanie Thielker2016-06-141-75/+74
| | | | this to be sealed!
* Mantis #7923. EDIT BEAMS ARE BACK!Diva Canto2016-06-121-2/+1
|
* Mantis #7858: DeleteSceneObject done slightly differently. ProcessEntities ↵Diva Canto2016-06-121-6/+27
| | | | | | now checks whether the objects have been deleted and, if so, sends an extra kill object packet, in order to compensate for potential race conditions encountered by the first one. Note: I still cannot reproduce this problem, but I was able to emulate it by adding an artificial delay on ProcessEntities, which did, indeed, result in objects not being deleted. This fix fixed my emulated scenario.
* Allow default permissions from the viewer to be applied to uploaded objects.Melanie Thielker2016-04-291-9/+38
| | | | Make inventory items reflect coorect permissions immediately unstead of after relog
* Make the SimulatorFeaturesModule nonshared as it ought to be from it's code ↵Melanie Thielker2016-04-271-5/+1
| | | | structure.
* Remove debug spamMelanie Thielker2016-03-151-1/+1
|
* Finish plumbing RezRestoreToWorld.Melanie Thielker2016-03-091-1/+16
|
* Plumb RezRestoreToWorld from the client (part done)Melanie Thielker2016-03-091-0/+7
|
* Make mesh upload respect the same parts maximum as what is set for theMelanie Thielker2016-03-052-1/+2
| | | | simulator
* Fix a couple of typosMelanie Thielker2016-03-051-1/+1
|
* Pushed AssemblyVersion up to VersionInfo, so that we don't need to do global ↵Diva Canto2016-02-212-2/+2
| | | | replace anymore.
* change the return list of objects costs, sending a answer if none foundUbitUmarov2015-12-201-38/+30
|
* dont try to process packets for deleted clientsUbitUmarov2015-12-091-34/+28
|
* disable useless and totally wrong incoming PoolsUbitUmarov2015-12-091-4/+7
|
* minor cleanupUbitUmarov2015-12-091-13/+4
|
* tick region health a bit fasterUbitUmarov2015-12-091-1/+1
|
* fix the queue type in useUbitUmarov2015-12-091-11/+12
|
* fix the queue type in useUbitUmarov2015-12-091-7/+8
|
generated by cgit v1.1 at 2024-10-16 22:28:42 +0000