aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region/Environment (follow)
Commit message (Collapse)AuthorAgeFilesLines
* last round of warning squashing. calling it a day now.Dr Scofield2008-06-277-25/+29
|
* more warnings to go.Dr Scofield2008-06-275-23/+35
|
* * Experiment with not bothering with a seperate assets metadata file - ↵Justin Clarke Casey2008-06-273-32/+69
| | | | | | | | extract necessary minimum metadata from asset filenames * Make all file extensions single - double extensions look ugly and probably aren't good for Windows
* move along, nothing to see here. just a couple of lazy variables.Dr Scofield2008-06-278-25/+31
|
* run NArrange on SOP to stack together the attributes,Sean Dague2008-06-271-1919/+1866
| | | | | | properties, private methods, public methods, etc.
* Mantis#1615. Thank you, Matth for a patch that:Charles Krinke2008-06-271-1/+1
| | | | | | causes deprecated LSL functions to throw the deprecated exception rather than not-implemented.
* dr scofield's continuing warnings safari:Dr Scofield2008-06-272-2/+3
| | | | | | | * code caught lingering around with no real purpose other than causing warnings will be taken down.
* * fix a bug where I was assuming all entities would be SceneObjectGroups - ↵Justin Clarke Casey2008-06-272-11/+18
| | | | | | | | of course, some are ScenePresences * would be nice to have a separate list in InnerScene in the future since a lot of methods end up screen entities for SOGs
* * Remove responsibilty for signalling scene object change from ↵Justin Clarke Casey2008-06-274-9/+43
| | | | | | | | | | | | SceneObjectPart.SendFullUpdate() * This means that we will no longer pointlessly repersist all the prims in the scene when OpenSim first starts up * This also means that force-update on the console will not trigger repersistence. * Also, in other places persistence is no longer done where it wasn't actually necessary * I think I changed the code for all instances correctly, but it's not possible that I missed some and some things which did persist properly have stopped * Please patch or mantis if this is the case
* * refactor: Remove group changed responsibilty from ScheduleGroupForTerseUpdateJustin Clarke Casey2008-06-272-3/+8
|
* * Remove SOG.SendGroupFullUpdate() group changed responsibilityJustin Clarke Casey2008-06-272-3/+4
|
* * refactor: Move responsibility for deciding whether a group should be ↵Justin Clarke Casey2008-06-272-5/+24
| | | | repersisted on to callers of SceneObjectGroup.SendGroupTeruseUpdate()
* * refactor: fission InnerScene.AddSceneObject()Justin Clarke Casey2008-06-275-9/+25
|
* * Add extra backup parameter to InnerScene.AddRestoredSceneObject()Justin Clarke Casey2008-06-275-11/+25
|
* Mantis#1612. Thank you, kindly, Matth for a patch that:Charles Krinke2008-06-272-0/+105
| | | | | | Adds the beginnints of llRemoteLoadScriptPin() and llSetRemoteScriptAccessPin().
* Mantis#1610. Thank you, Melanie for a patch that:Charles Krinke2008-06-262-2/+7
| | | | | | Implements the beginning of llInventoryDrop. Doesn't cover the actual dropping yet, just the permissions for it.
* Mantis#1603. Thank you, Matth for a patch to llGiveInventoryCharles Krinke2008-06-261-0/+21
| | | | | and Melanie for helping to finish it.
* Update svn properties. Minor formatting cleanup.Jeff Ames2008-06-261-2/+1
|
* * Put extensions on all assets in an archiveJustin Clarke Casey2008-06-263-14/+66
| | | | | | | * Choice of extensions is a scratch set, may not yet be correct * This facility is not useable yet - not least because I've just come across a significant save bug I need to investigate
* Mantis#1597. Thank you, Melanie for a patch that:Charles Krinke2008-06-264-1/+52
| | | | | | Adds handlers for the reclaim land functionality, plus all needed permissions checks.
* Mantis#1594. Thank you, Melanie for a patch that:Charles Krinke2008-06-261-50/+92
| | | | | | | | | Fixes: - Wearable icon and name sreset to default on copy/paste - Cache is not updated when renaming/moving folders - Partial refactor to make inventory less dependen on AssetBase having a "Name" field - Add llGiveInventoryList() function
* added the flag param to IClientAPI.SendMapBlockMW2008-06-253-5/+5
|
* first part of Requestmapblocks fixes: Adds uint flags param to the ↵MW2008-06-251-1/+2
| | | | OnRequestMapBlocks event (and handler), as when a client sends a map block request it also sends what layer it wants it for 0,1,2. It will always send two requests, one of them being for layer 2 (the overlay layer) and the other one either 0 or 1 depending on the tab that is selected in the client worldmap window. We should also be sending what layer the reply is for in IClientAPI.SendMapBlock (current always set to 0). That will come in next part (most likely at the weekend).
* forgotten methods in SceneManager for the terrain serialising.MW2008-06-251-0/+11
|
* Added support for terrain map to be serialised to xml(as base64 binary). ↵MW2008-06-256-5/+85
| | | | | | | | useful for places that the terrain map is needed in a serialised form. Also could add console commands to save and load from files, which should be faster than .raw files (these load/save commands are not included/implemented) Add util functions to compress and uncompress strings. Fixed a couple of modules so they use SceneCommunicationService rather than directly call functions on the CommsManager.
* the prefix "fix" turned out to be bollocks. sorry about that. hope this one Dr Scofield2008-06-251-1/+0
| | | | | | fares better.
* Minor formatting cleanup.Jeff Ames2008-06-253-66/+63
|
* taking another look at mantis #1502: adding necessary locks, checking for ↵Dr Scofield2008-06-251-3/+11
| | | | empty list.
* Mantis#1590. Thank you, Melanie for a preserve type patch.Charles Krinke2008-06-251-0/+1
|
* Mantis#1587. Thank you kindly, Melanie for a patch that:Charles Krinke2008-06-242-15/+38
| | | | | | | Add permissions magling to llGiveInventoryItem, correct some corner case functionality
* Mantis#5189. Thank you kindly, Matth for your patchCharles Krinke2008-06-241-13/+101
| | | | | to add the beginnings of llGiveInventory().
* Mantis#1569. Thank you kindly, Nebadon for a patch thatCharles Krinke2008-06-231-2/+2
| | | | | | | addresses: Rotation of terrains varies between all the formats, Raw and R32 seem to be the same, while PNG, JPG, BMP all seem to be rotated 180 degrees different.
* * Remove all use of asset.InvType, as outlined in mailing list discussionJustin Clarke Casey2008-06-234-40/+53
| | | | | | * This is stage 1, the field will be removed from AssetType and the assets table if this change doesn't prove problematic
* * archive each object as a separate xml file rather than put them all in one ↵Justin Clarke Casey2008-06-215-74/+42
| | | | | | | | single large file * this is being done for reasons of compositionality
* * refactor: Making some of the serialization names consistent, both within ↵Justin Clarke Casey2008-06-215-23/+54
| | | | | | | | their pairings and with the others * This does make some of the names less ideal than others - this should change in the future
* * Remove redundant <scene> tag from individual object xml representationJustin Clarke Casey2008-06-211-7/+19
| | | | | | | | | * WARNING: Until both regions upgrade to this revision, prim crossings from regions on this revision to earlier region revisions will not work properly * e.g. in the attachments cases, the attachments will remain visible but won't be individually detachable * This change may seem to have more costs than benefits, but I'm doing it because I can soon reuse this changed existing code in another context - it seems better not to proliferate similar but slightly different xml serializations.
* * Refactor: Fold sog.DeleteParts() functionality into existing sog.DeleteGroup()Justin Clarke Casey2008-06-212-22/+9
|
* * Remove a bug I created in r5171 where taking an object would terminate the ↵Justin Clarke Casey2008-06-216-14/+40
| | | | | | | | client session * change code to use an explicit state variable instead of using SOG.Name = null to signal deletion
* * minor: start saving current prims xml to an objects/ directory in the archiveJustin Clarke Casey2008-06-211-1/+1
|
* * minor: override the name of the sog such that it gets/sets the root part nameJustin Clarke Casey2008-06-212-1/+10
|
* * Adds Region ban capability to Regions. You access this by going to ↵Teravus Ovares2008-06-214-1/+133
| | | | | | | | World->Region/Estate. Then on the Estate tab, at the lower right hand corner, clicking the 'Add' button and picking an avatar. * It only persists across reboots for the mySQL datastore currently. * Currently have stubs in the other datastores.
* Mantis#1543. Thank you kindly, Jonc for a patch that:Charles Krinke2008-06-192-4/+81
| | | | | | Implements terrain bake from Region/Estate dialog and respects estate settings during terraforming
* * Temporary workaround for mantis 1568. If the avatar apperance factory now ↵Justin Clarke Casey2008-06-191-3/+20
| | | | throws any exception, we will carry on with the default appearance rather than terminating the client login
* * Refactors call to OutPacket out of AssetCache and into LLClientViewTeravus Ovares2008-06-181-0/+4
|
* * By popular demand, skipping trees from the map tile generation routine. ↵Teravus Ovares2008-06-181-0/+6
| | | | Skipping these because we have no good representation, they just appear as blocks that obscure the rest of the map items.
* revert 5134. Changing these bitmaps to 24bit just breaks things, and they ↵Sean Dague2008-06-182-2/+2
| | | | | | | | never display on the client.
* Minor formatting cleanup.Jeff Ames2008-06-183-69/+53
|
* make neb happy. I found where we initialized the dynamic textures toSean Dague2008-06-182-2/+2
| | | | | | 32bit images and changed them to 24bit images.
* change the default drawing font from Times -> Arial, asSean Dague2008-06-181-4/+12
| | | | | | | | san serif fonts are a bit easier on the eyes on textures. Add a new "FontName" attribute that can be used to override the font type.
* tweak dynamic texture stuff after the meeting today. This shouldSean Dague2008-06-171-1/+7
| | | | | | | now preserve the bulk of the texture attributes, and force on full bright. Not tested yet, but shouldn't bother most people.