aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region (follow)
Commit message (Collapse)AuthorAgeFilesLines
* When an object is duplicated, add the dupe to the uuid/local id indexes as ↵Justin Clark-Casey (justincc)2011-03-262-1/+52
| | | | | | | | | well as the basic entities list. Added a prim counts test to reinforce this - shift-copy was no incrementing prim count. This will sometime become a basic scene test. New code needs to be refactored so we just call SceneGraph.AddSceneObject(). This will happen in the near future. With this, basic owner prim counts on a single parcel appear to be working fine (with the same previous existing taint calls as used by the land management module). More work to do.
* Merge branch 'master' into primcountsJustin Clark-Casey (justincc)2011-03-261-28/+48
|\
| * Make SceneGraph.ForEachSOG() execute once for each SOG, not once for each ↵Justin Clark-Casey (justincc)2011-03-261-5/+25
| | | | | | | | | | | | | | | | prim (e.g. a SOG with 3 prims would have the action executed three times). To do this, a new SceneObjectGroupsByFullID index in SceneGraph which just index's prims by their root part UUID, in order to avoid the inefficiency of filtering existing lists. Existing callers to SceneGraph.ForEachSOG() did not fail due to the multiple per SOG action executions - they were probably just much less efficient. Code suggests that no callers expected ForEachSOG() to execute actions on sog multiple times
| * refactor: rename SOG collections in SceneGraph to make it clearer that they ↵Justin Clark-Casey (justincc)2011-03-261-30/+30
| | | | | | | | are indexing each part's UUID, not just the root part.
* | Add test for PCM taint. This currently fails due to unexpected behaviour of ↵Justin Clark-Casey (justincc)2011-03-263-16/+89
| | | | | | | | | | | | SceneGraph.ForEachSOG(). This will be corrected soon. Also adds lots of temproarily debug logging
* | Start using IPrimCounts populated by PrimCountModule instead of LandData ↵Justin Clark-Casey (justincc)2011-03-2511-241/+30
|/ | | | | | | counts populated by LandManagementModule. In order to pass ILandObject into IClientAPI.SendLandProperties(), had to push ILandObject and IPrimCounts into OpenSim.Framework from OpenSim.Region.Framework.Interfaces, in order to avoid ci Counts are showing odd behaviour at the moment, this will be addressed shortly.
* Add ILandObject.IPrimCounts for the new prim count module.Justin Clark-Casey (justincc)2011-03-255-31/+52
| | | | Not functional yet, but tests now act against this object rather than interrogating the module directly
* factor out common test setup code in PCM testsJustin Clark-Casey (justincc)2011-03-231-57/+53
|
* extend TestAddOwnerObject() to add a second objectJustin Clark-Casey (justincc)2011-03-231-4/+14
|
* Add PrimCountModuleTests.TestRemoveOwnerObject().Justin Clark-Casey (justincc)2011-03-231-1/+38
| | | | Also adds SceneSetupHelpers methods to easily create sogs with different part UUIDs
* refactor: simplify part of AddSceneObject() test setup by moving sog ↵Justin Clark-Casey (justincc)2011-03-231-10/+1
| | | | construction into SceneSetupHelpers.CreateSceneObject()
* use a 3 part object for the pcm test rather than a 1 part, for improved test ↵Justin Clark-Casey (justincc)2011-03-231-10/+14
| | | | coverage
* user a non UUID.Zero user in pcm test to avoid any special treatment of ↵Justin Clark-Casey (justincc)2011-03-231-4/+5
| | | | UUID.Zero
* Extend simple PCM add object test to check all countsJustin Clark-Casey (justincc)2011-03-231-0/+11
|
* Add method doc to the Get*() methods on PrimCountModuleJustin Clark-Casey (justincc)2011-03-231-0/+27
|
* remove a rogue Console.WriteLine() from the last commit.Justin Clark-Casey (justincc)2011-03-231-1/+0
|
* Add generic EventManager.OnObjectAddedToScene and get PrimCountModule to ↵Justin Clark-Casey (justincc)2011-03-235-5/+59
| | | | | | | listen for that rather than EventManager.OnParcelPrimCountAdd OnParcelPrimCountAdd had the wrong semantics for the PrimCountModule - it was invoked for every entity in the scene, not just new ones, which would screw up the untainted count. Extend automated test for this scenario.
* Add an initial confidence-building TestAddObject() for prim counts.Justin Clark-Casey (justincc)2011-03-224-25/+115
|
* On initial setup, include estate and regions names in questions to make it ↵Justin Clark-Casey (justincc)2011-03-212-5/+7
| | | | clearer what they relate to.
* In initial setup, stop a user being able to create a new estate with the ↵Justin Clark-Casey (justincc)2011-03-211-14/+31
| | | | same name as an existing estate.
* On initial opensim setup, don't ask the user whether they want to join an ↵Justin Clark-Casey (justincc)2011-03-211-39/+63
| | | | existing opensim estate when there aren't any. Proceed directly to estate setup instead.
* refactor: use EstateDataService property directly instead of loading it into ↵Justin Clark-Casey (justincc)2011-03-211-12/+7
| | | | a local variable
* minor: slightly adjust previous method doc.Justin Clark-Casey (justincc)2011-03-212-2/+2
|
* add method doc to IEstateDataService and IEstateDataStore interfacesJustin Clark-Casey (justincc)2011-03-212-2/+90
|
* On initial region registration, if the user chooses the option to make the ↵Justin Clark-Casey (justincc)2011-03-213-1/+40
| | | | region part of an existing estate, then list the existing region names.
* Thanks Kevin Cozens for a patch that:BlueWall2011-03-172-5/+5
| | | | Fixes several spelling mistakes
* When setting media on a prim values, use generic object permissions instead ↵Justin Clark-Casey (justincc)2011-03-141-2/+2
| | | | | | | of media permissions. Media permissions are just meant to be checked when we want to know if a user should be shown the navigation bar or allowed to navigate. It should not be checked when we're setting the media up. This bug was preventing a user from ever setting any more values if they had unchecked the owner settings.
* factor inventory file loading into its own methodJustin Clark-Casey (justincc)2011-03-121-57/+80
|
* factor out iar asset load code into its own methodJustin Clark-Casey (justincc)2011-03-121-29/+40
|
* refactor: make boolean load indicators on load iars instance fieldsJustin Clark-Casey (justincc)2011-03-121-8/+11
|
* introduce iar load checks which make sure archive.xml comes first, then ↵Justin Clark-Casey (justincc)2011-03-121-1/+32
| | | | | | inventory nodes, then assets this is necessary for correct loading. "save iar" always saves in this order so there shouldn't be any problems - these checks are to give better feedback to other systems that may construct IARs.
* Change how map blocks are encoded to make map searchMic Bowman2011-03-101-2/+4
| | | | work with viewer 2.
* Add test for simple case where creator account with appropriate uuid exists ↵Justin Clark-Casey (justincc)2011-03-101-0/+32
| | | | on the target system for an iar load
* refactor: rename test user account fieldsJustin Clark-Casey (justincc)2011-03-103-35/+35
|
* factor out common scene setup code in InventoryArchiveTestCaseJustin Clark-Casey (justincc)2011-03-102-41/+33
|
* extend TestLoadIarV0_1AbsentCreator() to check serialized object CreatorId ↵Justin Clark-Casey (justincc)2011-03-101-0/+6
| | | | as well
* re-enable item.CreatorId check in TestLoadIarV0_1AbsentCreator()Justin Clark-Casey (justincc)2011-03-102-5/+6
|
* For objects loaded from an IAR, make sure the CreatorID points towards the ↵Justin Clark-Casey (justincc)2011-03-103-10/+48
| | | | | | | | | OSP resolved ID if newer CreationData is not present. This should resolve issues where the creator for rezzed objects was being shown as "Unknown user" where previous behaviour was to show the OSP resolved account. This is being done by parsing the serialized objects and updating the CreatorID if no CreationData exists. This operation might be expensive for sculpties where the sculpt texture is inlined with the object data. Will just have to see. This relies on the IAR streaming inventory data before asset data (as is currently the case). Will need to introduce more stringent checks for file order on loading (much like JAR zips must start with the manifest file). This is for IAR loading only. Tests updated to check this behaviour.
* When setting up default iar for testing, use iar archiving code rather than ↵Justin Clark-Casey (justincc)2011-03-103-13/+17
| | | | constructing the tar manually
* Change existing users load iar test so that it fulfills it's original ↵Justin Clark-Casey (justincc)2011-03-102-11/+15
| | | | | | intention (i.e. OSPA resolution, which is still active). Will need to write a separate test for the simplest case where creator accounts with appropriate uuids exist on iar load.
* Split the inventory path testing parts of TestLoadIarV0_1ExistingUsers() ↵Justin Clark-Casey (justincc)2011-03-102-23/+46
| | | | into a new test TestLoadIarToInventoryPaths()
* Make the item created in the default test iar an object rather than a scriptJustin Clark-Casey (justincc)2011-03-091-2/+27
|
* minor: change the order of some default iar setup in test caseJustin Clark-Casey (justincc)2011-03-091-5/+7
|
* Upgrade nunit.framework.dll to version 2.5.9. Fix up tests appropriately.Justin Clark-Casey (justincc)2011-03-0921-21/+0
| | | | This version removes the NUnit.Framework.SyntaxHelpers namespace, so any modules with their own tests will need to delete this using statement.
* Add log messages on teleport failure to better pin down the cause.Justin Clark-Casey (justincc)2011-03-091-3/+9
|
* Split out path tests from InventoryArchiveTests. Factor common code into ↵Justin Clark-Casey (justincc)2011-03-093-531/+562
| | | | test case parent
* Simplify TestLoadIarV0_1AbsentUsers() to use common IAR test setup. Make ↵Justin Clark-Casey (justincc)2011-03-092-50/+17
| | | | static dictionaries on NullUserAccountData instance instead to stop user accounts being carried over between tests
* Make -m shortcut option for --merge on load iar specific. Correct some log ↵Justin Clark-Casey (justincc)2011-03-081-7/+7
| | | | message origins.
* factor out SetPartMediaFlags() function in MoapModule.Justin Clark-Casey (justincc)2011-03-051-11/+20
|
* Construct test load iar only once and reuse for each test, rather than ↵Justin Clark-Casey (justincc)2011-03-051-8/+18
| | | | recreating it every time