aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Danxors patch for >30prims with ODETedd Hansen2007-08-261-8/+4
|
* in-world run-time error-message no-w con-tains function-name.Tedd Hansen2007-08-263-12/+16
|
* Run-time script errors are now shown in-world. No line number though, might ↵Tedd Hansen2007-08-264-42/+108
| | | | require script to be compiled with (slow) debug information.
* Bugfix for last commit, { instead of (Tedd Hansen2007-08-261-1/+1
|
* Fixed bug that occurs sometimes on script unload where queued script event ↵Tedd Hansen2007-08-262-41/+57
| | | | was attempted executed after AppDomain was unloaded.
* Fixed error on shutdown caused by ThreadAbortException sending message ↵Tedd Hansen2007-08-252-3/+3
| | | | through already disposed logger. Thanks ckrinke
* Forgot mutex lock on List<> for timersTedd Hansen2007-08-251-11/+15
|
* Script compiler should now show error on correct line number in original ↵Tedd Hansen2007-08-254-13/+16
| | | | LSL-script.
* Applied dalien's libsl login patch (should now allow libsl clients to once ↵MW2007-08-251-4/+4
| | | | again login to opensim) [Seems I was the one who broke it...sorry ]
* Applied dalien's "show users" patchMW2007-08-251-1/+1
|
* oops , shouldn't have committed my local testing changesMW2007-08-251-1/+0
|
* Second attempt to commit : Test to see if this change improves or makes ↵MW2007-08-251-1/+1
| | | | worse the texture sending bug
* Removal of script that was unable to compile no longer crashes server.Tedd Hansen2007-08-253-14/+45
| | | | | Displays script compile error messages in-world.
* Test to see if this change improves or makes worse the texture sending bugMW2007-08-251-0/+2
|
* testing Tedds temporary CIA-standinTedd Hansen2007-08-251-1/+0
|
* testing Tedds temporary CIA-standinTedd Hansen2007-08-251-0/+1
|
* Added class for "long commands" (command that returns as event) with ↵Tedd Hansen2007-08-259-77/+259
| | | | dedicated thread for processing. Added support for llSetTimerEvent(). Deleting old compiled scripts before new compile is attempted (avoids loading wrong script on compile error).
* Scripts no longer crash sim after 5 minutes (override ↵Tedd Hansen2007-08-257-60/+190
| | | | InitializeLifetimeService). Loading/Unloading of scripts are now handled in separate thread so server is no delayed because of this. Each script is loaded into a single AppDomain (temporary test for script unload, eats ~15KB more memory for each script). Unload of scripts has been verified to free up memory.
* Hopefully fixed the bugs in primitives rotation editingMW2007-08-253-6/+9
|
* updated to include saving of object flags.Sean Dague2007-08-252-9/+17
| | | | | | People will need to blow away their databases after this point.
* Set some default values for "create user" in case anyone just pressing ↵MW2007-08-241-4/+4
| | | | return as the answer to some of the settings.
* Now also enabled Inventory persistence in standalone mode when account ↵MW2007-08-241-0/+4
| | | | Authentication is turned off.
* add some better verbose statements to get a better feel for what is goingSean Dague2007-08-241-6/+14
| | | | | | on in the data paths
* Hopefully fixed the issue of inventory not working for the master account. ↵MW2007-08-242-22/+30
| | | | (Note you will need to delete userprofile.yap for this to take effect.)
* Small bit of refactoring to the startup command script code (moved it into a ↵MW2007-08-242-19/+29
| | | | separate method), so that I could add a new CLI command of "command-script <fileName>", so that as well as the startup command script still being processed on startup. A user can create other command scripts and use the single command ("command-script <filename>") to run them at any time. Could be useful for trying out various configurations etc.
* it helps to actually call TestTables to get the new tables createdSean Dague2007-08-241-0/+2
|
* updated SQLite Inventory to newer model data definition,Sean Dague2007-08-241-485/+644
| | | | | | and add automatic generating of the inventory table
* Fixed a problem where some prims didn't show up when you crossed or ↵MW2007-08-241-0/+3
| | | | teleported into another region.
* Added temporary fix for the sqlite datastore exception in windows .Net, ↵MW2007-08-241-2/+11
| | | | Added a Try catch block around "shapeDa.Fill(ds.Tables["primshapes"]);" line. Seems if the database file is empty (ie opensim has just created it or nothing has been stored in it yet.) then the exception will be fired and catch, then opensim can continue and have no problems (it will still save prims fine), then on next restart if the database file has entries in it, the exception will no longer be thrown.
* Had "using OpenSim.Physics.Manager;" defined twice in SceneObjectGroup.cs, ↵MW2007-08-241-1/+0
| | | | so removed one of them.
* Hopefully fixed mantis bug #318 (exception when a prim moves across a ↵MW2007-08-242-3/+24
| | | | border, for now have stopped prims moving beyond a regions area, will add sending prims from one region to another soon).
* attempted fix for .NET issue with the databaseSean Dague2007-08-241-6/+6
|
* Scripts are now copied into a prim (ie new copy placed in the prim) rather ↵MW2007-08-243-9/+34
| | | | than moved into there.
* Updated sqlite3.dll to version 3.4.2MW2007-08-242-0/+9
| | | | | | Fixed it so now when you move a script to a Prim or delete a script from a prim the change should show up in the prims inventory straight away (without having to close the edit window and reopen it). When linking prims, all parts except for the root part of the new group are removed from the physics engine, as currently we only really support root parts in the physics engine.
* Add region to dbSean Dague2007-08-231-48/+50
|
* I think this fixes blob save/load issues (though I'm still not convinced this Sean Dague2007-08-231-20/+26
| | | | | | | | | will handle > 8k textures yet). Need MW to test to see if this gets rid of his issue. There is commented code left in here for now until we know it is fixed
* Lowered priority of script threads. Executing state_entry() event on script rez.Tedd Hansen2007-08-234-11/+2
|
* Added RemovePrim method to the physics plugins interface.MW2007-08-237-13/+54
| | | | | | | | Implemented that method in ODE plugin. Hooked it up so when deleting/taking prims into your inventory they will be removed from physics engine. Enabled the other physics hook ups in Scene.cs (and also added registering prims with physics plugin when they are rezzed from Inventory.) So now to get the avatar to prim collision testing working, just change to use the ODE plugin (in the OpenSim.ini file, physics = OpenDynamicsEngine). Remember though ODE only really works (without problems) when running with a single region.
* grouping of functions to make the overall logic easier to grasp for people,Sean Dague2007-08-231-349/+394
| | | | | | and start to show how this can be super classed with some common elements.
* fix typoSean Dague2007-08-231-5/+1
|
* Implemented Resize Method in OdePrim.MW2007-08-233-8/+31
| | | | | | | | attached the links to that from SceneObject, so now resizing works (as much as resizing currently works in opensim, fixing resizing in general is on my todo list for today). Rotation of a root prim also now updates the physics engine. So think there really is only deleteprim left, then it should be usable (Different shapes (other than boxes that it currently uses) can wait a little bit longer). [of course there are still the other issues of ODE not really working when there is more than one region in a instance of opensim].
* Added a PhysicsActor PhysActor member to SceneObjectPart, and made it so ↵MW2007-08-233-11/+23
| | | | | | | | this is set when registering the prims with the physics engine. Position changes of the prim is now updated straight away to physic engine. (note at the moment, only root prim is registered with physics engine. Think we need to decide how we are going to manage child prims and physics.) As before this is all currently disabled (in scene.cs) until its in a bit more working condition.
* Added danx0r's physics patch, although for now have disabled the lines in ↵MW2007-08-234-2/+50
| | | | Scene.cs, as any changes to prims (like size or position changes) are only updated to the physics engine when you restart opensim. Also prims aren't deleted from the physics engine. These shouldn't be hard to fix.
* More lsl functions (thanks to ldviopeng and wjordan!)Brian McBee2007-08-231-9/+40
|
* setup test tables function which lets us make sure that everythingSean Dague2007-08-221-111/+28
| | | | | | | | | | we are going to ask for from the database is actually there. This will let us bail early with a useful error message, instead of late with a hard to understand one. Do some other cleanups to get rid of debug input I put in
* Bit of refactoring of the sqlite storage code to build the Sean Dague2007-08-221-50/+187
| | | | | | | | data definition in ado.net objects up front. This makes auto generating the sql commands work a lot more reliably.
* Debug shows how many bytes (total) a script (assembly) uses after compile ↵Tedd Hansen2007-08-221-4/+4
| | | | and load.
* GC.GetTotalMemory(true) was blocking.Tedd Hansen2007-08-224-12/+17
| | | | | We now support individual scripts on individual prims. Do the script dance... \o/ \o\ /o/ \o/ .o.
* Added OnRemoveScript event handler to ScriptEngine. Fixed event queuing of ↵Tedd Hansen2007-08-223-2/+8
| | | | empty objects crash.
* Added OnRemoveScript(uint localID, LLUUID itemID) event , trigged when a ↵MW2007-08-2210-2/+73
| | | | script in a primitive is deleted.
generated by cgit v1.1 at 2024-11-19 12:19:41 +0000