* I believe that if the Linden client has not started to receive a texture after 15 seconds, it re-requests it.
* My hypothesis is that the texture packets are often still in the texture queue (esp. if the client has just cleared its cache), so another load of packets get added...
* If this is the cause, resolution is going to be rather complicated.
* This may improve region memory usage
* This is a short-term response to a problem whereby some clients keep requesting the same texture even after we've sent it
* This treats the symptom rather than the cause.
* n can be adjusted by changing the constant at the top of UserTextureDownloadService if necessary
* When a new prim is created and raytracing is called for, raytrace from the camera position to the ground in the direction of the Norm(RayEnd - RayStart).
* If we got a hit based on our camera, create the new prim at the edge of the prim we hit.
* Don't raytrace if the difference between any component of the vector exceeds 4.5meters.
* We should not be using ASCII anywhere except for legacy compatibility reasons.
* A large number of UTF8 Encoders are being used in places where we should be using Util.StringToField instead. These have been tagged with // ENCODING FAULT
* This should fix Mantis#799 - Japanese Profile Text does not work.
* Should work in multi-region standalone and grid modes
* This should also solve other non-home region caps issues (map requests, RC client inventory requests, etc)
* We now pass CAPS information on to the destination region on region crossing, and set up a CAPS object when an agent becomes a master
* Current limitation is that this will only work if your http_listener_port is 9000
* This is a very early code cut (lots of bad practice, hard coding and inefficiency). However, I wanted to get this out there for feedback and my own sanity. Next few patches will clean up the mess.
* This reveals that the problem with saving scripts in a non-home region in multi-region configurations is due to a CAPS setup issue
* For some reason the client is still using the CAPS on the region it just came from, causing the ScenePresence lookup to fail (since the presence is now, correctly, a child agent).
* Remember, your admin user and estate managers can move locked objects that are not owned by them. That functionality differs from the Linden way of thinking and it's by design! It is not a bug! Create a non-god user and use that as your normal account.
* The warning will be
[USER TEXTURE DOWNLOAD SERVICE]: Received {0} requests for already dispatched texture {1} from client {2}
This is to see whether the texture packet queue memory leak is caused by clients continually re-requesting textures they should already have
* Should help stop any InvalidOperationExceptions caused by concurrent read/write
* The extra locking should be okay, but I'm really surprised we've got away without mucho crashes due to this...
structs (such as LLUUID) are considered values by mono. comparing them against null makes no sense and the mono compiler will flag that as evaluating to always false --- except if "cleverly" disguised. the attached patch fixes such an occurrence in OpenSim/Region/ScriptEngine/Common/ScriptEngineBase/AsyncCommandPlugins/SensorRepeat.cs.
[yes, i'm on a crusade against comparing structs against null, go ask jradford from libsl ;-)]
* If a user logs in and they are noted as agentOnline. Set agentOnline = false and send a 'you're already logged in' message to the user asking them to wait 5 minutes. These 5 minutes are not enforced (because there's no foolproof interlock release yet without the grid operator getting a support call for every little sim crash). When the user gets the message, they can log-in immediately after it, but the user can expect weird results if they don't wait 5 minutes and log-in to the region they were in previously.
Here's a diff of the changes I have made in support of the following LSL
script functions.
llSetScriptState
llGetScriptState
llCSV2List
llListRandomize
llList2ListStrided
llListFindList
llResetOtherScript
llGetScriptName
It was necessary to modify ExecutorBase in support of the ScriptState
implementations.
I also modified SceneObjectPart and SceneObjectPart.Inventory to
corrects a quoting mismatch in the commentary that through off live
parsing of the files.
I also simplified the State definition at the start of BuiltinCommands.
So this commit doesn't actually fix inventory in that client, it just stops the "loading" message being displayed forever next to a folder, and instead shows empty folders.
Next part will be to fill in the details of the items in the folders.
* Even very rapid linking/delinking should now behave normally. Terse updates still occur as before
* Hopefully this ends the recent linking problems - please let us know if there are more
* Noise, and Noise-Area brushes now use Perlin noise, more closely simulating the method LL uses officially.
* TerrainModule has been cleaned up slightly.
* TerrainUtil class has several new functions related to seeded noise generation.
* Extracted ITerrainEffect, ITerrainFloodEffect, ITerrainLoader, ITerrainPaintableEffect, TerrainChannel to seperate files.
* This module more or less crashes every region in the instance if you enable it by moving it from local /bin to global /bin
* But hey, it crashes in lots of interesting ways.
* See OpenSim.ini.example for more details
* The old sqlite/mssql settings were redundant and effectively ignored anyway. Hence, there's no need for you to change your current settings, which will still work
* In fact, asset_database should probably be moved to [Network] since it's meaningless for standalone installations
* DelinkFromGroup was removing the parts from the delinked group, which later upset the update thread when it tried to do a queued update for that object
* Temporary fix is to stop deleting the parts, though it would be good later to stop sending out the now spurious updates
* This fix actually reveals another bug, where rapid linking and delinking will cause the non root prims to disappear (though they're actually still there if you relog). This is the next bug to tackle.
* Terrain Area-of-Effect brushes now should work properly.
* Updated the scale / effect of various brushes. Small & Medium brushes should now actually do something.
* This is in order to identify whether it is these which are somehow causing the massive texture packet bloat,
* since this is the only other packet put on to that queue other than the TextureSender ones
* Added osSetPrimFloatOnWater(BOOL) to make Physical prim float at the water level.
* osSetPrimFloatOnWater(TRUE); or osSetPrimFloatOnWater(FALSE);
* By default, prim do not float at the water level.
* More work is needed on the floating, but it's a start.
* W.I.P: Hydraulic Erosion (Spherical) paintbrush - code is all there, but some work is required on the initial parameters to get it to function correctly. Replaces the smooth brush when `newbrushes` is switched on.
* Implements the Optimised Erosion routine defined by Jacob Olsen in the paper 'Procedural Terrain Generation' (http://www.oddlabs.com/download/terrain_generation.pdf)
* Replaces the 'flatten' brush when 'newbrushes' is enabled.
* Started OpenSim on SQLite to a mass of red 'unable to load prim' messages.
* If you are getting this message, after this update, when you start OpenSimulator, you'll get it one more time and then it'll work again.
* Added "script terrain newbrushes <true|false>" to enable experimental terraforming brushes. Presently the 'revert' brush is replaced with the Weathering brush, when enabled.
Grass type is not persisted. Try to rez some grass several time to have different random grass types around, then immediately move a couple of them. Or restart the sim and logon again. You'll see all grass types reverted to a dull default one :-)
Also added osSetDynamicTextureDataBlend and osSetDynamicTextureURLBlend that will allow the various textures to be blended together, but currently there are still a few bugs in them. So not ready for use yet.
As per the suggestions in Mantis 687 (thanks Sakai) to resolve an exception
* This won't apply to any existing inventory tables you have, but shouldn't
cause any issues with old tables either
* Trying to store items other than textures, sounds and scripts in a prim's inventory should no longer cause an exception.
* Temporary solution is to ignore storage requests for these assets - actually implementing this requires changes to TaskInventoryItem, at least
* Enabled new TerrainModule. (The king is dead, long live the king!)
* Use the console command: "script terrain save file.r32" / "script terrain load file.r32" to load/save terrain. Now uses the extension to determine file format.
* MANY of the old terrain features do not have a replacement function in the new module yet, this needs to be corrected, but has not been done so far. This being said, the new module is faster and more efficient and should be a good replacement.
* *Much* faster terraforming (woot!)
* New "Brushes" design, so you can create custom terraforming brushes then apply those inplace of the standard tools. (ie an Erode Brush for example)
* New specialised "Flood Brushes" to do large area effects, ie, raise-area, now takes a bitmap rather than repeats the ordinary raise brush a thousand times.
* New modular file Load/Save systems -- write importers/exporters for multiple formats without having to hard code the whole thing in.
* Coming soon - effects system, ie the old Erosion functions, etc. for one-shot effects.
* ODE: Added support for larger box stacks. (they're slow, but they work)
* ODEPlugin no longer tries to 'catch up' with the simulator frame rate if it gets behind. Catching up was causing a lot of problems with larger box stacks and other things that stall the simulator (like saving prim in the datastore)
converts Parse calls to TryParse
Replaces string.empty and lluuid.zero calls on user load to use real
values from DB (related to user profile save/load work. Finer grained patching)
appropriate names consisten with their use.
All done with all 94 handlers from handler001
through handler094. Hopefully we can move
forward without numbered handlers.
* To play with this you must link your prim before setting it physical, otherwise they won't link in the physics engine properly. This will also be fixed.
* Currently the linked prim are extremely unstable because I have yet to implement combining of forces with the same normal. This will also be fixed. In fact, the whole PhysicsActor, ODEPrim relationship will be reworked to consider groups from the get-go.
* This implementation is better then it crashing your sim, so I'm commiting it for now.
to more meaningful names starting from the end and
working backwards.
handler094 -> handlerRequestAsset
handler093 -> handlerPacketStats
handler092 -> handlerGodKickUser
handler091 -> handlerUpdatePrimGroupRotation
handler090 -> handlerUpdatePrimRotation
Others to come shortly.
* Impelements llInstantMessage
* Reimplements llOwnerSay as llInstantMessage(llGetOWner(),msg).
* Try's to better identify the "True Name" of objects in llDetectedName
by checking for avatar name, scene Object Part name and entity name.
* Uses similar logic in the llSensor and llSensorRepeat functions.
Below is a patch for the smooth tool.
I factored out the essential computations and placed it in a
channel method to work similar to raise and lower.
It now performs about the same rate as raise and lower.
** Added more realistic calculations of mass for the rest of the supported prim shapes+holes+cuts+tapers. Previously they were the generic height * width * length. Spheres roll (Angular velocity) more realistically, etc.
This patch is intended to implement the following functions:
llIntegerToBase64
llBase64ToInteger
llParseStringKeepNulls
None of these functions are dependent upon state elsewhere in the SIM,
so they are appropriately self-contained. I've tested them out of
context, and from a script attached to an object in my test region.
So anyway, it moves the Asset downloading (packet sending) to a module (AssetDownloadModule).
So now at last, AssetCache should be just dealing with fetching assets from the asset server and caching them.
* This might stop some client's constant requests for unfound textures, which is a candidate for the memory leak
* If a texture is not found then the "Image not found" texture will now be displayed clientside
* If it works, this should resolve mantis 676
* Non texture image requests do not receive this packet yet
* This will require a prebuild
mingchen