4ae04dd0a8
extend exception catching for DoCreateChildAgentCall() from just WebException to Exception on feedback from Chris Hart.
...
See http://opensimulator.org/mantis/view.php?id=4810
2010-07-17 01:00:23 +01:00
61f9b3c2cf
Bug in 0.6.9 sometimes restoring script state causes region console to crash due to unhandled file lock exception. Attempt to resolve by wrapping several instances of file create / read logic in using statements and added some error handling for locked file exceptions. If it is IDisposable, it must be disposed! The close statements are unnecessary but harmless so I have left those in. The end of the using block will close and dispose automagically.
2010-07-17 00:44:54 +01:00
b2ef5658b1
Merge branch '0.6.9-post-fixes' of ssh://opensimulator.org/var/git/opensim into 0.6.9-post-fixes
2010-07-13 23:36:56 +01:00
a7f0af41ab
replace o with ; in oXmlRpcPort in OpenSim.ini.example
...
Thanks to Nebadon for the spot
2010-07-13 23:36:05 +01:00
c7a2e1b2d4
llVecNorm() now returns a zero-length vector when one is supplied as input. Addresses Mantis #4752
2010-07-08 10:17:28 -07:00
0116b80795
fix a potential division by zero
2010-07-08 10:14:02 -07:00
d743a221e8
Fix scripts in rezzed objects not starting (Mantis #4775 )
2010-06-30 21:27:55 +01:00
92edbd3e47
If an emergency pos is required for a new root agent, correctly preserve the z co-ord instead of accidentally resetting the incorrect x co-ord
...
This is a one-liner. It has already been addressed in master and 0.7-post-fixes in a more extensive way
2010-06-30 21:23:57 +01:00
cabb70d90a
Revert "stop exceptions in setting and getting state from propogating since they aren't fatal to operations"
...
This reverts commit ae24548216
2010-06-29 22:50:04 +01:00
ae24548216
stop exceptions in setting and getting state from propogating since they aren't fatal to operations
...
this will hopefully stop "save oar" from failing if a script asset is corrupt
2010-06-29 18:54:31 +01:00
85c20e1493
revert group membership checking to older code pending resolution of reported performance problems
2010-06-11 19:40:09 +01:00
821b8a5517
remove SimianGrid Groups OpenSim.ini.example stuff that accidentally crept in through a cherry-pick
2010-06-11 18:43:02 +01:00
491d898db8
If a script is deleted before it gets compiled, don't even bother to try
...
compiling it
2010-06-09 03:52:59 +01:00
5f60eb136e
Improve instructions for [Groups] setup in OpenSim.ini.example
2010-06-08 21:16:19 +01:00
8b3b89ecf4
backport xmlrpcgroups cache from master
2010-06-08 20:31:39 +01:00
0524458069
remove references to OpenSim.Framework.Commuications.Tests.dll since all relevant test code has been obsoleted
...
this allows the tests to pass on my local system
2010-06-08 16:10:40 +01:00
a160b44e07
Merge branch '0.6.9-post-fixes' into share-with-group
2010-06-08 15:50:21 +01:00
255e05beb5
minor: remove some commented out code and return ScenePresence.UpdatePriority() to private
2010-06-07 19:52:45 +01:00
912f0c7fde
Stop some hud components disappearing on region crossings
...
If viewers (or at least, Linden Viewer 1.23.5) receive child hud object updates before the root prim, then the children are not displayed.
Updates were being queued in LLClientView in the right order (root first) but were being sent in a random order since they were all at the same prioritization
This commit prioritizes the root prim of a hud to its highest level when queued.
I'm not sure if the periodic reprioritization triggered via ScenePresence might reset this, but boosting priority appears to work so far.
Also committed is a belt and braces mechanism in LLClientView to prevent child hud prim being sent out before their root, but since this doesn't appear to be needed it is currently commented out.
2010-06-07 19:14:40 +01:00
693b5d0838
reapply fix for double sending of attachment update on standalone region crossing
2010-06-07 19:14:30 +01:00
50ddb20204
commit code which stops full updates being fired multiple times when attachments cross standalone region boundaries
...
lots of messy debug code here too which would need to be removed
2010-06-07 19:14:22 +01:00
bdeda18b52
Revert "commit code which stops full updates being fired multiple times when attachments cross standalone region boundaries"
...
This reverts commit 5074d290e4
2010-06-04 18:54:58 +01:00
8eb70f9719
Fix bug where prim items were not loaded in the new sqlite database handler
...
This addresses mantis http://opensimulator.org/mantis/view.php?id=4739
2010-06-04 18:02:30 +01:00
2f6d401db1
Merge remote branch 'origin/0.6.9' into 0.6.9-post-fixes
...
This chiefly brings in the new sqlite adaptor and renames the old one to SQLiteLegacy
Existing configuratios should continue to work without changes unless you are using Mac OSX and mono 2.6 or later, in which case you will need to enable SQLiteLegacy instead. Please se the instructions in OpenSim.ini.example and the relevant config/include .ini files
Conflicts:
OpenSim/Framework/Servers/VersionInfo.cs
OpenSim/Region/Framework/Scenes/Scene.Inventory.cs
2010-06-04 18:00:48 +01:00
82d0d1c092
Revert "Next OSG test. Don't remove app domains if they contain running threads."
...
This reverts commit f831a1778b
2010-06-02 21:02:53 +01:00
f831a1778b
Next OSG test. Don't remove app domains if they contain running threads.
2010-06-02 19:57:03 +01:00
20e6b7a320
Next OSG test. Cowardly refuse to save the state of badly behaved scripts, instead of forcing the issue.
2010-06-02 18:05:58 +01:00
c38f7c96e7
Revert "Next OSG TEST RELEASE. Join aborted threads."
...
This reverts commit 698066d7e3
2010-06-02 18:00:44 +01:00
698066d7e3
Next OSG TEST RELEASE. Join aborted threads.
2010-06-02 17:41:28 +01:00
c5cfae7bac
Revert "After aborting a thread, wait for it to actually finish. EXPERIMENTAL for OSG"
...
This reverts commit 241ed5734e
2010-06-02 17:39:58 +01:00
241ed5734e
After aborting a thread, wait for it to actually finish. EXPERIMENTAL for OSG
...
testing. DO NOT UPDATE TO THIS UNLESS YOU KNOW WHAT YOU'RE DOING!
2010-06-02 17:14:37 +01:00
f5ca701273
Comment and remove JScript support. Mono 2.7Dev and 2.8 no longer include the
...
needed libraries
2010-06-01 14:19:26 +01:00
992efdc283
Fix a nullref in EventManager caused by RegionReady not setting the scene
2010-05-30 23:57:33 +01:00
c7d812802f
Adjust Scene.DeleteAllSceneObjects() to not delete objects attached to avatars.
...
This is going to be the right behaviour in all cases, I should think.
This means that avatars in region when an oar is loaded do not lose their attachments
2010-05-28 18:52:30 +01:00
5074d290e4
commit code which stops full updates being fired multiple times when attachments cross standalone region boundaries
...
lots of messy debug code here too which would need to be removed
2010-05-27 21:35:17 +01:00
d6e1333810
If a transfer request is received for a task inventory item asset, then route the permissions request through the existing CanEditScript() and CanEditNotecard() methods.
...
This implements the 'share with group' flag for notecards and scripts in prim inventory since the PermissionsModule checks group membership and permissions.
Other than that, the code in PermissionsModule duplicates the checks in LLClientView so there should be no change other than allowing group members to edit embedded notecards and scripts.
For all other asset types, the permission checking code in LLClientView continues to be used, pending refactoring of suitable permissions code
This means that 'share with group' will not yet work for prim inventory items other than notecards and scripts
2010-05-26 21:12:56 +01:00
66a833796c
check group membership and powers with the groups module rather than just the client's active group id
...
this resolves the earlier issue where notecards/scripts shared with group could only be edited if the user had that group active
2010-05-26 20:59:25 +01:00
32c43bf59a
Partially implement share with group option for object inventory items
...
If serverside permissions are off then this works as expected. Previously, it was impossible for more than one person to edit such items even if permissions were off.
If serverside permissions are on then this works as expected if the object was created by an avatar who had the required group active.
However, if the group for the object is later set then the contained item is still not editable. This may be linked to a wider bug where the object is still not modifiable by the group anyway
Resolve conflict in LLClientView
2010-05-26 20:58:40 +01:00
a6c7995b71
backport from master fix for attachment display
...
this fix stops two full updates being sent on attachment rather than one. Sending two can stop the client displaying attachments properly
2010-05-24 19:19:44 +01:00
8048a8a39d
backport some IGroupsModule method doc and an extra exposed method from git master
2010-05-24 17:33:58 +01:00
2603f9a134
stop "load oar" reporting ignored objects when those objects were actually loaded just fine
...
this fix also means that loaded scripts are properly/more promptly started
here, the fix is only done in the oar module. In master, the underlying problem was fixed instead but this is more invasive.
2010-05-24 16:53:43 +01:00
3bd2ec7b2e
Apply http://opensimulator.org/mantis/view.php?id=4724
...
Tweak to update appearance calls
Fix for deleting inventory items from a folder.
Thanks StrawberryFride!
2010-05-21 22:40:15 +01:00
e589a31bc4
Only send dialogs and notices to root agents, not child agents
...
Small change for backport
Conflicts:
OpenSim/Region/CoreModules/Avatar/Dialog/DialogModule.cs
2010-05-21 22:23:13 +01:00
e4114799c3
Replace CSJ2K with fixed version (Mantis #3318 )
2010-05-21 15:57:36 +01:00
59546a348a
Change the way the object perms are propagated to attempt to salvage
...
some older content
2010-05-21 15:57:16 +01:00
427ae1087f
minor: correctly print out missing item id when it can't be found rather than a NullReferenceException
2010-05-20 23:44:49 +01:00
1c63f6cd85
Don't append ACKs to zerocoded packets. Although this should fine in theory, I'm seeing the viewer ignore or fail to parse ACKs appended to our zerocoded packets. This should cut down on viewer->sim resend traffic
2010-05-20 23:38:38 +01:00
3b38cc10de
When saving an oar/iar, don't attempt to write out the data byte array if it's empty
...
On Mono 2.6 (and probably before) this causes a non-obvious IOException
2010-05-20 23:35:36 +01:00
d446b19e72
minor: commented out code removal
2010-05-20 23:34:44 +01:00
e9bac02fec
Completely prevent full update packets being sent after kill object packets
...
If a full update is sent after the kill, the object remains as in the linden viewer but in an undeletable and unowned state until relog
This patch prevents this by recording kills in LLClientView
Backport from master
2010-05-20 23:32:23 +01:00
Justin Clark-Casey (justincc)